BRPI1001332B1 - NON-INVASIVE MEASUREMENT INSTRUMENT FOR CONTINUOUS MONITORING OF GLUCOSE CONCENTRATION IN ARTERIAL BLOOD - Google Patents

Abstract

Non-invasive measuring instrument for continuous monitoring of arterial blood glucose concentration. patent of invention belonging to the field of measurement for diagnostic purposes, comprised of instrument (1) consisting basically of light emitting LED (2), photodetector (3), non-invasive optical sensor (4) and electronic circuit (5) capable to excite the sensor's optical components, read analog signals from the sensor, estimate the glucose concentration value and clearly present the value of this measurement, or send this information by any known means to a device capable of receiving it, present it and possibly store it, thus allowing a follow-up of the evolution of the treatment. in the invention, which has a margin of error of less than 10%, two fixed wavelengths (of 805 and 1350 nm) are preferably used to traverse the tissue, and the measurement of glucose concentration is made exclusively in arterial blood - identified by its pulsatile characteristic - by analyzing the continuous and alternating part of the light absorbed by living tissue separately, with the alternating component normally not exceeding 1% to 2% of the constant absorbance for the wavelength of 805 nm, and 0.05 at 0.5% for the wavelength of 1350 nm.

Description

Application field

[0001] The present invention patent belongs to the field of measurement for diagnostic purposes, more specifically measurement of blood characteristics in vivo using optical sensors (International Patent Classification A61B-5/1455), and was developed to enable continuous control of glucose concentrations in a non-invasive, simple, practical and portable way, using a measurement process with totally new characteristics that presents a significantly smaller margin of error than the similar ones, besides being able to be used in human beings and animals.

Background of the Invention / State of the Art

[0002] As is well known, diabetes or diabetes mellitus is a metabolic disease characterized by an abnormal increase in the concentration of glucose (or sugar) in the blood, which can cause serious health problems (such as heart attack, stroke, kidney failure, visual problems and injuries that are difficult to heal, among other complications) and has a high incidence in the world population - more than 220 million people are diabetic, according to surveys carried out in 2009 by the World Health Organization.

[0003] The most traditional form of diabetes control is by measuring capillary blood glucose (blood glucose concentration), which requires the analysis of a blood sample obtained through transcutaneous puncture, and for proper control, five measurements are recommended daily glycemic rate so that the treatment can be adjusted to the individual needs of each patient. It should be clarified that the term “capillary blood” refers to blood that travels through very thin blood vessels between the arteries and veins, through which the blood delivers oxygen to the tissues.

[0004] However, as it is an invasive procedure, relatively painful and unpleasant, it has serious practical limitations due to low patient compliance and incomplete data collection with few values measured during the day. Due to these factors, we began to research more comfortable methods, processes and measuring instruments for continuous monitoring of blood glucose, as a single blood glucose test as currently performed in clinics or laboratories is too little to make a good assessment of the patient.

[0005] Self-monitoring, in all types of diabetes, is extremely important for the maintenance of normal blood glucose levels. The most accurate of the tests is the measurement of capillary blood mentioned, but other alternative methods can be used, such as urine measurement or the use of reagent strips, which, however, are indirect methods of assessing blood glucose since when signs appear urine glucose values are usually above 180 mg/dl in blood - when normal values are in the range of 80 to 110 mg/dl. “Negative” values can also hide very low levels, masking hypoglycemia.

[0006] The methods currently used in measuring blood glucose can be divided into three large groups: laboratory tests, exams "fingertip" and measurement by continuous semi-invasive sensors.

[0007] Laboratory tests, despite having a low margin of error (in the order of 2 to 3%), require a blood sample that is collected from the patient and later sent for analysis in a specific laboratory. There are several methods used to measure blood glucose concentration in laboratory tests, but in general, blood plasma is separated by centrifugation and application of specific enzymes, such as glucose oxidase or hexokinase. Enzymes cause chemical reactions that lead to glucose degradation, with the formation of some product that is easily measured, either by generating an electric current (oxi-reduction reaction) or changing the color of the mixture. This method, however, has several inconveniences in daily blood glucose control: it is invasive, requiring a large amount of blood when compared to other methods; it demands a long waiting time for the result (usually two days), and involves the need to move the patient to a specialized center for the examination.

