HK1178253B - Underfill management system for a biosensor - Google Patents

Description

Underfill management system for biosensors

Cross Reference to Related Applications

This application claims priority from U.S. provisional application No.61/352,234 entitled "Underfill management system for a Biosensor," filed on 7/6/2010, the entire contents of which are incorporated herein by reference.

Background

Biosensor systems provide for the analysis of biological fluids such as whole blood, serum, plasma, urine, saliva, interstitial fluid (interstitial fluid) or intracellular fluid (intracellular fluid). Typically, these systems include a measurement device that analyzes a sample residing in a test sensor. Typically, the sample is in liquid form and may, in addition to being a biological fluid, be a derivative of the biological fluid, such as an extract, a dilution, a filtrate or a reconstituted precipitate (recovered precipitate) and the like. The analysis performed by the biosensor system determines the presence and/or concentration of one or more analytes, such as alcohol, glucose, uric acid, lactate, cholesterol, bilirubin, free fatty acids, triglycerides, proteins, ketones, phenylalanine, or enzymes, in the biological fluid. The analysis is useful in the diagnosis and treatment of physiological abnormalities. For example, a diabetic patient may use a biosensor system to determine the glucose level in whole blood in order to adjust diet and/or medication.

Biosensor systems may be designed to analyze one or more analytes, and may use different volumes of biological fluid. Some systems can analyze a single drop of whole blood, including red blood cells, for example, in a volume of 0.25 to 15 microliters (μ L). The biosensor system may be implemented using a bench-top measuring device, a portable measuring device, and the like. The portable measuring device may be hand-held and capable of qualitative and/or quantitative determination of one or more analytes in a sample. Examples of portable measuring devices include Bayer HealthCare, Inc. (Bayer HealthCare, Inc.) of Tarrytown, New York, TalytownAnda meter, and examples of bench-top measuring devices include an Electrochemical Workstation (Electrochemical Workstation) available from CH Instruments of Austin, Texas.

In electrochemical biosensor systems, the analyte concentration is determined from an electrical signal generated by an electrochemical oxidation/reduction or redox reaction of a measurable species upon application of an excitation signal to a sample. The measurable species may be an ionized analyte or an ionized species responsive to the analyte, such as a mediator. The excitation signal may be a potential or a current, and may be constant, variable, or a combination thereof, for example when an AC signal is applied with a DC signal offset. The excitation signal may be applied as a single pulse or in the form of multiple pulses, sequences or periods.

Electrochemical biosensor systems typically include a measurement device having electrical contacts that are connected to electrical conductors of a test sensor. The electrical conductor may be made of a conductive material such as solid metal, metal paste, conductive carbon paste, conductive polymer, or the like. Typically, the electrical conductors are connected to a working electrode, a counter electrode (counter electrode), a reference electrode and/or other electrodes extending into the sample reservoir. One or more electrical conductors may also extend into the sample reservoir to provide functions not provided by those electrodes described above.

The test sensor may include a reagent that reacts with the analyte in the sample. The reagents may include: an ionizing agent for promoting a redox reaction of the analyte; and a mediator or other substance that assists in the transfer of electrons between the ionized analyte and the electrode. The ionizing agent may be an analyte-specific enzyme such as glucose oxidase or glucose dehydrogenase that catalyzes the oxidation of glucose. The reagent may include a binder that binds the enzyme and the mediator together. The binder is a polymeric material that is at least partially water soluble and that provides physical support and containment of the agent while having chemical compatibility with the agent.

The mediator facilitates the transfer of electrons from the first species to the second species. For example, the mediator may assist in the transfer of electrons from the redox reaction between the analyte and the redox enzyme to or from the surface of the working electrode of the test sensor. The mediator may also assist in the transfer of electrons to or from the surface of the counter electrode to the sample. The mediator may be capable of transferring one or more electrons during the conditions of the electrochemical reaction. The mediator may be: organic transition metal complexes, such as ferrocyanide/ferricyanide; coordination compound metal complexes, such as ruthenium hexaammonium; electroactive organic molecules such as 3-phenylimino-3H-phenothiazine (PIPT) and 3-phenylimino-3H-phenoxazine (PIPO); and so on.

The test sensor may be placed in a measurement device and a sample may be introduced into a sample reservoir of the test sensor for analysis. A chemical redox reaction is initiated between the analyte, the ionizing agent and any mediator to form an electrochemically measurable species. For the analysis of the sample, the measuring device applies an excitation signal to electrical contacts connected to electrical conductors of the test sensor. These conductors carry electrical signals to the electrodes, which carry the excitation into the sample. The excitation signal causes an electrochemical redox reaction of the measurable species, which generates an analytical output signal. The electroanalytical output signal from the test sensor can be a current (as generated by amperometry or voltammetry), a potential (as generated by potentiometry/amperometry), or an accumulated charge (as generated by coulometry). The measurement device determines an analyte concentration in response to an analytical output signal from an electrochemical redox reaction of a measurable species.

In amperometry, a potential or voltage is applied to the sample. The electrochemical redox reaction of the measurable species generates an electrical current in response to the potential. The current is measured at a substantially constant potential at a fixed time in order to quantify the analyte in the sample. Amperometry measures the rate at which measurable species are electrochemically oxidized or reduced to determine the analyte concentration in the sample. Thus, amperometry does not measure the total amount of analyte in the sample, but rather determines the analyte concentration in the sample based on the rate of electrochemical redox reaction of the analyte in response to time. Biosensor systems using amperometry are described in U.S. Pat. Nos. 5,620,579, 5,653,863, 6,153,069 and 6,413,411

In coulometry, a potential is applied to a sample in order to exhaustively oxidize or reduce a measurable species within the sample. The applied potential generates a current that is integrated over time for the electrochemical redox reaction to produce a charge indicative of the analyte concentration. Coulometry generally attempts to capture the total amount of analyte within a sample, which necessitates knowledge of the sample volume to determine the concentration of analyte in the sample. A biosensor system using coulometry for whole blood glucose measurement is described in us patent No.6,120,676.

In voltammetry, a varying potential is applied to the sample. The electrochemical redox reaction of the measurable species generates an electrical current in response to the applied potential. The current is measured as a function of the applied potential to quantify the analyte in the sample. Voltammetry generally measures the rate at which a measurable species undergoes oxidation or reduction to determine the analyte concentration in a sample. Thus, voltammetry does not measure the total amount of analyte in a sample, but rather determines the analyte concentration in a sample based on the rate of electrochemical redox reaction of the analyte in response to a potential.

In gated amperometry and gated voltammetry, pulsed excitation may be used as described in U.S. patent publication No. 2008/0173552 filed 12/19 and U.S. patent publication No. 2008/0179197 filed 2/26 2006, respectively.

The measurement performance of a biosensor system is defined in terms of accuracy, which reflects the combined effect of the random error component and the systematic error component. The systematic error or trueness (trueneness) is the difference between the mean value determined from the biosensor system for the analyte concentration of the sample and one or more accepted reference values. The degree of truth may be expressed in terms of mean bias, where a larger mean bias value represents a lower degree of truth and thus contributes to lower accuracy. Accuracy is the closeness of agreement between multiple analyte readings relative to the mean. One or more errors in the analysis contribute to the bias and/or inaccuracy of the analyte concentration determined by the biosensor system. Thus, a reduction in the analysis error of the biosensor system leads to an increase in accuracy and thus an improvement in measurement performance.

Bias can be expressed in terms of "absolute bias" or "percent bias". The absolute bias may be expressed in measured units such as mg/dL, while the percent bias may be expressed as a percentage of the absolute bias value relative to 100mg/dL or a reference analyte concentration of the sample. For glucose concentrations less than 100mg/dL, the percent bias was defined as (100 mg/dL above the absolute bias ratio) 100. For glucose concentrations of 100mg/dL and higher, the percent bias was defined as the absolute bias over the reference analyte concentration 100. A recognized reference value for analyte glucose in a whole blood sample may be obtained using a reference instrument, such as YSI 2300STATPLUS, available from YSI Inc. of Yellow Springs, Ohio^TM. For other analytes, other reference instruments and means can be used to determine percent bias.

The percentage of the analysis that falls within the "percentage bias limit" of the selected percentage bias boundary indicates the percentage of the determined analyte concentration that is close to the reference concentration. Thus, the limit defines how close the determined analyte concentration is to the reference concentration. For example, 95 out of 100 performed analyses (95%) fall within the ± 10% percent bias limit is a more accurate result than 80 out of 100 performed analyses (80%) fall within the ± 10% percent bias limit. Similarly, 95 out of 100 performed analyses fall within the ± 5% percent bias limit are more accurate results than 95 out of 100 performed analyses fall within the ± 10% percent bias limit. Thus, an increase in the percentage of analysis that falls within the selected percentage bias limit or within a narrower percentage bias limit indicates an increase in the measurement performance of the biosensor system.

The mean may be determined for the percentage bias determined from the multiple analyses using the test sensor to provide a "mean percentage bias" for the multiple analyses. Since the mean percent bias can be determined, "percent bias standard deviation" can also be determined to describe how far the percent biases of the multiple analyses are from each other. The percentage biased standard deviation may be considered an indicator of the accuracy of the multiple analyses. Thus, a reduction in the percentage-biased standard deviation represents an enhancement in the measurement performance of the biosensor system.

Enhancing the measurement performance of a biosensor system by reducing errors from these or other sources means: for example, more of the analyte concentration determined by the biosensor system may be used by the patient for accurate therapy while monitoring blood glucose. In addition, the need for the patient to discard the test sensor and perform repeated analyses may also be reduced.

A test case is a collection (data population) of multiple analyses generated under substantially the same test conditions. For example, typically, analyte concentration values determined in the case of self-testing (self-testing) exhibit poorer measurement performance than in the case of a health care professional ("HCP") test; the analyte concentration values determined in the HCP test case demonstrated poorer measurement performance than in the controlled environment test case. This difference in measurement performance may be reflected in a greater percentage bias standard deviation of analyte concentration determined by the user self-test as compared to analyte concentration determined by the HCP test or by the controlled environment test. The controlled environment is an environment, preferably a laboratory environment, in which the physical properties of the sample and environmental factors can be controlled. Thus, in a controlled environment, the hematocrit (hemachrome) concentration may be fixed and the actual sample temperature can be known and compensated for. In the case of HCP testing, errors in operating conditions can be reduced or eliminated. In the case of user-self-test testing, such as clinical trials, the determined analyte concentration will likely include errors from all types of sources of error.

The biosensor system uses the analytical output signal to determine an analyte concentration of the sample. The biosensor system may provide an analysis output signal including one or more errors during analysis of the sample. These errors may be reflected in an abnormal output signal, for example, when one or more portions of the output signal or the entire output signal is not responsive or incorrectly responsive to the analyte concentration of the sample. These errors may arise from one or more causes of error, such as physical characteristics of the sample, environmental factors of the sample, operating conditions of the system, and so forth. Physical characteristics of the sample include the hematocrit (red blood cell) concentration of whole blood, interfering substances, and the like. Interfering substances include ascorbic acid, uric acid, acetaminophen, and the like. Environmental factors of the sample include temperature, etc. Operating conditions of the system include an underfill condition when the sample volume is not large enough, slow filling of the sample, intermittent electrical contact between the sample and one or more electrodes in the test sensor, degradation of reagents that interact with the analyte, and the like. There may be other causes or combinations of causes that may lead to errors.

If the test sensor is not filled with sample, the test sensor may provide an inaccurate analysis of the analyte in the sample. The biosensor system may include an underfill detection system for preventing or screening out analysis associated with an undersized sample volume. Some underfill detection systems have one or more indicator electrodes that may be separate or part of a working, counter, or other electrode used to determine the analyte concentration in the sample. Other underfill detection systems have a third or indicator electrode in addition to the counter and working electrodes. An additional underfill detection system has a sub-element in electrical communication with the counter electrode. Unlike the working and counter electrodes, conductive subelements, trigger electrodes, and the like are not used to determine the analyte response signal generated by the biosensor system. Thus, they may be bare conductive traces (conductive traces), conductors with non-analyte specific reagents (e.g., mediators), etc.

Typically, when a sample is present in the sample reservoir, an electrical signal is passed between the indicator electrodes, between the third electrode and the counter electrode, or between the subelement and the working electrode. The electrical signal indicates the presence or absence of a sample and may indicate whether the sample partially or completely fills the sample reservoir. A biosensor using an underfill detection system with a third electrode is described in us patent No.5,582,697. Biosensors using underfill detection systems with sub-elements having counter electrodes are described in U.S. Pat. No.6,531,040.

Other underfill methods may use the electrical properties of the sample as a function of sample volume to determine underfill. For example, U.S. patent 6,797,150 discloses using capacitance to determine whether a test sensor is too severely underfilled to be analyzed, or whether a test sensor is underfilled but is analyzable with an adjustment to the determined concentration. Unlike indicator electrode systems that rely solely on the sample being a conductive sample, electrical property-based systems rely on sample electrical properties that vary with sample volume. In the 6,797,150 patent, if the test sensor is severely underfilled, the analysis is stopped. If the test sensor is not full but analyzable with an adjustment, the method applies the same analytical method for full test sensor, but then the analyte concentration thus determined is adjusted with an offset value. Thus, the underfill analysis method can detect and analyze partially underfilled test sensors, but lacks the ability to correct for errors due to the need for additional sample for the test sensor to perform a proper analysis.

Although conventional biosensor systems using underfill detection systems may analyze a test sensor with a certain degree of underfill, or may reduce erroneous results due to an insufficient amount of Sample by stopping the analysis or by instructing the user to add more Sample, these underfill detection/analysis systems typically do not account for analysis errors due to more than one addition of Sample to the test sensor, variations in the Sample fill rate, or variations in the Sample addition profile. Sample addition morphology errors can occur when the sample does not flow uniformly over the reagent.

There is an ongoing and urgent need for improved biosensor systems, particularly those that can provide accurate and/or precise analyte concentrations determined from underfilled test sensors that are subsequently filled for analysis. Such an improved biosensor system can compensate for errors due to refilled test sensors, variations in sample fill rates, and/or sample addition patterns. The systems, devices, and methods of the present invention overcome at least one of the deficiencies associated with conventional biosensor systems.