[0008] Already the examination of "fingertip" is made from a small sample of capillary blood (one or two drops) collected through a small hole made in the patient's finger. The blood sample is applied to a tape which, in turn, is inserted into a portable device that can use different measurement methods. In general, similarly to laboratory tests, tapes with specific enzymes for glucose degradation are used, and the measurement of the formed products is carried out by the measuring device. This method allows the patient to undergo the exam in a quick and practical way, and from the beginning of the exam until the result is obtained, approximately 5 minutes are spent - which makes this method the most widespread and used among the diabetic patients, there are several types of devices for this purpose, such as those contained in documents PI0409488-3 of 20/04/2004 and PI9704928-0 of 30/09/1997 - which deal with methods of manufacturing test strips for measuring glucose in a blood sample for instruments of the "finger tip" type - and PI9903626-6 of 12/08/1999, which deals with a visual blood glucose test strip, provided with two membranes each with a reagent to cause a color change by analyzing a blood sample.

[0009] However, despite the small amount of blood required to perform this type of measurement, it is still an invasive test and, in addition, the accuracy of the reading can show deviations of the order of 15%. Another point to consider is the need for training of the patient who performs the exam, because according to their ability there can be a considerable variation in readings, possibly able to hide the need for specific control procedures and even lead to risk of worsening the condition of the clinician. It is also worth remembering that the need to use a lancet or needle to collect the blood needed to make the dosage means that some patients, for psychological or economic reasons, do not choose to use this method.

[0010] In recent years, the first continuous blood glucose measurement devices have emerged that use semi-invasive or non-invasive techniques, usually obtaining readings through skin or subcutaneous sensors in order to minimize patient discomfort - for example , presented in document PI0306653-3 of 10/22/2003, which deals with an implantable sensor for determining the concentration of glucose in the blood. The methods used by these sensors are quite diverse, but in general enzymatic sensors generate an electrical current proportional to the concentration of glucose in the blood through a chemical reaction of glucose oxidation, caused by an enzyme covering the electrode. It turns out, however, that such sensors need to be inserted through the skin to come into contact with the interstitial fluid and, therefore, they are also invasive. Furthermore, despite being considered “continuous”, such devices actually generate punctual readings in short time intervals - on the order of five to fifteen minutes.

[0011] Reverse iontophoresis is also another widely used method for evaluation by means of sensors, in which two electrodes placed on the skin, when subjected to a potential difference, force a flow of interstitial fluid to the surface of the skin to that the glucose concentration is measured using an enzymatic sensor. The instruments that use this method, however, have reading errors in the order of 15% to 20% when compared to those obtained through laboratory tests, which can reach, under certain conditions, an error of 50%.

[0012] Another measurement technique that has been much researched performs the reading through the emission of light through the blood, and some devices that use it are described in the following documents:

[0013] - US 4,890,621 of 01/19/1988: involves the step of separating a "clear liquid" from the blood of its constituent proteins and cells, where the glucose to be monitored is located; such clear liquid travels a predetermined path between a light emitter and a light sensing device, using the optical property to determine the blood glucose level. However, it uses an implantable sensor and is therefore an invasive technique;

[0014] - PI 9714562-9 of 06/05/1997: it is an instrument for measuring the intravascular blood glucose level (located inside any blood vessels) in a non-invasive way that uses light sources (four lengths different waveforms, with 650, 880, 940 and 1300 nm to illuminate the fluid), photodetectors and the transmitting DC levels of each light source;

[0015] - US 4,901,728 of 05/01/1989: deals with an instrument that uses infrared light emission for non-invasive determination of blood glucose, but involves optical rotation of polarized light to identify and measure the concentration of glucose;

[0016] - DE 10.2006.036.920 of 08/04/2006: describes a method that uses at least two wavelengths of radiated infrared light, measured in cycles that are not associated with cardiac cycles, but with measurement cycles;

[0017] - WO 9.300.855 of 07/03/1991: method for determining glucose by emitting a beam of infrared radiation whose wavelengths are chosen by filters (which are changed during the scanning measurement process - they are not fixed wavelengths) coming from a light source, such beams being polarized and reflected.