Disclosure of Invention

A method for determining an analyte concentration in a sample, comprising: determining a fill state of the test sensor; signaling addition of a supplemental sample to substantially fill the test sensor; applying an analytical test excitation signal to the sample; generating at least one analytical output signal value responsive to the analyte concentration in the sample and the analytical test excitation signal; compensating for underfill errors in the at least one analytical output signal value in response to a fill state of the test sensor; and determining an analyte concentration in the sample from the at least one analytical output signal value and the compensation.

A biosensor system for determining an analyte concentration in a sample, comprising: a test sensor having a sample interface in electrical communication with a reservoir formed by the test sensor; and a testing device having a processor connected to a sensor interface, the sensor interface in electrical communication with the sample interface, the processor in electrical communication with a storage medium. The processor determines a fill status of the test sensor, signals addition of a supplemental sample to substantially fill the test sensor, instructs a charger to apply an analytical test excitation signal to the sample, measures at least one analytical output signal value responsive to an analyte concentration in the sample and the analytical test excitation signal, compensates for an underfill error in the at least one analytical output signal value responsive to the fill status of the test sensor, and determines an analyte concentration in the sample from the at least one analytical output signal value and the compensation.

A method for determining an analyte concentration in a sample, comprising: applying a regular polling sequence and an extended polling sequence to the sample, the extended polling sequence comprising at least one different extended input pulse; and generating at least one analytical output signal responsive to the concentration of the analyte in the sample. The method further comprises the following steps: selecting an error parameter in response to the at least one different extended input pulse; determining at least one slope deviation value from the error parameter; and determining an analyte concentration in the sample from the at least one analytical output signal and a slope compensation equation responsive to at least one exponential function, wherein the slope compensation equation comprises at least one reference correlation and at least one slope deviation.

A method for determining an analyte concentration in a sample, comprising sequentially detecting a sample fill of test sensors, wherein the sequentially detecting comprises: determining when two different electrode pairs of the test sensor are contacted by the sample; generating at least one analytical output signal responsive to the concentration of the analyte in the sample; selecting an error parameter in response to when two different electrode pairs of the test sensor are contacted by the sample; determining at least one exponential function responsive to the error parameter; and determining an analyte concentration in the sample from the at least one analytical output signal and a slope compensation equation responsive to the at least one exponential function, wherein the slope compensation equation includes at least one reference correlation and at least one slope deviation.

Drawings

The invention can be better understood with reference to the following drawings and description. The components in the figures are not necessarily to scale, emphasis instead being placed upon illustrating the principles of the invention. Moreover, in the figures, like reference numerals designate corresponding parts throughout the different views.

FIG. 1A shows a schematic view of a test sensor.

FIG. 1B shows a schematic diagram of a test sensor with indicator electrodes.

Fig. 2A shows a gated amperometric pulse sequence in the case where the test excitation signal applied to the working electrode and the counter electrode comprises a plurality of pulses.

Fig. 2B shows a gated amperometric pulse sequence in the case where the test excitation signal applied to the working and counter electrodes comprises a plurality of pulses and in the case where a second excitation signal is applied to the additional electrode to generate the auxiliary output signal.

Fig. 3A illustrates a regular polling sequence and an extended polling sequence of polling input signals and a test excitation signal for a biosensor system with a binary underfill management system.

Fig. 3B illustrates the regular and extended polling sequences of the polling input signal and the test excitation signal of a biosensor system having an underfill management system capable of discerning the degree of underfill.

Fig. 3C and 3D illustrate regular and extended polling sequences of other polling input signals and other test excitation signals for a biosensor system with a binary underfill management system.

FIG. 4A shows S_cal、S_hyp、ΔS、A_corr、A_calAnd Δ a.

Fig. 4B illustrates an underfill compensation method including a transfer function, main compensation, and residual compensation.

FIG. 5A illustrates an analytical method for determining the concentration of an analyte in a sample using a binary underfill management system.

FIG. 6A illustrates an analytical method for determining an analyte concentration in a sample using an underfill management system that determines an initial degree of underfill.

FIG. 7A shows the Δ S value (Δ S) before compensation in the case where the subsequent SFF compensation equation includes an exponential function that relates the ratio error parameter (R7/6) to slope_{Is not compensated}) And the compensated Delta S value (Delta S)_{Has been compensated}) The correlation between them.

Fig. 7B and 7D show percentage Bias (% -Bias) values for a plurality of uncompensated and compensated analyses for subsequent SFF test sensors and an initial SFF test sensor.

FIG. 7C illustrates the percentages of uncompensated and compensated determined glucose analyte concentrations that fall within the + -15% percent bias limit when the test sensor is initially underfilled and the subsequent SFF is for analysis.

FIG. 7E shows the measured performance of a binary compensation system using a complex exponential function.

Fig. 8A, 8B, 8C and 8D show the performance of the LUF compensation system using a main function and different first residual functions.

Fig. 9A, 9B, 9C and 9D show the performance of HUF compensation systems using different master functions.

Fig. 10A shows a schematic diagram of a biosensor system with an underfill management system.

Detailed Description

The underfill management system includes: an underfill recognition system that evaluates whether to analyze the sample in response to an initial test sensor fill state, or to wait for a supplemental sample to be added to the test sensor; and an underfill compensation system that compensates the analyte analysis for one or more errors caused by the initial and subsequent fills of the test sensor. The underfill recognition system may detect the presence of a sample, determine whether the test sensor is initially substantially full or underfilled, indicate when the sample volume is underfilled so that a supplemental sample may be added to the test sensor, and start or stop sample analysis in response to the sample volume. The underfill recognition system may also determine an initial degree of underfill. After the underfill recognition system determines the initial fill state of the test sensor, the underfill compensation system compensates the analysis based on the initial fill state of the test sensor to improve the measurement performance of the biosensor system for the initial underfill test sensor. The underfill recognition system may also determine one or more subsequent fill states, and the underfill compensation system may compensate the analysis based on the one or more subsequent fill states.

The underfill recognition system may be binary in operation or capable of detecting the degree of underfill. If binary, the underfill recognition system determines that there is a sample and there is enough sample to continue the analysis from the initial fill, or that there is a sample but not enough sample to continue the analysis from the initial fill. If insufficient sample is present to continue from the initial fill, the binary system signals the user to add additional sample, preferably for a predetermined period of time, and then directs the system to continue analysis after the sensor is substantially full. Thus, the underfill management system implements one of two underfill compensation systems in response to one of: (1) whether the initial fill resulted in the test sensor being substantially full (SFF); or (2) whether a subsequent fill is provided to reach the SFF of the test sensor. One or more subsequent fills may be used to SFF test the sensor.

An underfill recognition system capable of detecting underfill levels may provide the underfill management system with the following capabilities in addition to providing binary underfill recognition: one of at least three underfill compensation systems is implemented based on whether the initial fill provides (1) substantially full (SFF), or (2) low volume underfill (LUF), or (3) high volume underfill (HUF). Thus, different compensation systems may be implemented in response to different initial fill states. Further, the underfill detection system may be capable of determining and implementing a different compensation system in response to whether a first subsequent fill results in SFF or whether a second or third subsequent fill results in SFF. For example, the compensation system may be implemented to compensate for the following: when the initial fill provides a LUF state; when the first subsequent fill provides a HUF state; and when the second subsequent fill provides the SFF state.

After the underfill recognition system determines that the test sensor is SFF, the biosensor system applies an analytical test stimulus to the sample. The underfill compensation system applies one or more compensation equations in response to an initial and/or subsequent fill state of the test sensor. Preferably, the compensation equation comprises an exponential function extracted from the intermediate signal of the analytical output signal and from the auxiliary output signal to adjust the correlation for determining the analyte concentration in the sample from the analytical output signal. Preferably, the exponential function is a complex exponential function and may be paired with one or more residual functions to provide an analyte concentration that is underfilled.

In a biosensor system having an underfill management system, the underfill recognition system is preferably selected to reduce or substantially eliminate any irreversible alteration of the analyte concentration in the sample prior to application of an analytical test stimulus that electrochemically oxidizes or reduces a measurable species to determine the analyte concentration of the sample. An "irreversible change" is a change in mass, volume, chemical property, electrical property, or combination thereof, etc., from an original condition to another condition that cannot be incomplete or that cannot be substantially returned to the original condition. In an assay that relates the rate of an electrochemical redox reaction to the concentration of an analyte, the original reaction rate cannot be obtained once a portion of the analyte is irreversibly altered by a stimulus having a relatively large amplitude and/or a long pulse width. In these assays, the pulse width is more likely to alter the analyte concentration.

Underfill recognition systems that determine the fill state of a test sensor without irreversibly altering the analyte concentration prior to application of an excitation signal generally fall into two types: (1) sequential detection of sample filling; and (2) polling the input signal. However, other underfill recognition systems may be used, preferably that do not irreversibly alter the analyte concentration of the sample prior to application of the excitation signal, and are capable of providing notification of the addition of a supplemental sample to the test sensor.

Underfill detection systems that use sequential detection of sample filling do not irreversibly oxidize, reduce, or otherwise alter the analyte in the sample due to the use of relatively short pulse widths to detect electrical connections between electrodes that are placed in series as the sample enters the test sensor. Underfill detection systems that use a polling input signal use shorter pulse widths that do not irreversibly oxidize, reduce, or otherwise alter the analyte in the sample. The pulse of the polling input signal is distinct from a larger amplitude or longer pulse width of the test stimulus that irreversibly oxidizes, reduces, or otherwise alters the analyte signal of the analyte in the sample.

In general, the underfill recognition system is selected based on the electrode design of the test sensor and the desired level of compensation for the underfill management system. The more sophisticated the underfill management system, the better the measurement performance of the system with varying initial underfill levels. The test sensors can have various configurations, including those having multiple electrodes and conductors. The test sensor may have 2, 3,4 or more electrodes. Test sensors that use a polling input signal for underfill detection typically require two electrodes, while test sensors that use sequential detection of sample fill typically require at least three consecutive electrodes.

A binary underfill recognition system for detecting underfill may be implemented on the test sensor 100 shown in fig. 1A. The test sensor 100 forms a reservoir 104, and the reservoir 104 includes a counter electrode 106 and a working electrode 108 located in the reservoir 104. "located in. Counter electrode 106 includes a sub-element 110, sub-element 110 being located upstream of working electrode 108 in reservoir 104. The mediator may be disposed on the counter electrode 106, on the working electrode 108, in the reservoir 104, a combination thereof, or the like. Other components are omitted from the test sensor 102 for clarity. Counter electrode 106 and subelement 110 can have different redox potentials, such as when a mediator is disposed on counter electrode 106 but not on subelement 110 or when a different mediator system is disposed on subelement 110.

When the test sensor includes enough sample to accurately analyze the concentration of one or more analytes in the sample using the initial SFF compensation system, then the sensor 100 is SFF. The sample volume required to test the sensor SFF for accurate initial SFF compensation can be determined experimentally, theoretically, or a combination thereof. When the working electrode is covered with the sample, the test sensor 100 can be considered to be SFF. A substantial fill of the test sensor is obtained when at least 85%, preferably at least 90%, more preferably at least 95% of the volume of the sample reservoir filled with the test sensor is filled. For example, a test sensor having a reservoir volume of 0.5 μ L can be considered SFF when at least 0.42 μ L of sample is present in the reservoir, preferably when at least 0.45 μ L of sample is present in the reservoir, more preferably when at least 0.48 μ L of sample is present in the reservoir. Thus, the underfill recognition system may be configured to determine the SFF at one or more of these reservoir fill volumes as a function of the design and placement of the working electrode in the reservoir 104.

When applied to the test sensor 100, the polling input signal generates one or more polling output signals from the sample that can be used to detect when the sample is present, when the test sensor is underfilled, and when the test sensor is SFF. When the test sensor is SFF, an analytical test excitation signal is applied to the sample and the analytical test excitation signal generates one or more output signals that can be used to determine one or more analyte concentrations in the sample. Upon underfill, the underfill detection system requests the user to add more biological fluid to the test sensor. The biosensor may use multiple sample thresholds for detecting a replenishment sample in the sensor, for example, an initial sample threshold for detecting the presence of a sample in the test sensor and a second or refill sample threshold for detecting when more sample has been added to the test sensor.

The polling signal has a regular polling sequence of one or more regular input pulses followed by an extended polling sequence of one or more extended input pulses. The regular input pulses are essentially the same, but different regular input pulses may be used. The polling signal is essentially a sequence of polling pulses separated by polling relaxation (polling relaxation). During the polling pulse, the electrical signal is turned on. The turn-on includes a period of time when the electrical signal is present. During the polling relaxation, the electrical signal is significantly reduced in amplitude relative to when the electrical signal is on. The reducing includes when the electrical signal is reduced by at least one order of magnitude relative to when the electrical signal is on. Reducing also includes when the electrical signal is reduced to off. Shutdown includes periods when no electrical signal is present. Turning off does not include periods where an electrical signal is present but the electrical signal does not have an amplitude per se. The electrical signal may be switched on and off by closing and opening an electrical circuit, respectively. The above-described circuit may be opened and closed in a mechanical, electrical or similar manner. Other on/off mechanisms may be used.

The extended polling sequence is part of a polling signal. The extended polling sequence has one or more extended input pulses. One or more or none of these extended input pulses may be essentially the same as the regular input pulses. At least one extended input pulse in the extended polling sequence is different from a regular input pulse of the regular polling sequence. The different extended input pulse may be the last or another extended input pulse in the extended polling sequence. The different extended input pulses may be stepped down, stepped up, or a combination thereof relative to the regular input pulses. The step down comprises extending the input pulse with an extension amplitude decreasing with each subsequent input pulse. The step-up comprises extending the input pulse with an increasing amplitude with each subsequent input pulse. The extended polling sequence may generate one or more volume output signals responsive to the sample volume. The volume output signal can be used to determine whether the sample is initially SFF or underfilled.

When a polling signal is applied to a sample in a biosensor, each pulse of the polling signal typically generates a corresponding output pulse from the sample. The one or more output pulses form a polling output signal. Each regular input pulse of the regular polling sequence generates a regular output pulse in the sample output signal. The biosensor detects the presence of a sample when at least one of the regular output pulses reaches a sample threshold and then applies an extended polling sequence. Each extended input pulse of the extended polling sequence generates an extended output pulse in the volume output signal. The different spread input pulses generate different spread output pulses that are responsive to a fill state of the test sensor.

The regular and extended polling sequences may have a pulse width of less than about 500 milliseconds (ms) and a pulse interval of less than about 2 seconds (sec). The polling sequence may have an input pulse width of less than about 100ms and a pulse interval of less than about 500 ms. The polling sequence may have an input pulse width in the range of about 0.5 milliseconds to about 75ms and an input pulse interval in the range of about 5ms to about 300 ms. The polling sequence may have an input pulse width in the range of about 1 millisecond to about 50ms and an input pulse interval in the range of about 10ms to about 250 ms. The polling sequence may have an input pulse width of about 5ms and an input pulse interval of about 125 ms. Thus, each of the regular and extended polling sequences may have a pulse width and a pulse interval selected from these or other values, so long as the extended polling sequence includes an extended input pulse that is different from the regular input pulse width and the pulse interval.

One or more volume thresholds may be used to detect when the test sensor is initially SFF or underfilled. The test sensor is SFF when the different expanded output pulses reach the selected volume threshold. When the different expanded output pulses do not reach the volume threshold, the test sensor is underfilled and requires more sample for analysis. The sample covers fewer electrodes in the test sensor when the test sensor is not full than when the test sensor is SFF. The underfill state and the SFF state may be selected in response to experimental data, theoretical analysis, desired accuracy and/or precision of volume or analysis, mediators used, electrode configurations, combinations thereof, and the like.

To determine binary underfill by sequential detection using the test sensor 100, an electrical potential having a relatively short pulse width (e.g., 50 milliseconds or less) can be applied across the working electrode 108 and the counter electrode 106 using the electrically connected subelement 110. By monitoring the current output when a sample is introduced into the sample reservoir 104, it can be determined when the sample will contact the working/subelement and thus the working/counter electrode. If only the working electrode/subelement is contacted by the sample, the biosensor system requests that a supplemental sample be added to the test sensor 100 in SFF. Although less preferred due to some irreversible alteration of the analyte concentration, binary underfill may also be determined during the initial phase of the application of the analytical input signal. A more detailed description of the use of analytical input signals to determine Underfill may be found in U.S. patent publication No.2009/0095071 entitled "Underfill detection System for a Biosensor".

The test sensor 100 may be operated in a binary manner using a polling signal or sequential detection of an underfill recognition system, in which: the analysis continues from the initial SFF or the biosensor system signals a supplemental sample to test the sensor SFF after the initial fill but before the analysis continues. When the test sensor is SFF, the biosensor system may apply the test excitation signal immediately after the extended polling period or at other selected times. The underfill management system implements a compensation system for an initial SFF test sensor or for an initial underfill and subsequent SFF test sensor. Since the underfill management system selects the appropriate underfill compensation based on the initial fill state of the test sensor, the underfill compensation system may also compensate for situations when an analysis input signal is used to detect underfill, yet to a lesser degree than when the initial fill state of the test sensor is determined prior to application of the analysis input signal.

An underfill recognition system that determines one or more degrees of underfill using polling may also be implemented on the test sensor 100 of FIG. 1A. In an underfill recognition system that determines one or more degrees of underfill, a plurality of different extended input pulses are used to determine the degree of underfill.

With respect to binary underfill recognition systems using polling, additional volume thresholds may be used to detect when the test sensor is the initial SFF or a range with an initial underfill volume. The test sensor is SFF when the different expanded output pulses reach the selected volume threshold. When more than one different extended output pulse reaches a volume threshold, or reaches one volume threshold but not another, the test sensor is underfilled, more sample is required for analysis, and the degree of underfill can be determined.

Thus, depending on whether a binary or degree underfill recognition system is used, the volume threshold may be selected to distinguish between multiple fill states, including an initial SFF, an initial underfill, a different initial volume or range of volumes of underfill, a minimum and/or maximum volume, combinations thereof, and so forth. For example, if the degree underfill recognition system detects an initial underfill, a volume threshold may be selected to distinguish a low volume underfill (LUF) from a high volume underfill (HUF) initial fill state.

The volume threshold may be a predetermined threshold stored in a memory device, a predetermined threshold obtained from a look-up table, or the like. The predetermined threshold may have been developed theoretically or from a statistical analysis of laboratory work. The volume threshold may be a threshold measured or calculated in response to one or more of the polling output signals. The volume threshold may be selected to identify when a change in one or more output signals is responsive to a volume condition.

The underfill management system may use multiple volume thresholds to determine the volume of the sample or the degree of underfill of the biosensor. When the volume output signal exceeds one volume threshold but not another, the volume output signal will indicate that the sample volume is between the volumes associated with the volume thresholds. For example, if the volume threshold of the initial LUF is exceeded, but the volume threshold of the initial SFF is not exceeded, the volume output signal will indicate the initial HUF. More volume thresholds may be used to provide more accurate volume determinations.

The periods in the extended polling sequence may be used to create a buffer or delay of slow-filling samples. While an initial expanded output pulse in the volume output signal may indicate underfill, a later or final expanded output pulse may indicate SFF when the sample has substantially completed filling. The period in the extended polling sequence may be used for other criteria, such as with or without multiple thresholds to determine the volume or volume range of the sample.

When the last low extended polling output does not meet the volume threshold, a regular polling sequence and an extended polling sequence will be generated. This cycle may continue indefinitely until the sample volume meets the volume threshold or for a selected number of polling sequences. During this time, a supplemental sample may be added to the test sensor to trigger meeting the volume threshold and achieving SFF of the test sensor.

An underfill recognition system that determines the degree of underfill using sequential detection of sample fill on successive electrodes may be implemented on the test sensor 120 of FIG. 1B. In addition to having the electrodes of the test sensor 100, the test sensor 120 adds additional, electrically independent electrodes 122 and 124. The upstream electrode 124 may be an electrode for providing an auxiliary output signal in response to the hematocrit content of the sample. The downstream electrode 122 may be used to detect that the sample has reached the end of the sample reservoir 104, and thus that the SFF of the test sensor 120 has occurred.

To determine the degree of underfill of the test sensor 120, potential pulses of relatively short duration may be applied to different electrode pairs in sequence to determine which electrode pairs are contacted by the sample. For example, electrodes 124 and 110 may be considered a first electrode pair, electrodes 110 and 108 may be considered a second electrode pair, and electrodes 108 and 122 may be considered a third electrode pair. Contact between the hematocrit electrode 124 and the subelement 110 can be used to indicate the presence of a sample. If the initial fill results in contact between the hematocrit electrode 124 and the subelement 110, but not between the subelement 110 and the working electrode 108, then an initial LUF has occurred. If the initial fill results in contact between the working electrode 108 and the counter electrode 106, but not between the counter electrode 106 and the additional electrode 122, an initial HUF has occurred. If the initial fill results in contact between working electrode 108 and additional electrode 122, an initial SFF has occurred and the analysis can continue to analyze the analyte with the test stimulus.

In addition to the use of individual contacts, the time it takes for the sample to traverse each successive pair of electrodes can be used to determine the initial fill state of the test sensor 120. For example, the underfill management system may determine the time it takes for the sample to contact the subelement 110 and the working electrode 108 after first contacting the hematocrit electrode 124 and the subelement 110. If the time is above the threshold, the test sensor 120 may be considered the initial LUF. Similarly, the underfill management system may determine the time it takes for the sample to contact the working electrode 108 and the additional electrode 122 after first contacting the working electrode 108 and the subelement 110. If the time is above the threshold, the test sensor 120 may be considered the initial HUF.

The volume threshold or sequential detection factor corresponding to the LUF may be selected so as to fill, for example, approximately 40% to 50% of the test-sensor reservoir. Similarly, the value corresponding to the HUF may be selected so as to fill approximately 58% to 70% of the test-sensor reservoir. Other fill percentages of the test sensor reservoir may be selected to represent a LUF, HUF, or other fill state. Preferably, a threshold or sequential detection factor corresponding to the LUF state indicates an initial underfill in the event that the reagent of the working electrode is not substantially contacted by the sample. Similarly, preferably, where at least the reagent of the working electrode is substantially contacted by the sample, a threshold or sequential detection factor corresponding to the HUF state indicates an initial underfill.

If the underfill recognition system determines that a sample is present, LUF or HUF, the system requests replenishment of the sample until SFF occurs. An analytical test stimulus is then applied to determine the analyte concentration of the sample. The value from the analytical output signal can be correlated with the analyte concentration by a correlation equation. To determine the analyte concentration after underfill compensation, the underfill management system implements the underfill compensation system in response to the initial fill state or in response to a combination of the initial fill state and any subsequent fill states.

Fig. 2A shows a gated amperometric pulse sequence in the case where the test excitation signal applied to the working electrode and the counter electrode comprises a plurality of pulses. The values of the analytical output signal current resulting from these pulses are shown above each pulse. The intermediate signal current values are shown as filled circles. Each of the i values is a current value of the analysis output signal in response to the excitation signal. The first digit in the subscript of the i values represents the pulse number, while the second digit in the subscript represents the order in which the signals are output when measuring the current values. For example, i_2，3A third current value measured for the second pulse is shown.

The exponential function described below with respect to the compensation system includes one or more exponents. The exponent characterizes the error parameter and may include a ratio of intermediate signal current values as shown in fig. 2A. For example, the intermediate current value may be compared to the respective pulse signal decay periods to provide a ratio within the pulse, such as a ratio R3 ═ i_3，3/i_3，1、R4＝i_4，3/i_4，1And the like. In the example within these pulses, the last current value recorded from the pulse is divided by the current value recorded from the same pulseThe pulse recorded first current value to form a ratio. In another example, intermediate current values may be compared between separate pulse signal decay periods, such as a ratio R3/2 ═ i_3，3/i_2，3、R4/3＝i_4，3/i_3，3And the like. These are inter-pulse ratios that divide the current value later in the time pulse by the current value earlier in the time pulse.

The exponential function may also include a combination of ratios extracted from the analysis output signal shown in fig. 2A. In one example, the exponential function may be a linear function including a Ratio of ratios, e.g., Ratio 3/2-R3/R2, Ratio 4/3-R4/R3, and so on. In another example, the exponential function may include an algebraic or other combination of exponentials. For example, the combination Index-1 may be expressed as Index-1 ═ R4/3-Ratio 3/2. In another example, the combined Index, Index-2, may be expressed as Index-2 ═ (R4/3)^p-(Ratio3/2)^qWherein p and q are each independently positive numbers.

FIG. 2B shows a gated amperometric pulse sequence where the excitation signals applied to the working and counter electrodes comprise a plurality of pulses and where a second excitation signal is applied to the additional electrode to generate an auxiliary output signal in response to the hematocrit content of the sample. The excitation signal applied to the additional electrode is applied after the analysis of the excitation signal is completed, but may be applied at other times. For example, the current values from the additional electrodes may be used in an exponential function that relates the current values measured from the additional electrodes to the percent Hct of the sample.

Although a gated amperometric test excitation signal is used in the following examples of polling and sequential underfill recognition, other test excitation signals providing a desired compensation system may be used.

In fig. 3A, a polling signal for a binary underfill recognition system is shown, having a regular polling sequence of six regular input pulses and an extended polling sequence of four extended input pulses. The extended polling sequence has three similar extended input pulses followed by a different extended input pulse. The three similar extended input pulses have an extended amplitude of about 400mV, while the different extended input pulse is the last extended input pulse and has an amplitude of about 100 mV. The pulse widths of the regular and extended polling sequences are short, e.g., at most 50ms or at most 20 ms. The regular and extended pulse widths are in the range of about 1ms to about 15ms or in the range of about 5ms to about 10 ms. The reverse arrow indicates that the regular polling sequence and/or the extended polling sequence may be restarted when desired (e.g., when no sample is present, when the test sensor is initially underfilled, or if other criteria are met or not met). The polling signal may be used with a binary underfill detection system to determine whether a sample is present in the test sensor, whether the test sensor is initially SFF, or whether the test sensor is initially underfilled.

The analytical potential sequence shown in FIG. 3A has two test pulses with an excitation pulse width of about 1 second and a relaxation width of about 0.5 second. The first excitation pulse essentially starts at the end of the last extended input pulse in the extended polling sequence. The pulse width of the test excitation is substantially longer relative to the pulse width of the polling pulse, which can result in irreversible alteration of the analyte concentration of the sample.

In fig. 3B, the polling signal of the underfill recognition system capable of discriminating the degree of underfill has a regular polling sequence of six regular input pulses and an extended polling sequence of four extended input pulses. The extended polling sequence has one similar extended input pulse followed by three different extended input pulses. The similar extended input pulse has an extended amplitude of about 400mV, which is essentially the same as the regular amplitude of the regular input pulse. The different extended input pulses step down or have decreasing extended amplitudes of about 300mV, about 200mV, and about 100mV, which are different from the regular amplitudes of the regular input pulses. The polling signal may be used with an underfill recognition system capable of discerning the degree of underfill to determine whether a sample is present in the test sensor, whether the test sensor is initially SFF, whether the test sensor is initially LUF, or whether the test sensor is initially HUF. The polling signal may be used to discern additional degrees of underfill.

The polling output signal includes a sample output signal and a volume output signal. The sample output signal is generated in response to a regular polling sequence. The volume output signal is generated in response to an extended polling sequence. The sample output signal can have a current in the range of about 5nA to about 800nA, in the range of about 50nA to about 500nA, in the range of about 100nA to about 400nA, or in the range of about 200nA to about 300 nA. The volume output signal can have a current in the range of about 5nA to about 800nA, in the range of about 50nA to about 500nA, in the range of about 100nA to about 400nA, or in the range of about 200nA to about 300 nA. Other output current values may be obtained in response to polling input signals based on the nature of the sample and the temperature of the analysis. Preferably, different thresholds may be selected for different temperature ranges.