[0018] By the detailed analysis of the equipment presented in the cited documents and also several other documents that describe devices that use the non-invasive infrared light emission, it is noted that, in general, the measurement is made in veins or arteries, using one or several wavelengths with punctual values or situated between a predetermined range. Furthermore, they involve spectrophotometry - that is, there is an incidence of light after which angles, lags and lengths are measured. In a fundamentally didactic description, for the identification of glucose molecules, the known absorption values of other tissues in the medium - such as bone, vessels, epidermis, etc. - are subtracted from the total radiation captured by the photodetectors. - where the final value corresponds to a certain concentration of glucose. Another disadvantage of these instruments of the current technique lies in the high margin of error - situated between 10 and 15% and which, as previously mentioned, can cause several inconveniences.

Invention Presentation

[0019] In view of the demand for a non-invasive, accurate, portable, and easy-to-operate measuring device capable of encouraging users to control the blood glucose rate with the appropriate frequency, the non-invasive measuring instrument was developed for continuous monitoring of the arterial blood glucose concentration object of this application for privilege, comprised of a device that uses the pulsatile characteristic of blood to perform, by means of infrared light, the measurement of exclusively arterial blood glucose concentration by by analyzing the continuous and alternating part of the light absorbed by living tissue separately, the alternating component normally not exceeding 1% to 2% of the constant absorbance for the wavelength of 805 nm, and 0.05 to 0.5% for the wavelength of 1350 nm, which ends up significantly increasing the accuracy of the measurement resulting in values with a margin of error of less than 10% - ie. a, much smaller than that obtained through the use of similar portables in use today.

[0020] Thus, there is a portable instrument, easy to operate and extremely discreet, even allowing its use in public places without causing embarrassment to users, which can operate continuously, thus increasing the adherence of diabetic patients to the treatment. In addition, its functional characteristics allow it to be produced in different formats - for example, in the form of rings and/or earrings, and the improvement of existing technologies may in the future enable constructions in the form of bracelets or similar pieces with the same reliability and practicality, but always involving the transmission of the readings made to any devices that may be cell phones, portable computers or any “wireless” type receivers.

[0021] This highly portable instrument comprises a non-invasive optical sensor and an electronic circuit capable of exciting the sensor's optical components, reading analog signals from the sensor, estimating the glucose concentration value and clearly presenting the value of this measurement , or send this information by any known means to a device capable of receiving, displaying and possibly storing it, thus allowing a follow-up of the evolution of the treatment.

[0022] In the invention presented herein, two fixed wavelengths are preferably used to traverse the tissue, and the measurement will be made exclusively in arterial blood, identified through the use of the pulsatile characteristic of blood to eliminate all other components of the medium .

[0023] For illustrative purposes and for better understanding, the invention will be better described and detailed based on the attached drawings, in which:

[0024] Figure 1 illustrates the application of Beer's Law in arterial blood;

[0025] Figure 2 illustrates the absorption and transmission of light in living tissue;

[0026] Figures 3.1 and 3.2 respectively illustrate the "raw" signals and the normalization of the intensities of the light signals;

[0027] Figure 4 presents an overview of the system for measuring pulse blood glucose;

[0028] Figures 5.1, 5.2, 5.3 and 5.4 illustrate some of the possible forms of construction for the transmissive sensor, whereby light from two LEDs passes through the region of the body in which the measurement will be made (preferably on the finger or on the earlobe,) alternately, focusing on the photodetectors;

[0029] Figure 6 shows an example of a block diagram of the microcontrolled blood glucose plate;

[0030] Figure 7 presents an example of current voltage converter circuit;

[0031] Figure 8 presents an example of a time diagram of the LEDs excitation circuit;

[0032] Figure 9 presents an example of differential transimpedance amplifier;

[0033] Figure 10 presents an example of sampling and retention circuit;

[0034] Figure 11 presents an example of a time diagram of the multiplexing circuit, and

[0035] Figure 12 presents an example of programmable gain amplifier.