Fig. 3C and 3D illustrate regular and extended polling sequences of other polling input signals and other test excitation signals for a biosensor system with a binary underfill management system. In fig. 3C, the polling signal is shown to have a regular polling sequence of 7 regular input pulses and an extended polling sequence of 21 extended input pulses, while in fig. 3D, the polling signal is shown to have a regular polling sequence of 15 regular input pulses and an extended polling sequence of 7 extended input pulses. The extended polling sequence has a plurality of periods (seven shown in fig. 3C and three shown in fig. 3D) of extended input pulses, with two higher and one lower extended amplitudes in each period. Each cycle has a start cycle pulse, a middle cycle pulse, and an end cycle pulse. The start and middle periodic pulses are similar extended input pulses having a magnitude of about 450mV, which is essentially the same as the regular magnitude of the regular input pulses. The end period pulse is a different extended input pulse having a magnitude of about 100mV, which is different from the regular magnitude of the regular input pulse. The pulse width and the relaxation width of the regular polling signal and the extended polling signal are essentially the same. Although fig. 3C and 3D illustrate a regular polling sequence followed by an extended polling sequence having seven or three cycles, respectively, the regular polling sequence may be implemented after each cycle or multiple cycles of the extended polling sequence. In fig. 3C and 3D, the regular polling sequence detects the presence of a sample, while the extended polling sequence detects the fill status. Thus, the number of extended input pulses varies depending on how quickly the initial underfill test sensor is subsequently filled to SFF.

The analytical electrical potential sequences shown in fig. 3C and 3D have seven or eight analytical pulses, respectively, with different pulse widths from about 0.25sec to about 0.5sec and different relaxation widths from about 0.25sec to about 1sec, respectively. The first analysis pulse has an analysis pulse potential of about 400 mV. The second analysis pulse has an analysis pulse potential of about 200 mV. The third through sixth analysis pulses in fig. 3C and the third through seventh analysis pulses in fig. 3D each have an analysis pulse potential of about 250 mV. The seventh analysis pulse in fig. 3D and the eighth analysis pulse in fig. 3D have analysis pulse potentials that vary from about 250mV to about 600 mV. The first analysis pulse essentially starts at the end of the last extended input pulse in the extended polling sequence of both figures.

In addition to identifying SFF, underfilling, and requesting replenishment of the sample, the underfill management system also compensates for errors in the analysis by adjusting the correlation used to determine the analyte concentration in the sample. Preferably, the compensation accounts for errors associated with variations in the initial sample fill and any subsequent sample fills of the test sensor. Preferably, different compensation systems are used for the initial SFF or subsequent SFF test sensors. When the underfill recognition system discerns the initial degree of underfill, the subsequent SFF test sensor may be considered an initial HUF or an initial LUF. The compensation system for a particular initial fill state may use one or more different compensation equations and use different values for each equation. The preferred underfill compensation system includes slope-based compensation of the main compensation paired with optional residual compensation. Although these compensation systems are described later, other compensation systems may also be used to provide different underfill compensations in response to whether the test sensor is an initial SFF or a subsequent SFF. Thus, the underfill management system may select between the plurality of compensation systems in response to the determination of the initial fill state and any subsequent fill states by the underfill recognition system.

Slope-based compensation uses a prediction function that compensates for errors in the analyte analysis. Such errors can lead to bias, thereby reducing the accuracy and/or precision of the determined analyte concentration. Fig. 4A illustrates a method of slope-based compensation useful for biosensor systems having a linear or near-linear relationship between an analytical output signal and an analyte concentration. The figure shows S_cal、S_hyp、ΔS、A_corr、A_calAnd Δ a. Line A represents a reference correlation having a slope S_calAnd correlating the output signal in the form of a current value from the biosensor system with an analyte concentration value obtained for the sample from the YSI or other reference instrument. When used during analysis of a sample by a biosensor system, the reference correlation of line a may include an analysis output signal current value having one or more errors that may provide inaccurate and/or imprecise analyte concentration values. Line B represents the error compensation correlation, which has a slope S_hypAnd correlating the current value obtained from the biosensor system with a sample analyte concentration value obtained from a reference instrument. The error compensation correlation has been adjusted or modified to reduce or substantially eliminate one or more of the errors. Δ S is S_calCorrelation line and S_hypThe slope deviation between the correlation lines and may be expressed as a difference or by other mathematical operators. Δ A is the analyte concentration determined uncompensated or uncorrected (A)_cal) And the analyte concentration (A) determined after error compensation or correction_corr) The difference between them.

Thus, the slope-based compensation equation using Δ S can be expressed as follows:

(equation 1)

Wherein A is_corrIs the compensated analyte concentration, i is the value of the output signal from the biosensor system, Int is the intercept of the reference correlation equation, S_calIs the slope of the reference correlation equation, Δ S denotes S_calAnd the assumed slope (S)_hyn) The slope deviation between, the hypothetical slope is the slope of the line providing the analytical output signal value for the analyte concentration of the sample in an error-free manner. Int value and S of reference correlation equation_calThe value may be implemented as a table of item number assignments (PNAs), another look-up table, etc. in the biosensor system. The slope deviation term can be normalized to give Δ S/S and the compensation equation rewritten as follows:

(equation 1A)

Other slope compensation equations may be used including at least one slope deviation value and the analysis output signal described above. Although equations set forth throughout this application and the claims may include an "═ symbol," the symbol is used to denote equivalence, relationship, prediction, and the like.

In the case of no compensationThe specific analysis output signal value will be provided from S_hypError compensation lines for samples with different analyte concentrations from S_calThe sample analyte concentration of the correlation line is referenced. From S_hypError compensation line derived A_corrThe values provide a more accurate value of the analyte concentration in the sample. Therefore, equations 1 and 1A use Δ S to convert the current values, S_calAnd Int to a compensated analyte concentration value A_corr。

If the value of Δ S is experimentally determined from the samples and substituted into equation 1 or equation 1A, the bias in the determined analyte concentrations for these samples will be fully compensated. Alternatively, if Δ S is replaced by a prediction function, the ability of the compensation equation to correct for the bias in the determined analyte concentration will depend on how relevant the value generated from the prediction function is to Δ S. Thus, for equation 1, Δ S may be replaced by a prediction function f (predictor), which may be rewritten as follows:

(equation 2)

Although the prediction function f (predictor) may have b₁*f(Index)+b₀But other values or exponents may be used in conjunction with the exponential function f (index) to provide f (predictor). For example, b may be present with or without₁(representing the slope) value and b₀The prediction function is provided using an exponential function f (index) with one or both of the (representative intercept) values. Therefore, when b₁1 and b₀When 0, f (predictor) is f (index). Multiple exponential functions may also be combined to provide f (predictor) to provide a corrected sample analyte concentration. The prediction or exponential function will better correct errors in the analysis when it has a greater correlation with slope deviation.

The prediction function includes at least one exponential function, one or more of which may be complex. The exponential function is responsive to at least one error parameter. The error parameter may be any value responsive to one or more errors in the output signal. The error parameter value may be determined before, during or after the analysis. The error parameters may be: a value from an analysis of the analyte, e.g. an intermediate signal from the analysis output signal; or from an auxiliary output signal independent of the analytical output signal, e.g. from thermocouple current or voltage, additional electrode current or voltage, etc. Thus, the error parameter may be extracted directly or indirectly from the analyzed output signal and/or obtained independently from the analyzed output signal. Other error parameters may be determined from these or other analysis output signals or auxiliary output signals. Any error parameter may be used to form one or more terms of the compositional exponential function, such as those described in international publication No. wo 2009/108239 entitled "Slope-Based Compensation," filed on 12/6 2008, and so forth. More detailed processing of error correction using exponential functions and slope deviation values can also be found in this publication.

The calculated number is generated from an exponential function related to an error parameter (e.g., hematocrit or temperature) that represents the effect of the error parameter on bias. The exponential function may be experimentally determined as a regression equation or other equation of a plot between the deviation from the reference slope and the error parameter. Thus, the exponential function represents the effect of the error parameter on the slope deviation, the normalized slope deviation, or the percentage bias. In normalization, slope deviation, exponential functions, or other parameters are adjusted (multiplied, divided, etc.) by variables to reduce the statistical impact of changes in the parameters, improve the variance of the variations in the parameters, normalize measurements of the parameters, combinations thereof, and so forth. In addition to referencing the correlation equation, the exponential function may be predetermined and stored in the biosensor system.

An exponential function is complex when it comprises at least two terms, each of said terms being modified by a weighting coefficient. Thus, the weighting coefficients of the complex exponential function provide the following capabilities: the relative significance of the plurality of error parameters is resolved in response to an amount of error that each error parameter contributes to the determined analyte concentration. Preferably, the combination is a linear combination, but other combining methods that provide weighting coefficients for the terms may be used. Each term may include one or more error parameters. More detailed processing on the use of predictive Functions and Complex Index Functions for analyte analysis can be found in international application No. pct/US2009/067150 entitled "complete Index Functions" filed 12, 8, 2009.

An example of a complex exponential function is represented as follows:

f(CIndex)＝a₁+(a₂)(Hct)+(a₃)(R4/3)+(a₄)(R5/4)+(a₅)(R6/5)+

(a₆)(R6/4)+(a₇)(Hct)(G_raw)+(a8)(R₄/3)(G_raw)+

(a₉)(R5/3)(G_raw)+(a₁₀)(R6/5)(G_raw)+(a₁₁)(R6/4)(G_raw)+

(a₁₂)(Temp)(Hct)+(a₁₃)(Temp)(R5/3)+(a₁₄)(Temp)(R6/5)

+(a₁₅)(Hct)(R5/4)+(a₁₆)(Hct)(R6/5)+(a₁₇)(Hct)(R6/4)

+.. (Eq.3)

Wherein, a₁Is a constant value, a₂To a₁₇Are respective independent weighting coefficients, G_rawIs the determined analyte concentration of the sample without compensation, Temp is the temperature, Hct is the current from the additional electrode. Weighting factor (a)₂To a₁₇) Each of which is immediately followed by its associated entry.

There are at least three basic types of terms in this complex exponential function: (1) respective ratio indices extracted from the analysis output signal, e.g., R3/2 and R4/3; (2) ratiometric index to temperature, Hct current and/or G extracted from the analysis output signal_rawThe interactive items between, e.g., (Temp) (R5/3) and (R4/3) (G_raw) (ii) a And (3) temperature, Hct or G_raw. These terms may include values other than error parameters, including G_rawInside it. When these terms are replaced with appropriate values, the complex exponential function generates a complex exponential value. Statistical processing may be performed on the plurality of terms to determine one or more constants and weighting coefficients. The statistical processing may be performed using statistical package software including MINITAB (MINTAB, inc., State College, PA).

One or more mathematical techniques may be used to select the terms to be included in the complex exponential function to determine the exclusion value for each potential term. One or more exclusion tests are then applied to the exclusion values to identify terms to be excluded from the complex exponential function. For example, the p-value may be used as a rankExcept for a portion of the test. Constant a₁May be determined by regression or other mathematical techniques. While a single constant is shown in the complex exponential function, a constant may not be required; more than one constant may be used, and the constant may be equal to 0. Thus, one or more constants may or may not be included in the complex exponential function. In forming the prediction function, it is also possible to use, for example, b described later₀Constants, or one or more constants, are combined with the complex exponential function.

The complex exponential function comprises at least two terms modified by weighting coefficients. The weighting coefficients are values other than 1 or 0. Preferably, each term comprising an error parameter is modified by a weighting factor. More preferably, each non-constant term of the complex exponential function is modified by a weighting coefficient. The weighting coefficients may have positive or negative values. The weighting coefficients may be determined by statistical processing of experimental data collected from combinations of multiple analyte concentrations, different hematocrit levels, different temperatures, and the like.

These slope-based compensation methods and other compensation methods can be paired with residual compensation in order to further improve the measurement performance of the biosensor system. By focusing on the residual error and finding the residual function associated with the residual error, the total error in the analysis can be reduced. Errors from a biosensor system may have multiple sources or causes of error due to different processes/behaviors that are partially or totally independent. By compensating the main errors (e.g., temperature and hematocrit) with a main compensation function to remove at least 50% of the total error, residual errors can be determined that remain, and residual functions associated with these residual errors can be determined. A more detailed discussion of residual Error compensation can be found in international application No. pct/US2011/029318 entitled "residual compensation Underfill Error", filed on day 22/3/2011.

Residual error compensation can substantially compensate for the total error in the analysis until the error becomes random. Random errors are errors that are not attributable to any cause of error and are not described by a residual function at a level statistically considered significant. Compensation from the combined main and residual functions may improve the measurement performance of the biosensor system in more than one way. For example, the combined main and residual compensation may improve the measurement performance of the biosensor system with respect to, for example, a percentage bias limit or a percentage bias standard deviation.

Residual error compensation may provide the greatest benefit to samples analyzed by the user himself during "self-testing". Residual error compensation may also provide benefits to samples analyzed by a Health Care Professional (HCP). While not wishing to be bound by any particular theory, it is believed that self-test errors may result from different behaviors or processes that are substantially independent of controlled environment or HCP test errors.

Fig. 4B illustrates an error compensation method including a transfer function 410, main compensation, and residual compensation. The output from the conversion function 410 including the total error 415 is compensated with a main compensation in the form of a main function 420. The remaining residual error 425 is compensated for using residual compensation in the form of at least a first residual function 430. The total error 415 includes a main error and a residual error. The total error 415 may also include random errors and/or other types of errors. The transfer function 410, the main function 420, and the first residual function 430 may be implemented as three separate mathematical equations, a single mathematical equation, and so on. For example, the transfer function 410 may be implemented as a first mathematical equation, and the main function 420 and the first residual function 430 may be combined and implemented as a second mathematical equation.

In fig. 4B, the unmodified output value 405 may be an output current generated in response to amperometry, voltammetry, coulometry, or other input signal used to generate an output signal having a current component. The output signal is responsive to the measurable species in the sample. The measurable species may be an analyte of interest or a mediator whose concentration in the sample is responsive to the concentration of the analyte of interest.

Preferably, the transfer function 410 is a correlation between the uncorrected output values 405 generated from the sample in response to the input signals from the measurement device and one or more reference analyte concentrations determined at known physical characteristics and environmental factors of the sample. For example, the sample may be a whole blood sample having a known hematocrit content of 42%, wherein the analysis is performed at a known constant temperature of 25 ℃. The correlation between the known sample analyte concentration and the uncorrected output signal value can be represented graphically, mathematically, a combination thereof, and the like. The correlation may be represented by a table of item numbers (PNAs), another look-up table, or the like, which is predetermined and stored in the measuring apparatus.

The master function 420 providing the master compensation may include a slope-based function, a complex exponential function, or other compensation functions that are focused on reducing errors in the analysis (e.g., temperature and hematocrit). For example, the total error observed for a biosensor system comprising a measurement device and a test sensor may be expressed in terms of Δ S/S (normalized slope deviation) or Δ G/G (relative glucose error). The master function 420 may compensate for at least 50%, preferably at least 60%, of the total error 415. The remaining analytical error in analyte concentration that is not compensated for by the master function can be considered to be due to operating conditions, manufacturing variations, and/or random errors. Since the master function 420 is a function, it may be expressed mathematically, for example using an equation, or may be represented by a look-up table that is predetermined and stored in the measurement device. The transfer function 410 may be mathematically combined with the main function 420 to provide a combined equation or look-up table. Suitable main Compensation techniques are described previously and may include additional details found, for example, in international publication No. wo 2009/108239 entitled "Slope-Based Compensation" and international application No. pct/US2009/067150 entitled "Complex Index Functions". Other main functions may be used.

When the sample is whole blood and the analyte is glucose, the compensation provided by the master function 420 may be substantially limited to compensation for analytical errors due to temperature and hematocrit. Thus, by characterizing the biosensor system with respect to temperature and hematocrit variation, the effects from temperature and hematocrit can be compensated by the master function 420. Preferably, other sources of error (e.g., operating conditions of the system) that are independent of temperature and hematocrit are not characterized and, therefore, are not included in the master function 420.

In addition to compensating for the main error using the main function 420, a first residual function 430 is applied that provides at least a portion of the residual compensation. Residual errors from error causes other than temperature and hematocrit may be identified and correlated with one or more exponential functions. The difference in error between the analysis performed in the controlled environment or by the HCP and the user self-test can be generally represented by the following equation: residual error is the total non-random error observed-the main function value. Thus, the residual error may be considered to be the non-random error and the manufacturing variation error minus the error expected to be compensated by the primary compensation (e.g., by the primary function).

The observed residual error is substantially free of errors that are removed from the total error by the value of the master function 420. The total error includes errors from substantially different sources and/or test cases, such as temperature and hematocrit errors determined in the controlled environment (substantially described by the master function), errors with respect to operating conditions originating outside the controlled environment (substantially described by the residual function), and manufacturing variations. The first residual function 430 may compensate for at least 5%, preferably at least 10%, more preferably at least 20% of the total error 415. At the same time, the main function 420 and the first residual function 430 may compensate for at least 60%, preferably at least 70%, of the total error 415.

The residual error remaining after application of the first residual function 430 may be further reduced if the second residual function is applied. Although the errors described by the second residual function may be from a controlled environment or a non-controlled environment, preferably these errors are non-random errors remaining after the main compensation and/or errors remaining after the main function and the first residual function compensation. For example, the second residual function may be selected to compensate for errors caused at extreme temperatures and/or sample hematocrit levels (e.g., at 5 ℃ and 70% Hct). Thus, the second residual function may be selected to compensate for errors outside the normal condition range of the main function or the main function and the first residual function. The second residual function may also be selected to compensate for system imperfections in the compensation provided by the main function or by the main function and the first residual function. Additional information about the second Residual function can be found in international application No. pct/US2011/029318 entitled "Residual Compensation adapting Underfill Error".

In addition to including the main compensation and at least one residual compensation, the error compensation method shown in fig. 4B may include the following capabilities: the compensation provided by the main compensation is adjusted with respect to the compensation provided by the residual compensation. When more than one residual function is used, residual compensation may also include the following capabilities: the compensation provided by the first residual function and the second residual function is adjusted. The error compensation provided by the main compensation in relation to the compensation provided by the residual compensation may be adjusted, since the function(s) constituting the residual compensation may be taken from predetermined values stored as a database in the measuring device or for a limited temperature and/or hematocrit range, whereas the main function may be determined from the temperature and the full range of hematocrit. Thus, the main function may be determined from inputs taken during analysis of the sample, while a limited number of residual functions may be predetermined and stored in the measurement device. The error compensation provided by the main compensation in relation to the compensation provided by the residual compensation may also be adjusted, since some overlap may occur between the errors described by the main function and the one or more residual functions. There may be other reasons to adjust the error compensation provided by the main compensation in relation to the compensation provided by the residual compensation.

The general form of compensation that adjusts the error compensation provided by the main compensation with respect to the compensation provided by the residual compensation can be expressed as: the master function + WC residual function, where WC is the residual weighting coefficient. The residual weighting coefficient WC may be chosen as a function for varying the temperature and/or hematocrit of the compensating contribution from the residual function. Similarly, compensation comprising one or more residual functions, each of which is modified by a residual weighting coefficient, may take the following general form:

compensated analyte concentration-current nA/(Slope)_Cal(1+ master function + WC1 residual 1+ WC2 residual 2.) (equation 4)

Or use an alternative general form of residual:

compensated analyte concentration-current nA/(Slope)_Cal(1+ primary function) ((1 + WC 1) × (1+ WC 2) · residual 2) ·(equation 5)

Where WC1 and WC2 are residual weighting coefficients having values between 0 and 1, and WC1 and WC2 allow reducing or eliminating the effect of the residual function when conditions are outside those used to spread the residual function. Residual 1 is the first stage of Residual compensation after the main compensation function, while Residual 2 is the next stage of Residual compensation, but may not be available if no source of error/exponential function is found. Preferably, Residual 1 and Residual 2 are independent of each other and independent of the master function.

The weighting coefficients of the main compensation with respect to the residual compensation and/or the weighting coefficients of one or more residual functions may be predetermined and stored in the measuring means in the form of a table or by other means. For example, WC1 values and WC2 values may be characterized in a two-dimensional table as a function of temperature and hematocrit. In this way, a table of weighting coefficients may be constructed to improve the measurement performance of the biosensor system by reducing the effect of the residual function(s) on the determined analyte concentration when the hematocrit content of the sample and the temperature at which the analysis is performed are closer to the conditions used to obtain the data used to determine the conversion function 410.

FIG. 5A illustrates an analytical method 500 for determining an analyte concentration in a sample using a binary underfill management system. In step 502, the biosensor system is enabled. In step 504, the biosensor system applies a regular polling sequence of polling signals to the sample. In step 506, the biosensor system detects the presence of the sample in the test sensor. In step 508, the biosensor system applies an extended polling sequence of polling signals to the sample. In step 510, the underfill recognition system detects whether the test sensor is initially SFF. If so, the underfill management system proceeds to step 514; if not, the underfill management system proceeds to step 512. In step 512, the biosensor system requests a supplemental sample and returns to step 510 to detect whether the test sensor is SFF. Although not shown, if the test sensor is still underfilled, step 512 may be repeated. In step 514, the biosensor applies a test excitation signal to the sample. In step 516, the biosensor measures an output signal in response to a redox reaction of a measurable species in the sample. In step 518, an underfill-corrected analyte concentration of the sample is determined based on the initial or subsequent SFF compensation equation and the output signal. In step 520, the analyte concentration may be displayed, stored for future reference, and/or used for additional calculations.

In step 502 of fig. 5A, the biosensor system is enabled. The system may be enabled by a power switch or button, a sensing mechanism for determining whether the measurement device is touched or held by a user, another mechanism for determining when a test sensor is placed within the measurement device, and the like. Upon activation, the biosensor is substantially ready to receive a sample and to determine the concentration of one or more analytes in the sample.

In step 504 of fig. 5A, the biosensor applies a regular polling sequence of polling signals to the sample. There may be one or more regular polling sequences in the polling signal. Fig. 3A and 3C both show regular polling sequences of polling signals of a binary underfill management system. Other regular polling sequences and polling signals may be used.

In step 506 of fig. 5A, the biosensor detects when the biologic fluid sample is available for analysis in the test sensor. When no sample is present, the biosensor continues the regular polling period, loops through one or more regular polling periods, starts or restarts the regular polling period, disables the biosensor, enters a sleep mode, combinations thereof, and the like. The biosensor detects the presence of a sample when at least one of the regular output pulses reaches a sample threshold and then applies an extended polling sequence. The biosensor may display the sample output signal on a display and/or may store the sample output signal in a storage device.

In step 508 of fig. 5A, the biosensor applies an extended polling sequence of polling signals to the sample. The biosensor may apply the extended polling sequence immediately at the end of the regular polling sequence, or after a transition period or at another selected time. "immediately" includes little or no time transition from the regular polling sequence to the extended polling sequence. There may be one or more extended polling sequences in the polling signal. Fig. 3A and 3C illustrate an extended polling sequence of polling signals suitable for use with a binary underfill management system. Other extended polling sequences and polling signals may be used.

In step 510 of FIG. 5A, the biosensor system detects whether the test sensor is SFF. If the test sensor is not SFF, the analysis moves to step 512. If the test sensor is SFF, the analysis moves to step 514. As previously discussed, one or more thresholds may be used to determine whether the test sensor is initially SFF. Values other than the threshold from the polling output signal may also be used.

In step 512 of fig. 5A, the biosensor system requests addition of a supplemental sample. The biosensor generates one or more error signals or other indicators to the user. Indicators on the measuring device or elsewhere may indicate that the sample size is not large enough for the user, for example, using icons, flashing lights, light emitting diodes, audio sounds, text messages, etc. The indicator may also indicate that the sample volume is not large enough for the biosensor, which may perform some function or action in response to the insufficient sample volume, such as stopping the analysis, restarting the polling signal, disabling the biosensor, and so forth. The biosensor system may generate the one or more indicators immediately after detection and/or prior to analysis of the analyte. The one or more indicators may be displayed on a display device and/or maintained in a storage device.

In step 514 of fig. 5A, the biosensor system applies an analytical test excitation signal to analyze the measurable species in the sample. The biosensor applies a test excitation signal to the sample. The test excitation signal may be applied immediately after the extended polling sequence of polling signals. The test excitation signal may be applied for a selected period of time after an extended polling sequence of polling signals. The test excitation signal may be a gated amperometric excitation signal or another excitation signal.

In step 516 of fig. 5A, the biosensor system measures an analytical output signal in response to a redox reaction of the measurable species in response to the analyte concentration in the sample. The sample generates one or more analytical output signals in response to the test excitation signal. The biosensor may measure the output signal continuously or intermittently. For example, the biosensor may intermittently measure the output signal during the pulses of the gated amperometric excitation signal, resulting in a plurality of current values recorded during each pulse. The system may display the output signal on a display and/or may store the output signal or portions of the output signal in a storage device.

In step 518 of fig. 5A, the biosensor system selects a compensation system in response to whether the test sensor is an initial SFF or a subsequent SFF. The compensation system is selected in response to at least one parameter associated with the polling signal. The parameters associated with the polling signal may include the time of the regular polling sequence, the time of the extended polling sequence, the current or voltage values of the regular polling output signal, the current or voltage values of the extended polling output signal, and so forth. The biosensor system correlates an output signal responsive to the analyte concentration in the sample with the analyte concentration in the sample and compensates responsive to the initial fill state of the test sensor.

Although the binary underfill management system of the analysis method 500 of FIG. 5A uses polling underfill recognition, the method 500 may similarly be implemented using a sequential detection underfill recognition system as previously described. Instead of applying a polling sequence in step 504, a relatively short pulse width voltage would be applied to the electrodes and the output current measured. Thus, the polling sequence of steps 504 and 508 would be replaced with a relatively short pulse width of voltage applied across the continuous electrodes, and the output current would be measured to determine which electrode pairs contacted the sample, and optionally the time required for the sample to pass through the continuous electrodes. In step 506, the presence of the sample is detected when the output current reflects that the sample is contacting the subelement of the counter electrode and the working electrode. If, in step 510, the presence of a sample is detected, but sufficient sample contact with the working and counter electrodes is not detected, the method proceeds to step 512 and replenishment of the sample is requested. If sufficient sample contact with the working and counter electrodes is detected in step 510, then the method will move to step 514 since the test sensor is of the initial SFF. The other parts of the analysis method 500 will be performed similarly to the polling method.

FIG. 6A illustrates an analytical method 600 for determining an analyte concentration in a sample using an underfill management system that determines an initial degree of underfill. The method 600 uses polling to identify an initial underfill level. In step 602, the biosensor is enabled. In step 604, the biosensor system applies a regular polling sequence of polling signals to the sample. In step 606, the biosensor system determines the presence of the sample in the test sensor. In step 608, the biosensor system applies an extended polling sequence of polling signals to the sample with the ability to discern an unfilled volume. In step 610, the underfill recognition system detects whether the test sensor is an initial SFF, or an initial HUF, or an initial LUF. If it is an initial SFF, the underfill management system proceeds to step 614; if it is the initial HUF or LUF, the underfill management system proceeds to step 612. In step 612, the biosensor system requests a supplemental sample and returns to step 610 to determine if the test sensor is SFF. Although not shown, if the test sensor is still underfilled, step 612 may be repeated. In step 614, the biosensor applies an analytical test stimulus signal to the sample. In step 616, the biosensor measures an output signal in response to a redox reaction of a measurable species in the sample. In step 618, a compensated analyte concentration of the sample is determined from the initial SFF compensation equation, the initial HUF compensation equation, or the initial LUF compensation equation and the output signal. In step 620, the analyte concentration may be displayed, stored for future reference, and/or used for additional calculations.

In fig. 6A, a biosensor activation step 602, a polling signal application step 604, a sample detection step 606, an extended polling sequence application step 608, a request for replenishment sample step 612, and an analyte concentration display, storage, and/or further processing step 620 may be implemented similarly to the corresponding steps in fig. 5A. As discussed previously, extending the polling sequence will allow more than one volume threshold to be met.