Detailed Description of the Invention

[0036] As illustrated in the above-related figures, the non-invasive measuring instrument for continuous monitoring of arterial blood glucose concentration object of this patent application is comprised of instrument (1) consisting basically by light emitting LED (2), photodetector (3), non-invasive optical sensor (4) and electronic circuit (5).

[0037] The measurement used in the instrument (1) proposed for continuous and non-invasive monitoring of arterial blood glucose concentration adopts the principles of absorbance and light transmittance in a medium, which can be better understood by studying Beer's Law -Lambert or Bouguer's Law, which describes the attenuation of light that propagates through a uniform medium containing absorbing substances.

[0038] For the determination of arterial blood glucose concentration by the proposed instrument, using the method of measuring the absorbance or transmittance of light in living tissues, several different wavelengths can be used. The pulsatile characteristic of arterial blood will also be used, in an unprecedented way until the current technological development of the sector, to distinguish between the absorbance of arterial blood and that of other absorbers.

[0039] To exemplify the method, this approach will adopt only two different wavelengths to carry out the measurement process, but this does not limit the proposed instrument to using several different wavelengths to obtain better measurement accuracy to be obtained.

[0040] Different reasons lead to the most common choice for the wavelengths used to determine the concentration of glucose in arterial blood. The red pigmentation of the skin absorbs a large part of the light at wavelengths below 600nm and therefore it is not appropriate to measure the absorbance of light in this range. As a reference, the wavelength of 805nm will be used, as this value has a low influence of hemoglobins, since the absorbance spectra of reduced hemoglobin and oxygenated hemoglobin are relatively flat and very close.

[0041] For significant variations in the intensity of light transmitted through glucose, one must have a wavelength in the range of 1300nm to 2400nm. A good choice for this wavelength is 1350nm, as there are commercial light emitting diodes for this value - which ultimately reduces manufacturing costs - and is within the recommended range.

[0042] It is worth mentioning that these wavelengths were used in a preferential assembly, but any wavelengths that meet the requirements of the method used to perform the measurement can be used without, thereby, escaping from the scope of the intended protection, which mainly involves the use of the pulsatile characteristic of blood as an element of distinction from the reading medium.

[0043] The light that propagates through biological tissues (such as the finger or the earlobe, for example) is absorbed by different absorbing substances. Light absorbers in the region of interest are pigmentation of the skin, bones, arterial blood and venous blood. Instead of measuring arterial blood glucose concentration in vitro with an arterial blood sample, the arterial pulse will be used. Figure 1 shows the amount of light absorbed and transmitted in living tissue in relation to distance.

[0044] Arteries contain more blood during systole than during diastole and, consequently, their diameter increases due to increased pressure. This effect only occurs in arteries and arterioles, but not in veins. The absorbance of light in tissues with arteries increases during systole, mainly because of the increase in the amount of absorbing substances and due to the fact that the optic path length d in the arteries increases. This alternating part of the total absorbance allows us to differentiate the absorbance due to venous blood, a constant amount of arterial blood, and other non-pulsatile components - such as skin pigmentation (continuous component of total absorbance) - from the relative absorbance of the pulsatile component of arterial blood (alternate component). The alternating part of the light absorbed by living tissue usually does not exceed 1% to 2% of the constant absorbance. The electrical signal resulting from transmitted light, which is time-varying, is known as a plethysmographic signal.

[0045] The intensity of light that propagates through a tissue during diastole is high, represented by IH, indicated in Figures 1 and 2. The absorbers that are present during diastole are the continuous components (DC). All DC components, except non-pulsatile arterial blood, can be represented using the extinction coefficient relative to the specific wavelength, substance concentration, and light path length.

[0046] The diameter of arterial vessels is minimal (dmin) and then the absorbance relative to arterial hemoglobin is minimal and the amount of transmitted light is high (IH) and has a peak as shown in Figures 1 and 2.

[0047] The length of the optical path in arteries increases during systole at dmax indicated in Figures 1 and 2. The amount of absorbed light finds its maximum value and the amount of transmitted light finds its minimum value IL indicated in Figures 1 and 2.

[0048] The light intensity I that falls on the photodetector (3) is a function of the diameter d of the arteries and arterioles. During a cardiac cycle this diameter changes from dmin to dmax. Substituting d by dmin + Δd, one can express I as a function of IH, Δd and the part of the diameter that changes from zero to dmax-dmin with respect to time.