In step 610 of FIG. 6A, the biosensor system determines whether the test sensor is an initial SFF, HUF, or LUF. Different thresholds may be used to distinguish between the initial SFF, HUF and LUF states. For example, when the output from the extended polling sequence satisfies a first threshold, the test sensor is considered an initial LUF. If the extended polling sequence output satisfies the second threshold, the test sensor is considered an initial HUF. If the extended polling sequence output satisfies the third threshold, the test sensor is considered to be the initial SFF. The first, second and third thresholds are responsive to a fill state of the test sensor. For example, the LUF threshold may be met when 40% to 50% of the volume of the test sensor is filled, and the HUF threshold is met when 58% to 70% of the volume of the test sensor is filled. Values other than the threshold from the polling output signal may also be used to determine the initial fill state of the test sensor. Other percentages of test sensor fill may be selected to correspond to the initial fill states LUF, HUF, and SFF.

In step 618 of FIG. 6A, the biosensor system selects a compensation system in response to whether the test sensor is of an initial SFF or is subsequently filled to an SFF after an initial HUF or LUF state. The compensation system is selected by the underfill recognition system in response to at least two parameters associated with the polling signal. The underfill management system correlates the output signal responsive to the analyte concentration in the sample with the analyte concentration in the sample and compensates in response to the initial fill state of the test sensor.

Although the underfill management system of the analysis method 600 of FIG. 6A uses polling to determine the degree of underfill, the method 600 may similarly be implemented using a sequential detection underfill recognition system as previously described. Thus, the polling sequence of steps 604 and 608 would be replaced with a relatively short pulse width of voltage applied across the successive electrodes and the output current would be measured to determine which electrode pairs contacted the sample and optionally the time required for the sample to pass through the successive electrodes. The other parts of the analysis method 600 will be performed similarly to the polling method.

When the test sensor is an initial SFF, the underfill management system performs an initial SFF compensation. Slope-based compensation equations are preferred for the initial SFF compensation system. An example slope-based compensation of the initial SFF can be represented as follows:

(equation 6)

Here, f (index) temp is an exponential function representing the change in slope (Δ S) from the reference correlation attributable to the temperature error parameter, and f (index) hct is an exponential function representing the change in slope (Δ S) from the reference correlation attributable to the hematocrit error parameter.

More preferably, a slope-based compensation equation including a complex exponential function is used. The complex index function may combine the f (index) temp index function and the f (index) hct index function into a single mathematical form. The slope-based compensation equation for an initial SFF that includes a complex exponential function with a combined temperature and hematocrit function is previously represented as equation 3. More preferably, to also reduce the error introduced by user self-testing for the initial SFF test sensor, the underfill management system will implement the initial SFF compensation using a slope-based compensation equation that includes a complex exponential function that is the main function P1 in addition to the first and second residual functions R1 and R2. The initial SFF compensation equation including the main function P1 and the first and second residual functions may be generally expressed as follows:

A_comp＝i/[S_cal*(1+P1+WC₁*R1+WC₂*R2)](equation 7)

Here, A_compIs the compensated analyte (e.g., glucose) concentration of the sample, i is the current value, e.g., the final current value from the fifth excitation pulse shown in FIG. 2B, S_calIs the slope from the reference correlation equation, P1 is the master function, WC₁Is a first residual weighting coefficient, R1 is a first residual function, WC₂Is a second residual weighting coefficient and R2 is a second residual function. Although a second residual function is shown, it is not required.

The appropriate master function, first and second residual functions and their associated residual weighting coefficients used in equation 7 may be expressed as follows:

the main function P1 ═ 17.5252-0.012154 ═ i_7-Hct′-0.0258*′R3/2′-15.057*′R5/4′-20.04*′R6/5′+16.318*′R6/4′-5.1e-7*′i_7-Hct*G_raw′+0.0029343*′R4/3*G_raw′+0.01512*′R5/4*G_raw′-0.0191066*′R6/5*G_raw′-1.55e-6*′Temp*i_7-Hct′+0.030154*′Temp*R5/4′-0.006368*′Temp*R5/3′-9.476e-4*′i_7-Hct*R4/3′+0.011803*′i_7-Hct*R5/4′+8.112e-4*′i_7-Hct*R5/3′+0.013868*′i_7-Hct*R6/5′-0.01303*′i_7-Hct*R6/4′-9.1e-6*′i_7-Hct*R5/4*G_raw′+1.02e-5*′i_7-Hct*R6/5*G_raw′

(equation 8)

The first residual function R1 ═ 4.4084+5.683 ═ R4/3 ' -5.1348 × ' R5/4 ' -4.2282 × ' R5/3 ' -7.971 × ' R6/5 ' +7.40 × ' R6/4 ' +1.08e-5 × ' i ' of the first residual function_7-Hct*G_raw′-0.0015806*′R32*G_raw′-0.018626*′R43*G_raw′-0.044513*′R54*G_raw′+0.01978*′R53*G_raw′+0.04634*′R65*G_raw′+0.001481*′Temp*R32′+0.03006*′Temp*R54′-0.03737*′Temp*R64′-0.001453*′i_7-Hct*R43′+7.836e-4*′i_7-Hct*R53′+6.61e-4*′i_7-Hct*R65′+1.75e-5*′i_7-Hct*R54*G_raw′-2.89e-5*′i_7-Hct*R65*G_raw' (Eq. 9)

Wherein i_7-HctIs the current from the hematocrit sensing electrode at 7 seconds as shown in fig. 2B; temp is the measured device temperature; r3/2, R4/3, R5/4, R6/5, R5/3, and R6/4 are examples of inter-pulse ratio terms having the general format of the last current later in the time pulse divided by the last current earlier in the time pulse; and G_rawAre uncompensated analyte values.

When the test sensor is initially underfilled and then subsequently filled to SFF, the binary underfill management system will implement the subsequent SFF compensation. The binary underfill management system is generally configured to detect an initial HUF as opposed to an initial LUF state as an initial underfill, since the working electrode of the test sensor is generally in contact with the sample to indicate the presence of the sample in the binary system. Slope-based compensation is preferred for subsequent SFF compensation systems. An example of slope-based compensation for subsequent SFF can be represented as follows:

(equation 10)

Here, f (index) SubSFF is an exponential function representing the change in normalized slope deviation (Δ S/S) from a reference correlation attributable to the error introduced into the analysis by the initial underfill of the test sensor and the subsequent SFF.

More preferably, the subsequent SFF compensation system includes a slope-based compensation equation including a complex exponential function, wherein a different primary function P2 is used than when used for the initial SFF compensation. Although different residual functions may also be used, the residual functions may be less beneficial than for the initial SFF state, since errors attributable to self-testing are likely to be altered or reduced by subsequent padding. Thus, although different residual functions are preferred for each fill state determined by the underfill recognition system, they are not required.

The rationale for selecting different master functions P2 for subsequent SFF compensation is explained below for the initial HUF compensation system. In the case where the binary underfill recognition system determines underfill without the sample substantially contacting the working electrode, the initial LUF-type compensation system may be used for subsequent SFF compensation. The subsequent SFF compensation equation with a different primary function P2 for use with a binary underfill recognition system that detects an initial HUF type underfill may be expressed as follows:

different main functions P2 ═ 0.602-0.28941 ═ R3/2 ' -22.651 ═ R6/4 ' -9.204 ═ R7/6 ' +22.807 ═ R7/5 ' -26.617 ═ R8/7 ' +15.771 ' R8/6 ' -0.019103 ═ R4/3 ═ G_raw′+0.018181*′R5/3*G_raw′-0.009982*′R6/4*G_raw′+0.033009*′R8/7*G_raw′-0.022485*′R8/6*G_raw′+0.012486*′R3/2*Temp′+0.939*′R6/4*Temp′-0.9769*′R7/5*Temp′+0.56133*′R4/3*EPF_WE′-1.1673*′R5/4*EPF_WE′+0.57756*′R7/6*EPF_WE′-0.002448*′R4/3*G_raw*EPF_WE′+0.005993*′R5/4*G_raw*EPF_WE′+0.009662*′R6/5*G_raw*EPF_WE′-0.0013429*′R6/4*G_raw*EPF_WE′-0.011844*′R7/6*G_raw*EPF_WE' (EQUATION 10A)

Wherein, EPF_WEIs an extended polling factor that represents an underfill condition where the working electrode is significantly contacted by the sample. In case of sequential detection of an underfill recognition system, it may be directed to EPF_WEUsing Sequential Detection Factor (SDF)_WE)。

Fig. 7A, 7B, 7C, and 7D illustrate a comparison between an uncompensated glucose analyte concentration and a compensated glucose analyte concentration determined from a whole blood sample including red blood cells when the test sensor is initially underfilled and subsequent SFF whole blood. The test sensor is initially filled with a sample volume of less than 0.5 ml to create an underfill test sensor, where 0.5 ml is the SFF volume of the sample reservoir of the test sensor. A supplemental sample was added to the underfill test sensor to provide a subsequent SFF test sensor, and the glucose concentration of each sample was then determined. These readings are also compared to the readings from the sensor of the initial SFF.

FIG. 7A shows the Δ S value (Δ S) before compensation with subsequent SFF compensation including an exponential function relating the ratio error parameter (R7/6) to slope_uncomp) And a compensated Δ S value (Δ S)_comp) The correlation between them. The ratio error parameter R7/6 represents the relationship between the analytic output signal currents generated by the measurable species in response to the 6 th and 7 th pulses of the gated amperometric test excitation pulse sequence comprising at least 7 pulses. Other output signal currents and pulse references may be used. The ratio error parameter R7/6 is an example of an error parameter determined from analyzing the output signal. The exponential function relating the ratio error parameter R7/6 to the slope may be selected from various exponential functions also relating other error parameters to the slope.

FIG. 7B shows the percentage bias (% -bias) values for a plurality of uncompensated and compensated analyses for a subsequent SFF test sensor and an initial SFF test sensor when the correlation of FIG. 7A is used as an exponential function in accordance with equation 10. Fig. 7D shows similar data when the exponential function relating the ratio error parameter (R7/6) to the slope is replaced with the complex exponential function of equation 10A used as a different master function. The diamond symbols correspond to the bias values of the determined analyte concentration for uncompensated subsequent SFF, while the square symbols correspond to the bias values of the analyte concentration for compensated subsequent SFF. The determined analyte concentration of the test sensor from the initial SFF is identified on the right side of the graph. The remaining readings were from the initial underfill of the test sensor with a second fill and subsequent SFF prior to analysis.

FIG. 7C illustrates the percentages of uncompensated and compensated determined glucose analyte concentrations that fall within the + -15% percent bias limit when the test sensor is initially underfilled and the subsequent SFF is for analysis. The right side of the graph shows: initial fills of about 0.4 ml or greater do not benefit from a subsequent SFF compensation system. Thus, for this underfill management system, an underfill recognition system may be set to account for about 0.4 ml of SFF. A volume of about 0.45 ml may also be selected as SFF because there are no disadvantages to using an initial or subsequent SFF compensation system in the volume range of about 0.4 to about 0.45 ml. The underfill recognition system is configured to recognize a sample volume of about 0.25 milliliters as present but initially underfilled.

The darker line in fig. 7C shows the percentage of determined analyte concentration that falls within the ± 15% percent bias limit when the test sensor is initially underfilled, subsequent SFF, but no subsequent SFF compensation is applied. The lighter line shows the percentage of the determined analyte concentration that falls within the ± 15% percent bias limit when the test sensor is initially underfilled, the subsequent SFF, and the subsequent SFF compensation is applied. The smaller the initial underfill volume, the greater the improvement provided by the subsequent SFF compensation. At the lowest initial underfill volume of about 0.25 ml, only 63% of the uncompensated glucose readings fall within the ± 15% percent bias limit, while 96% of the compensated glucose readings fall within the ± 15% percent bias limit.

Fig. 7E shows the measured performance provided by the binary compensation system when the test sensor is analyzed for initial underfill and subsequent SFF and the subsequent fill follows the initial fill for up to about 30 seconds. The X-axis of the graph shows the time delay between an initial sample fill of the test sensor and a subsequent sample fill of the test sensor. A subsequent fill delay of about 3 seconds to about 30 seconds is used. In this example, the complex exponential function of equation 10A provides comparable measurement performance to equation 10 when used with an exponential function that relates the ratio error parameter (R7/6) to the slope.

When the test sensor is initially underfilled and then subsequently SFF's, the underfill management system, which is able to determine the initial degree of underfill, will implement either the initial LUF compensation or the initial HUF compensation. This capability may improve the measurement performance of the biosensor system, particularly for test sensor underfill volumes that exhibit little improvement when compensated with a subsequent SFF compensation system of the binary underfill management system. For example, the volume range from about 0.35 to about 0.42 milliliters of the test sensor described with reference to fig. 7C may benefit from an underfill compensation system that is different from that used for about 0.25 to about 0.35 underfill volumes. Although two initial underfill levels LUF and HUF are described below, the underfill management system may determine and manage other initial underfill levels.

Preferably, the initial LUF compensation system includes the same master function P1 as was used when testing the initial SFF of the sensor. Preferably, however, the master function P1 is paired with at least a first residual function that is different from the first residual function used for the initial SFF test sensor. Therefore, P1 is preferably used with a different first residual function R3. The initial SFF second residual function may be used, a second residual function different from the initial SFF second residual function may be used, or the initial LUF compensation system may not use the second residual function.

Although different master functions may be used, the master function P1 from the initial SFF state is preferred for the initial LUF compensation system because the initial sample fill does not substantially react with the reagent of the working electrode for the initial LUF state. Preferably, different first residual functions are used in the initial LUF compensation system to account for the substantial impact of self-test type errors on the analysis due to the initial LUF. While not wishing to be bound by any particular theory, the initial LUF state may be considered a serious self-test type error. The preferred initial LUF compensation equation can be expressed as follows:

A_comp＝i/[S_cal*(1+P1+WC₁*R3)](equation 11)

Here, A_compIs compensation of the sampleThe latter analyte (e.g., glucose) concentration, i is the current value, e.g., the final current value from the fifth excitation pulse shown in FIG. 2B, S_ca1Is the slope from the reference correlation equation, P1 is the primary function previously expressed as equation 8, WC₁Is a first residual weighting coefficient and R3 is a different first residual function. Although a different second residual function is not used, a different second residual function may be included. Preferably, the master function P1 will compensate about 90% of the total non-random error in the analysis, while the first different residual function will compensate the remaining 10% of the non-random error.

A suitable different first residual function R3 used in equation 11 may be expressed as follows:

different first residual functions R3 ═ -11.8098+0.0039471 ×' i_7-Hct′-0.46222*′R2/1′+9.2972*′R4/3′+6.4753*′R5/4′-9.0922*′R5/3′+5.6898*′R6/5′-0.00000113*′i_7-Hct*G_raw′-0.00034435*′R2/1*G_raw′+0.0024328*′R4/3*G_raw′-0.0034962*′R5/3*G_raw′+0.0022624*′R6/5*G_raw′-0.052217*′Temp*R4/3′+0.046291*′Temp*R5/3′+0.00024631*′i_7-Hct*R2/1′-0.0057016*′i_7-Hct*R4/3′+0.0056713*′i_7-Hct*R5/3′-0.0041934*′i_7-Hct*R6/4′+0.00000085*′i_7-Hct*R6/5*G_raw'+ 0.0040847 × SDF R2/1' +0.025846 × SDF R4/3 '-0.032782 × SDF R5/4' (equation 12)

Where SDF is a sequential detection factor representing an underfill condition when the working electrode is not significantly contacted by the sample. In the case of polling the underfill recognition system, an Extended Polling Factor (EPF) may be used for SDF.

Fig. 8A, 8B, 8C and 8D show the measured performance of the LUF compensation system using the main function of equation 8 and the different first residual functions of equation 12. About 100 test sensors were of the initial SFF and glucose concentrations were determined with and without the initial SFF compensation system as described previously. About 600 test sensors were initially filled to a volume of LUF, which for these test sensors was about 0.25 ml, and subsequently filled to SFF before glucose concentration was determined with and without the LUF compensation system. Whole blood samples analyzed for glucose include samples representing the full range of glucose concentrations, hematocrit content, and analysis temperatures.

Fig. 8A shows the measured performance provided by the LUF compensation system when the test sensors of the initial and subsequent SFFs were analyzed and the subsequent fill followed by the initial fill for up to almost 40 seconds. The X-axis of the graph shows the time delay between an initial sample fill of the test sensor and a subsequent sample fill of the test sensor. A subsequent fill delay of about 3 seconds to about 35 seconds is used. For example, analysis 801 is an uncompensated determined glucose concentration from the following analysis: in this analysis, the test sensor was an initial LUF and a subsequent SFF after about 30 seconds from initial fill. Fig. 8B shows the measured performance according to the data set of fig. 8A and using the LUF compensation system for whole blood samples including hematocrit contents of about 20%, 40% and 55% (volume/volume). Fig. 8C shows the measured performance for samples analyzed at approximately 15 ℃, 22 ℃, and 35 ℃ also according to the same dataset described above and utilizing the LUF compensation system. FIG. 8D shows the measured performance using the LUF compensation system for samples having glucose concentrations of approximately 50mg/dL, 75mg/dL, 330mg/dL, and 550 mg/dL.

Table I below summarizes the measured performance results of the test sensors for the initial and subsequent SFFs without compensation and with the use of the LUF compensation system. Table I also summarizes the overall performance results of the initial SFF test sensors without compensation and with the initial SFF compensation system utilized for comparison. Table I shows the mean percent bias determined from 596 initial LUFs and subsequent SFFs and from 112 initial SFF test sensors and the percent bias standard deviation determined therefrom. The percentages of the analysis falling within the ± 5%, ± 8%, ± 10%, ± 12.5% and ± 15% percentage bias limits relative to the reference glucose concentration of the blood sample determined using the YSI reference instrument are also shown.

TABLE I

For about 600 or fewer test sensors, the use of a LUF compensation system for test sensors with an initial LUF and subsequent SFFs resulted in over 95% of the analysis being within the + -10% percent bias limit, over 85% of the analysis being within the + -8% percent bias limit, and over 75% of the analysis being within the + -5% percent bias limit. This shows that: there was an improvement of greater than 240% ((98.7-28.7)/28.7 x 100) at a ± 10% percent bias limit and an improvement of greater than 400% ((77.5-13.6)/13.6 x 100) at a ± 5% percent bias limit relative to an uncompensated analysis of test sensors from the initial and subsequent SFFs. In fact, similar or better compensated measurement performance was observed for test sensors of the initial and subsequent SFFs as compared to the initial SFF test sensor.

The use of a LUF compensation system for test sensors with an initial LUF and subsequent SFFs also provides a percent biased standard deviation of less than 5 for 600 or fewer analyses performed with 600 or fewer test sensors. This shows that: there was an improvement of greater than 80% ((23.05-4.02)/23.05 x 100) in percent bias standard deviation over uncompensated analysis.

These performance measurements were obtained using a LUF compensation system for whole blood samples having a hematocrit content of about 20% (v/v) to about 55% over a sample temperature range from about 15 ℃ to about 35 ℃ and for glucose concentrations in the range of about 50mg/dL to 500 mg/dL. The underfill management system provides these results for the following test sensors: the test sensor is an initial LUF and subsequent SFFs within 6 seconds or less from initial fill, within 15 seconds or less from initial fill, within 30 seconds or less from initial fill, and within 35 seconds or less from initial fill. Thus, the LUF compensation system provides a significant improvement in the measurement performance of the biosensor system for an initial LUF test sensor that is subsequently filled to SFF in about 40 seconds.

Preferably, the initial HUF compensation system includes a different master function P2 than the master function used when the test sensor is the initial SFF. Optionally, a different master function P2 may be paired with a first residual function that is different from the first residual function for the initial SFF test sensor. Thus, if the initial HUF compensation system includes a first residual function, P2 is used with a first residual function that is different from the first residual function for the initial SFF test sensor, however, the initial HUF compensation system may not use the first residual function. The initial SFF second residual function may be used, a second residual function different from the initial SFF second residual function may be used, or the initial HUF compensation system may not use the second residual function. If the first residual function is used with the initial HUF compensation system, it is different from the residual function for the initial SFF state because the main function has changed from P1 to P2, and the residual function is compensating for the substantially uncompensated error of the main function.

The different primary functions from the initial SFF state are preferred for the initial HUF state because for the HUF state, the initial filling of the sample begins to chemically react with the reagent of the working electrode to generate a measurable species. Thus, the measurable species is generated both before and after the subsequent sample fill is provided to the test sensor. This situation may lead to: when an analytical test stimulus is applied to the sample, there are more measurable species present in the sample when the initial HUF occurs than when the initial SFF occurs. Thus, the initial HUF state may provide the following relationship between the electrochemical redox rate of the measurable species and the potential analyte concentration of the sample during the analytical portion of the assay: this relationship is different from the relationship that occurs when the test sensor is in the case of the initial SFF. Thus, the initial HUF state of the future SFF may be considered a substantially different analysis than when the initial SFF test sensor was analyzed.

Although not preferred, the initial HUF compensation system may use the same master function P1 as was used when the test sensor was the initial SFF. However, in this example, a different first residual function will be used with respect to the initial LUF compensation system. In practice, this will likely result in this different first residual function taking over more compensation from the main function P1 than is desired, since the post-HUF analysis (post-HUF analysis) can be considered to be a substantially different analysis than the post-SFF analysis or post-LUF analysis. Thus, for using an initial SFF or LUF master function with a different first residual function, it would likely result in the different first residual function compensating for more than 10% of the non-random error in the uncompensated analyte concentration — a situation that is likely to be realistic but not preferred. Furthermore, this situation leads to: this different first residual function compensates the "defect" in the main function P1 more than the error in the analysis. This will likely result in the first residual function being less effective in compensating for errors in the analysis. Thus, while the initial HUF compensation system may use the initial SFF master function P1 alone or with a different first residual function P1, a different master function P2 is preferred for the initial HUF compensation system with or without a different first residual function. The preferred initial HUF compensation equation can be expressed as follows:

A_comp＝(i-Int)/[S_cal*(1+P2)](equation 13)

Here, A_compIs the compensated analyte (e.g., glucose) concentration of the sample, i is the current value, e.g., the final current value from the fifth excitation pulse shown in FIG. 2B, Int is the intercept from the reference correlation equation, S_calIs the slope from the reference correlation equation and P2 is a different master function than the one previously represented in equation 8. Although substantially because the reaction time of the sample with the working electrode reagent has been extendedThe self-test error is reduced so that a different first residual function is not used, but a different second residual function may be included.

The appropriate different master function P2 used in equation 13 may be expressed as follows:

different main functions P2-8.9398-0.0034873 ═ i_7-Hct′+0.09534*′Temp′+0.56865*′R1′-0.67442*′R2/1′+1.7684*′R5/3′-11.9758*′R6/5′-0.00029591*′i_7-Hct*R1′+0.00044337*′i_7-Hct*R2/1′+0.0024269*′i_7-Hct*R5/4′+0.0051844*′i_7-Hct*R6/5′-0.0038634*′i_7-Hct*R6/4′-0.00073925*′R2/1*G_raw′-0.00188086*′R3/2*G_raw′-0.033466*′R4/3*G_raw′+0.041547*′R5/3*G_raw′+0.040176*′R6/5*G_raw′-0.045438*′R6/4*G_raw′-0.061549*′Temp*R4/3′-0.31944*′Temp*R5/4′+0.30496*′Temp*R6/4′-0.0077184*′SDF_WE*R1′+0.0036398*′SDF_WE*R21′-0.0018913*′SDF_WER43' (equation 14)

Wherein R1 is the current ratio within the pulse (i)_1，5/i_1，1) Examples of items, and SDF_WEIs a sequential detection factor that represents an underfill condition when the working electrode is significantly contacted by the sample.

Fig. 9A, 9B, 9C and 9D show the performance of the HUF compensation system using the different primary functions of equation 14. About 100 test sensors were of the initial SFF and glucose concentrations were determined with and without the initial SFF compensation system as described previously. About 650 test sensors were initially filled to a HUF volume, which for these test sensors was about 0.43 ml, and subsequently filled to SFF before glucose concentration was determined with and without the HUF compensation system. Whole blood samples analyzed for glucose include samples representing the full range of glucose concentrations, hematocrit content, and analysis temperatures.

Fig. 9A shows the measurement performance provided by the HUF compensation system when the test sensors of the initial HUF and subsequent SFF are analyzed and the subsequent fill follows the initial fill for a maximum of almost 40 seconds. The X-axis of the graph shows the time delay between an initial sample fill of the test sensor and a subsequent sample fill of the test sensor. A subsequent fill delay of about 3 seconds to about 35 seconds is used. Fig. 9B shows the measured performance according to the data set of fig. 9A and using the HUF compensation system for whole blood samples including hematocrit contents of about 20%, 40% and 55% (volume/volume). Fig. 9C shows the measured performance from the same dataset as above and using the HUF compensation system for samples analyzed at approximately 15 ℃, 22 ℃ and 35 ℃. FIG. 9D shows the measured performance for samples having glucose concentrations of approximately 50mg/dL, 75mg/dL, 330mg/dL, and 550mg/dL using the HUF compensation system.

Table II below summarizes the measured performance results of the test sensors for the initial HUF and subsequent SFF without compensation and with the HUF compensation system. Table II also summarizes the overall performance results of the initial SFF test sensors without compensation and with the initial SFF compensation system utilized for comparison. Table II shows the mean percent bias determined from 648 initial HUFs and subsequent SFFs and from 108 initial SFF test sensors and the percent bias standard deviation determined therefrom. The percentages of the analysis falling within the ± 5%, ± 8%, ± 10%, ± 12.5% and ± 15% percent bias limits relative to the reference glucose concentration of the blood sample determined using the YSI reference instrument are also shown.

TABLE II

For about 650 or fewer test sensors, the use of the HUF compensation system for test sensors with initial HUFs and subsequent SFFs was such that over 95% of the analyses were within the + -10% percent bias limit, such that over 90% of the analyses were within the + -8% percent bias limit, and such that over 75% of the analyses were within the + -5% percent bias limit. This shows that: there was almost 200% ((98.6-34)/34 x 100) improvement at the ± 10% percent bias limit and almost 400% ((79-16.4)/16.4 x 100) improvement at the ± 5% percent bias limit relative to uncompensated analysis of test sensors from the initial HUF and subsequent SFFs. In fact, similar post-compensation measurement performance was observed for test sensors of the initial HUF and subsequent SFF compared to the initial SFF test sensor.

The use of the HUF compensation system with test sensors of the initial HUF and subsequent SFF also provides a percent biased standard deviation of less than 4 for 650 or fewer analyses performed with 650 or fewer test sensors. This shows that: there was greater than 80% ((24.18-3.46)/24.18 x 100) improvement in percent bias standard deviation compared to uncompensated analysis.

These performance measurements were obtained using a HUF compensation system for whole blood samples having a hematocrit content of about 20% (v/v) to about 55% over a sample temperature range from about 15 ℃ to about 35 ℃ and for glucose concentrations in the range of about 50mg/dL to 500 mg/dL. The underfill management system provides these results for the following test sensors: the test sensor is an initial HUF and a subsequent SFF in 6 seconds or less from initial fill, 15 seconds or less from initial fill, 30 seconds or less from initial fill, and 35 seconds or less from initial fill. Thus, the HUF compensation system provides a significant improvement in the measurement performance of the biosensor system for an initial HUF test sensor that is subsequently filled to the SFF in approximately 40 seconds.

Fig. 10A shows a schematic diagram of a biosensor system 1000 with an underfill management system. The biosensor system 1000 determines the analyte concentration in the sample. The biosensor system 1000 may be used to determine one or more analyte concentrations, such as alcohol, glucose, uric acid, lactate, cholesterol, bilirubin, free fatty acids, triglycerides, proteins, ketones, phenylalanine, enzymes, and the like, in a biological fluid, such as whole blood, urine, saliva, and the like. Although a particular configuration is shown, system 1000 may have other configurations, including configurations with additional components.