[0049] The reading of the proposed instrument (1) will be an estimate of the concentration of glucose in arterial blood, and the measurement with two wavelengths allows to distinguish the concentrations of only two different absorbers (glucose and others).

[0050] But in animals are found other types of absorbers present in arterial blood, also varying their concentration. These other types absorb light in the same way as glucose and therefore influence the measurement. Thus, until measurements are made with as many wavelengths as there are absorbers in the blood, it will not be possible to determine the correct concentrations of glucose in animals.

[0051] Arterial blood glucose concentration can be obtained based on Beer's Law as a function of the ratio of absorbances at two wavelengths; but due to the non-linearity of the LED's (2), the photodetector (3) and the light absorption by tissues, the absorbances have to be normalized. This model results in a theoretical calibration curve, which is not used in practice.

[0052] The light intensity measured at a given wavelength has to be normalized before being compared with the result of the measurement performed at another wavelength, due to the fact that each LED (2) can emit light with different intensities, because the absorption characteristics of the continuous components and the sensitivity of the photodetector (3) are different for different wavelengths, as well as because the tissue absorption and the absorber length can vary greatly from patient to patient. The normalized signal In is calculated by dividing the transmitted light intensity by its maximum peak value (IH.1 for the light length of 805nm and IH.2 for the light length of 1350nm).

[0053] The normalized signals of the transmitted light intensity for the wavelengths are independent of the incident light level and the non-linearity of the photodetector (3), as shown in Figure 3. The AC component of the normalized signal represents only the variations of the Transmitted light caused by arterial blood pulsatility and can be compared with other AC components. These AC components depend on the absorbers present in the arterial blood (Glucose and Others) and on the optical length d, through the variation in the volume of the arteries.

[0054] The absorbance of light is obtained in a normalized way by calculating the natural logarithm of the normalized transmitted light intensity. Dividing the original signal by the intensity of light transmitted during diastole IH, and calculating the total absorbance, then calculate the total absorbance just because of the AC components. The R-ratio of these normalized absorbances at the wavelengths of 805nm and 1350nm (respectively represented by “V” and “IV” in Figure 6) depends only on the light absorbers present in the arterial blood.

[0055] Assuming that the optical path length d1 for the light length of 805nm and d2 for the light length of 1350nm are equal, and that only the arteries change their diameters, the following result is reached:

[0056] where "ε" represents the extinction coefficient of the substance, "X" the wavelength used in the measurement method and "C" the concentration of the substance.

[0057] Thus, the R ratio is not a function of the optical path length and can be obtained from the concentration of glucose in arterial blood.

[0058] The incident light that propagates through a human tissue is not only decomposed into absorbed light and transmitted light as proposed by Beer's Law. With some components of light, the phenomena of reflection and scattering occur, changing the result of this theoretical equation.

[0059] The surface of the skin, tissues, muscles, bone and especially blood cause light scattering, which increases the absorbance of light. Blood is not a homogeneous liquid and is capable of non-linear light absorption, for example, when glucose concentration varies. The variation in light absorption is not only due to the increase in the length of the optical path during systole, due to the change in the axis of the red cells, which also alter their absorption. Red cells have the shape of a biconcave disc, with its larger diameter aligned parallel to the direction of blood flow during diastole, while during systole it is aligned perpendicular to the direction of flow. As a result of these properties, the absorbance and reflectance of moving blood vary within a cardiac cycle and with the velocity of blood flow.

[0060] For this measurement method, it is suggested, for the most part, to use the R ratio, known as ratio of ratios or modulation ratio, which can be defined approximately by:

[0061] In this equation, the AC component is the peak-to-peak variation of the signal in heart rate and the DC component is the average of all transmitted light intensity, for the two light lengths used.

[0062] The results of measurements of arterial blood glucose concentration differ from the results obtained based on Beer's Law. A physical phenomenon called light scattering significantly increases the light absorption measurement result. However, the proposed instrument (1) estimates the concentration of glucose in arterial blood with sufficient accuracy for clinical use, with an error of less than 10%. This is achieved due to the fact that a calibration curve based on empirical data will be used, as the mathematical modeling of the effects of light scattering for different conditions is very complex.

[0063] A certain amount of data must be acquired in clinical studies for instrument calibration purposes, containing information of the R ratio calculated by the proposed instrument and the corresponding value of the glucose concentration measured using another more accurate meter. Tables or research equations should then be used to find the relationship of these two variables for reading the instrument (1) presented here.

[0064] To relate the calculated values of the R ratio with the glucose concentration value presented by the instrument (1), the theoretical calibration curve equation based on Beer's Law can be modified, resulting in:

[0065] In this equation, the extinction coefficients were replaced by the constants ki. Another approximation for the mathematical representation of the calibration curve is the use of a polynomial, that is, Glucose = k + k2R + KR2

[0066] Since the constants k1, k2 and k3 are determined through clinical studies, in order to obtain the best fit for the instrument's calibration curve (1).

[0067] The following is an example of an instrument (1) for measuring glucose in arterial blood just to illustrate the method, therefore, the instrument is not limited to the diagrams and circuits presented.

[0068] Measurements of arterial blood glucose concentration use two different wavelengths, as explained above. Two light-emitting diodes or LEDs (2) - one with a wavelength of 805 nm and the other with a wavelength of 1350 nm, for example - emit light through the tissues of a patient's finger (D) until it finds its respective photodetector (3). Each photodetector (3) has a response curve centered on the wavelength of the corresponding emitting diode. The LEDs (2) must be alternately activated in order to avoid cross influence between the emitters and their receivers. The signal coming from the photodetectors (3), proportional to the incident light coming from the LEDs (2), must be amplified separately. These signals are then filtered to remove noise from the switching frequency, the electrical network and the environment.

[0069] The continuous and alternating components of each signal coming from the photodetector (3) are then separated by means of specific filters and then amplified. There is also a separation of light interference from the environment, a signal that is amplified when the two LEDs (2) are off. The signals are then digitized by an analog to digital converter, being read by a microcontroller that estimates the glucose concentration value, displays this data and sends it to a computer for storage and analysis through a standard RS-232C serial port.

[0070] Figure 4 illustrates a sensor (4) with two light-emitting diodes and two photodetectors aligned, one set for the wavelength of 805 nm and the other for 1350 nm. From this definition, there is a micro-controlled board (5) containing the necessary circuits to control the sensor LEDs, condition the signal captured by the sensor's photodetectors, control A/D and D/A converters, among other functions and with software shipped. This board (5) also has the ability to perform a digital pre-filtering of the converted signals, display the measurement of arterial blood glucose concentration and communicate with a personal computer (6) using its own communication protocol, through one RS-232C type serial port (7). This board (7) works with floating earth in accordance with the requirements of the national and international Technical Standards for electromedical equipment (IEC 60601 and ABNT NBR IEC 60601 Standards series), ensuring the safety of the patient and operator who will use this system, using an isolated source (8).

[0071] The blood glucose sensor (4) used consists of two LEDs and two photodetectors, one set for the wavelength of 805nm and the other for 1350nm. The light emitted by LEDs is partially reflected, transmitted, absorbed and dispersed through the skin, other tissues and the blood before striking the photodetector. The mounting of the sensor (4) must be done in such a way as to protect the photodetectors (3) from the action of ambient light, in the spectral response range of the photodetector (3). There must be a flexible cable that connects the sensor (4) to the proposed instrument (1) and is capable of both providing enough energy to operate the LEDs (2) and also conducting the electrical signal from the photodetectors (3). We chose to use a transmissive sensor to measure the concentration of glucose in the blood. Figure 5 shows a generic transmissive sensor.

[0072] In transmissive sensors, the photodetector (3) has to detect the light transmitted through the tissue. Therefore, it must be placed in line with the LEDs (2) so that the greatest amount of transmitted light is incident on it, thus ensuring the greatest possible detection of light energy. The photodetector (3) should be placed as close to the skin as possible, exerting as little force on the tissue as possible. If the force exerted by the sensor is high, the blood under the tissue can clot due to the external pressure applied. Furthermore, if the distance between the LEDs (2) and the photodetectors (3) is increased - increasing the optical path -, the amount of light detected decreases according to Beer's Law as seen above. Transmissive sensors are usually placed on the patient's fingers (D), ear (O) or nose. In the case of the proposed instrument, a sensor for placement on the patient's finger (D) was illustrated.