The underfill management system improves the accuracy and/or precision of the system 1000 in determining the concentration of an analyte in a sample after an initial underfill occurs. The underfill management system includes an underfill recognition system and an underfill compensation system. The underfill recognition system indicates when the sample of biological fluid has initially SFF or initially underfilled the test sensor reservoir 1008. If the test-sensor reservoir 1008 is initially underfilled, the underfill recognition system instructs the system 1000 to request a replenishment of the sample. The underfill compensation system compensates the analyte concentration for one or more errors in the analysis in response to the initial fill state of the reservoir 1008 as determined by the underfill recognition system.

The biosensor system 1000 includes a measurement device 1002 and a test sensor 1004. The measurement device 1002 may be implemented as a desktop device, a portable device, or a handheld device, among others. A handheld device is a device that can be held in a human hand and is portable. An example of a handheld device is available from Bayer healthcare, Inc. of Elkhart (IN), IndMeasurement device of Elite blood glucose monitor.

The test sensor 1004 has a base 1006, the base 1006 forming a reservoir 1008 having an opening 1012. An optional channel 1010 may provide fluid communication between the reservoir 1008 and the opening 1012. The reservoir 1008 and the channel 1010 may be covered by a lid (not shown) having a vent. The reservoir 1008 defines a partially enclosed volume. Reservoir 1008 may include components that help retain a liquid sample, such as a water-swellable polymer or a porous polymer matrix. Reagents may be deposited into the reservoir 1008 and/or the channel 1010. The reagents include one or more enzymes, mediators, binders, and other active or inactive species. The test sensor 1004 may have a sample interface 1014 in electrical communication with the partially enclosed volume of the reservoir 1008. The test sensors 1004 may have other configurations.

In an electrochemical system, sample interface 1014 has conductors that connect to working electrode 1032 and counter electrode 1034. The sample interface 1014 may also include conductors coupled to one or more additional electrodes 1036 from which auxiliary output signals from the one or more additional electrodes 1036 may be measured. The electrodes may be substantially in the same plane. The electrodes may be disposed on the surface of the substrate 1006 that forms the reservoir 1008. The electrodes may extend or protrude into the volume formed by the reservoir 1008. The dielectric layer may partially cover the conductors and/or electrodes. The mediator may be disposed on or near the working electrode and the counter electrode. The sample interface 1014 may have other components and configurations.

The measurement device 1002 includes circuitry 1016 connected to a sensor interface 1018 and optionally a display 1020. The circuit 1016 includes a processor 1022 coupled to a signal generator 1024, an optional temperature sensor 1026, and a storage medium 1028. The measurement device 1002 may have other components and configurations.

The signal generator 1024 provides electrical excitation signals to the sensor interface 1018 in response to the processor 1022. The electrical stimulus signals may include polling signals and analytical test stimulus signals used in underfill management systems. The electrical excitation signal may be transmitted by the sensor interface 1018 to the sample interface 1014. The electrical excitation signal may be an electrical potential or current, and may be constant, variable, or a combination thereof (e.g., when an AC signal is applied with a DC signal offset). The electrical excitation signal may be applied as a single pulse or in the form of multiple pulses, sequences or periods. The signal generator 1024 may also record a recording signal received from the sensor interface 1018 as a generation recorder.

An optional temperature sensor 1026 determines the temperature used during analysis of the sample. The temperature of the sample may be measured directly or calculated from the output signal, or it may be assumed that the temperature of the sample is equal or similar to the measured ambient temperature or the temperature of the measurement device 1002 used to implement the biosensor system 1000. The temperature may be measured using a thermistor, thermometer, or other temperature sensing device. Other techniques may be used to determine the sample temperature.

The storage medium 1028 may be a magnetic, optical, or semiconductor memory, or other processor-readable storage device, etc. The storage medium 1028 may be a fixed storage device or a removable storage device such as a memory card.

Processor 1022 implements the underfill management system and other data processing using processor-readable software code and data stored in storage medium 1028. The processor 1022 initiates the underfill management system in response to the presence of the test sensor 1004 at the sensor interface 1018, application of the sample to the test sensor 1004, user input, etc. Processor 1022 directs signal generator 1024 to provide electrical excitation signals to sensor interface 1018.

Processor 1022 receives and measures output signals from sensor interface 1018. The output signal may be an electrical signal, such as a current or a potential. The output signals include polling output signals used in the underfill management system. The output signal includes an analytical output signal generated in response to a redox reaction of a measurable species in the sample for determining an analyte concentration of the sample. Processor 1022 may compare the polling output signal to one or more polling thresholds, as discussed previously.

When the sample is not in the SFF reservoir 1008 as previously discussed, the processor 1022 provides an error signal or other indicator of an underfill condition. Processor 1022 may display the error signal on display 1020 and may store the error signal and associated data in storage medium 1028. Processor 1022 may provide an error signal at any time during or after the analyte analysis. Processor 1022 may provide an error signal when an underfill condition is detected and prompt the user to add more sample to test sensor 1004. Processor 1022 may stop the analyte analysis upon detection of an underfill condition.

Processor 1022 determines the analyte concentration after underfill compensation from the output signal using the correlation equation as previously described. The results of the analyte analysis may be output to display 1020 and may be stored in storage medium 1028. The correlation equation between analyte concentration and output signal and the compensation equation for an underfill compensation system may be represented graphically, mathematically, a combination thereof, and the like. The equation may be represented by a table of item numbers (PNAs), another lookup table, etc., stored in the storage medium 1028. The constants and weighting coefficients may also be stored in the storage medium 1028. Instructions related to the implementation of analyte analysis may be provided by computer readable software code stored in the storage medium 1028. The code may be object code or any other code that describes or controls the functions described herein. One or more data processes may be performed on the data from the analyte analysis, including the determination of decay rates, K constants, ratios, functions, etc. in processor 1022.

The sensor interface 1018 has contacts that connect to or are in electrical communication with conductors of the sample interface 1014 of the test sensor 1004. Electrical communication includes by wire, wireless, and the like. The sensor interface 1018 transmits electrical excitation signals from the signal generator 1024 through these contacts to the connector in the sample interface 1014. The sensor interface 1018 transmits output signals from the sample interface 1014 to the processor 1022 and/or the signal generator 1024.

Display 1020 may be analog or digital. The display 1020 may be an LCD, LED, OLED, vacuum fluorescent lamp, or another display suitable for displaying digital readings. Other displays may be used. The display 1020 is in electrical communication with the processor 1022. The display 1020 may be separate from the measurement device 1002, such as when it is in wireless communication with the processor 1022. Alternatively, the display 1020 may be removed from the measurement device 1002, such as when the measurement device 1002 is in electrical communication with a remote computing device, drug dosage pump, or the like.

In use, the biosensor system 1000 enables and performs one or more diagnostic routines or other preparatory functions prior to analysis of a sample. The sample interface 1014 of the test sensor 1004 is in electrical and/or optical communication with the sensor interface 1018 of the measurement device 1002. The electrical communication includes: allowing input and/or output signals to be communicated between contacts in sensor interface 1018 and conductors in sample interface 1014. The test sensor 1004 receives a sample, preferably a biological fluid in liquid form. By introducing the sample into the opening 1012, the sample is transferred into the volume formed by the reservoir 1008. The sample flows through optional channel 1010 into reservoir 1008, filling the volume and simultaneously venting the previously contained air. The liquid sample chemically reacts with reagents deposited into the channel 1010 and/or reservoir 1008.

Processor 1022 identifies when a sample of biological fluid is present or absent for analysis. Sample interface 1014 provides a sample output signal to sensor interface 1018. Processor 1022 receives the sample output signal from sensor interface 1018. Processor 1022 may display the sample output signal on display 1020 and/or may store the sample output signal in storage medium 1028. Processor 1022 detects the presence of a sample when the sample polling output signal reaches one or more sample thresholds or when electrical conduction occurs between two or more electrodes. Processor 1022 may detect the absence of a sample when the sample polling output signal does not reach one or more sample thresholds or when electrical conduction does not occur between two or more electrodes.

Processor 1022 detects when the sample SFF or underfilling reservoir 1008. Sample interface 1014 provides a volume output signal to sensor interface 1018. Processor 1022 receives the volume output signal from sensor interface 1018. Processor 1022 may display the volume output signal on display 1020 and/or may store the volume output signal in storage medium 1028. Processor 1022 compares the volume output signal to one or more volume thresholds. Processor 1022 identifies that the sample has been SFF reservoir 1008 when the number of sequential contacts or volume polling output signal reaches one or more volume thresholds. The processor 1022 recognizes that the reservoir 1008 is not filled with sample when the number of sequential contacts or the volume polling output signal does not meet one or more volume thresholds.

When the processor identifies that the reservoir 1008 is not filled with sample, the processor 1022 prompts the user to add additional sample to the test sensor 1004 before proceeding with the analyte analysis. When the volume output signal indicates that the reservoir 1008 is not SFF, the processor 1022 may provide an error signal or other indicator of an underfill condition. The error signal may include a request or symbol from a user to replenish the sample. When more sample is provided to the reservoir 1008 by subsequent filling after the reservoir 1008 is underfilled, a larger sample volume generates another sample output signal. Processor 1022 determines that a supplemental sample is present when the further sample output signal reaches the same threshold or a further sample threshold as described above.

When the processor 1022 recognizes that the reservoir 1008 is SFF, the processor 1022 directs the signal generator 1024 to apply an analytical test excitation signal to the sample. The sample generates one or more output signals in response to the test excitation signal. Processor 1022 measures output signals generated by the sample. From the measured output signals, processor 1022 determines the analyte concentration of the sample. Based on the initial fill state and any subsequent fill states determined by the processor 1022 during underfill recognition, the processor applies the appropriate underfill compensation. For example, if the underfill recognition system determines an initial SFF, processor 1022 applies an initial SFF compensation. Processor 1022 utilizes the at least one slope deviation value to adjust the output signal, the correlation between the analyte concentration and the output signal, and/or the underfill to compensate for the analyte concentration. The analyte concentration may be determined from the slope adjusted correlation and the output signal. As previously mentioned, normalization techniques may also be used.

While various embodiments of the invention have been described, it will be apparent to those of ordinary skill in the art that other embodiments and implementations are possible within the scope of the invention.

Claims (19)

1. A method for determining the concentration of an analyte in a sample, comprising the steps of:

determining a fill state of the test sensor;

signaling addition of a supplemental sample to substantially fill the test sensor;

applying an analytical test excitation signal to the sample;

generating at least one analytical output signal value responsive to the analyte concentration in the sample and the analytical test excitation signal;

compensating for underfill errors in the at least one analytical output signal value in response to a fill state of the test sensor; and

determining an analyte concentration in the sample from the at least one analytical output signal value and the compensation.

2. The method of claim 1, further comprising, prior to the step of determining the fill status of the test sensor, the steps of: detecting the presence of the sample in the test sensor.

3. The method of claim 1, wherein,

the step of determining the fill state of the test sensor comprises determining an initial fill state of the test sensor; and is

The step of compensating for underfill errors in the at least one analytical output signal value in response to the fill state of the test sensor is performed in response to an initial fill state of the test sensor.

4. The method of claim 1, wherein the step of determining the fill status of the test sensor comprises:

applying a polling sequence to the sample, the polling sequence comprising a regular polling sequence and an extended polling sequence, and the extended polling sequence comprising at least one different extended input pulse; or

Sample fill was sequentially detected.

5. The method of claim 4, wherein the at least one different extended input pulse comprises two or more different extended input pulses of reduced amplitude.

6. The method of claim 4, wherein the sequentially detecting comprises: determining when two different pairs of electrodes are in contact through the sample.

7. The method of claim 4, further comprising adjusting at least one reference correlation with at least one exponential function for said compensating, wherein said compensating for underfilling errors comprises determining said at least one exponential function from an error parameter, wherein said error parameter is derived from at least one output signal value.

8. The method of claim 7, wherein the at least one exponential function is responsive to a slope deviation between a reference correlation and a hypothetical sample analyte concentration, wherein the hypothetical sample analyte concentration is indicative of the analyte concentration in the sample in an error-free manner.

9. The method of claim 7, further comprising:

selecting the error parameter in response to the polling sequence or the sequential detection,

wherein the error parameter is a value responsive to a volume threshold.

10. The method of claim 9, wherein the error parameter is a value corresponding to a duration of time.

11. The method of claim 1, wherein the step of compensating for underfill errors comprises: a subsequent substantially full underfill compensation system that includes a primary function that is different from the primary function used for the initial substantially full compensation, the primary function being used to compensate for a primary error in the total error.

12. The method of claim 1, wherein the step of compensating for underfill errors comprises: an initial low volume underfill compensation system that includes a first residual function that is different from a first residual function used for initial substantially full compensation, the first residual function being used to compensate for residual errors in a total error.

13. The method of claim 12, wherein the initial low-volume underfill compensation system further comprises a master function used in the initial substantially full compensation to compensate for a master in total error.

14. The method of claim 1, wherein the step of compensating for underfill errors comprises: an initial high volume underfill compensation system that includes a master function that is different from a master function used for initial substantially full compensation, the master function being used to compensate for a master error in the total error.

15. The method of claim 1, wherein the sample is whole blood comprising red blood cells, the analyte is glucose, and more than 95% of the glucose concentration determined for 600 or fewer analyses falls within a ± 10% percent bias limit.

16. The method of claim 15, wherein more than 95% of the glucose concentrations determined for 600 or fewer analyses fall within a ± 10% percent bias limit when the test sensor is initially underfilled and subsequently substantially flooded within 35 seconds or less from initial filling.

17. The method of claim 1, wherein the sample is whole blood comprising red blood cells, the analyte is glucose, and more than 75% of the glucose concentration determined for 600 or fewer analyses falls within a ± 5% percent bias limit.

18. The method of claim 1, wherein the sample is whole blood comprising red blood cells, the analyte is glucose, and the glucose concentration determined for 600 or fewer analyses provides a percent bias standard deviation of less than 5.

19. A biosensor system for determining an analyte concentration in a sample, comprising:

a test sensor having a sample interface in electrical communication with a reservoir formed by the test sensor; and

a testing device having a processor connected to a sensor interface in electrical communication with the sample interface, the processor in electrical communication with a storage medium,

wherein the processor performs the method of claim 1,

the storage medium stores at least one reference correlation predetermined with a reference instrument, and

the measuring device is portable.