[0073] The LEDs (2) are activated alternately, so that only one wavelength passes through the tissue at a time and its transmitted light is detected by the corresponding photodetector (3).

[0074] Figure 6 shows the block diagram of the microcontrolled blood glucose plate used in the proposed System for measurement shown in Figure 2.

[0075] The microcontroller indicated in Figure 6 must internally contain a 12-bit resolution analog-digital converter with a multiplexer for 8 analog inputs and maximum conversion rate of 200,000 samples per second, a direct memory access (DMA) controller for high-speed analog-to-digital conversions, two digital-to-analog converters also with 12-bit resolution, internal reference voltage with a maximum variation of 40 parts per million per degree centigrade, built-in temperature sensor connected to the analog-digital converter, 8 kbytes of programmable flash memory via the serial port for program code storage, 640 internal bytes of flash memory for data storage without the need for extra font for data writing, 256 internal bytes of RAM memory for data storage, possibility of access to 16 Mbytes of external RAM memory, possibility to access 64 kbytes of memory with program code, 3 circuits 16-bit timers/counters, 32 programmable input/output lines, 9 interrupt sources with 2 priority levels, ability to operate at three power consumption levels, one asynchronous serial communication channel, one compatible communication channel with the I2C standard from PHILIPS CORPORATION, a communication channel compatible with the SPI standard from MOTOROLA, Inc, internal "watchdog" circuit and monitor for power supply (in the models tested for implementation of the proposed instrument, the ADμC812 microprocessor from ANALOG was used DEVICES, INC., for being complete for the application in question, which has a core compatible with INTEL CORPORATION's 8051 family).

[0076] The microcontroller is responsible for generating all signals for synchronism of the circuits on the board. The voltage-current conversion circuit indicated in Figure 6, on the other hand, is responsible for the excitation of the LEDs (Figure 7). This circuit is capable of delivering a pulsed current of 120 mA to each LED. The microcontroller determines the current to be supplied to each LED as a function of light absorption by the tissues where the sensor (4) is being used, when it then makes available in one of its internal digital-analog converters a voltage (V) equivalent to desired current. The current must be as high as possible without reaching the saturation of the photodetector (3).

[0077] The microcontroller controls the amount of current that will be supplied to each LED, dynamically adjusting the voltage (VLED) delivered to the voltage-current converter circuit at the non-inverting input of the U9B operational amplifier, and this voltage is changed every time toggles which LED is lit.

[0078] Said microcontroller also controls the LEDV and LEDIV signals, when at low level they activate the 805 nm LED and the 1350 nm LED, respectively. The LEDs are never turned on simultaneously, but there is a time interval in which both are off, allowing the microcontroller to also get the influences of ambient light.

[0079] When the LEDV line is at a low logic level, with the LEDIV line at a high logic level, the microcontroller provides in its digital-analog converter the voltage proportional to the desired current for the 805 nm LED. With the LEDV line at a low logic level, transistor Q6 is off, thus allowing transistor Q4 to turn on and carry the LED current through resistor R10. Still with the LEDV line at low logic level, transistor Q1 turns on, allowing the flow of current from the positive source (+5VA), through resistor R6, towards the 805 nm LED anode (LVA), then turning on the LED of 805 nm. With the LEDIV line at a high logic level, transistor Q3 is off, with no current conduction, and transistor Q5 is on, grounding the base of transistor Q2, keeping it off. In this case the current flow is from the positive source (+5VA), through R6, transistor Q1, 805 nm LED, transistor Q4 and feedback resistor R10. The potential difference at R10 is fed back to the operational amplifier U9B, which compares it with the reference voltage VLED and changes the current conduction of Q4 as necessary.

[0080] When the LEDIV line is at a low logic level, with the LEDV line at a high logic level, the microcontroller provides in its digital-analog converter the voltage proportional to the desired current for the 1350 nm LED. With the LEDIV line at a low logic level, transistor Q5 is off, thus allowing transistor Q2 to turn on and carry the LED current through resistor R10. Still with the LEDIV line at low logic level, transistor Q3 turns on, allowing the flow of current from the positive source (+5VA), through resistor R6, towards the anode of the 1350 nm LED (LVK), then turning on the LED of 1350 nm. With the LEDV line at a high logic level, transistor Q1 is off, with no current conduction, and transistor Q6 is on, grounding the base of transistor Q4, keeping it off. In this case the current flow is from the positive source (+5VA), through R6, transistor Q3, 1350 nm LED, transistor Q2 and feedback resistor R10. The potential difference at R10 is fed back to the operational amplifier U9B, which compares it with the reference voltage VLED and changes the current conduction of Q2 as necessary.

[0081] In case the LEDV and LEDIV lines are both at high logic level, all bipolar transistors are off, with no current conducting through any of the LEDs.

[0082] The sensor detection circuit, which detects whether or not there is a sensor connected, simply checks the saturation of the U9B operational amplifier through the analog-digital conversion of the INOP line, which is saturated when the sensor is not connected.

[0083] In Figure 8 is presented the time diagram of the demultiplexer circuit that appears in Figure 6. Each phase shown in Figure 8 is equivalent to a time interval of approximately 520 μs. The "LED V" signal corresponds to the activation of the 805 nm LED, while the signal "LED IV" corresponds to the activation of the 1350 nm LED. The demultiplexer needs a delay in its activation in relation to the LEDs, due to the response time of the analog circuits that are between the photodetectors (3) and the demultiplexer. Therefore, the demultiplexer trigger signals present a delay in relation to the corresponding LED trigger signal. There is also a signal that is demultiplexed and corresponds to the phase when both LEDs are off. This signal corresponds to the interference of ambient light added to the “offsets” of the analog circuits, and this signal must also be read by the microcontroller in order to eliminate it from the signals corresponding to the LEDs.

[0084] In phase 1, the 805 nm LED is activated, while the 1350 nm LED remains off, the demultiplexer output is disabled and the necessary response time of the analog circuit prior to the demultiplexer is awaited. In phase 2, with the 805 nm LED still on and the 1350 nm LED off, the demultiplexer is enabled and its output corresponding to the sample circuit is released and the 805 nm LED is retained. In phase 3, the two LEDs are turned off, the demultiplexer output is disabled and again the necessary response time of the analog circuit prior to the demultiplexer is waited. In phase 4, keeping the same condition as the LEDs of phase 3, the demultiplexer is enabled and its output corresponding to the sample circuit is released and it retains the interference of the ambient light. In phase 5, the 1350 nm LED is activated, while the 805 nm LED remains off, the demultiplexer output is disabled and the necessary response time of the analog circuit prior to the demultiplexer is awaited. In phase 6, with the 1350 nm LED still on and the 805 nm LED off, the demultiplexer is enabled and its output corresponding to the sample circuit is released and the 1350 nm LED is retained. Phases 7 and 8 are identical to phases 3 and 4, where again the interference of ambient light is sampled. After phase 8, phase 1 is run again.

[0085] As the signal of interest from the photodetectors (3) implemented by two photodiodes is a current, a differential transconductance amplifier should be used, for example, the one shown in Figure 9. The photodetector (3) of the sensor ( 4) it is connected directly to this circuit, its anode being connected to the FDA line and its cathode to the FDK line. An advantage of this configuration is that larger gains are achieved with smaller resistors, which reduces the circuit output “offset”. Another advantage of this circuit is the great common-mode rejection, as the noise will appear as a common-mode signal on both inputs, being canceled at the amplifier output. It is important to emphasize that this configuration does not eliminate the need for an electrostatic shield, but it works well together with it, removing the noise that crosses the imperfections of the shield.

[0086] As the signal from the differential transimpedance amplifier has information about the signals referring to the LEDs and ambient light, the different contents of this signal need to be separated to be individually treated by the filtering circuits (in the tested models, to demultiplex these PHILIPS CORPORATION integrated component CD4051B was used, developed exactly for this purpose, controlled by the microcontroller, as described above). After the signals are demultiplexed, they enter a simple sampling and holding circuit that can be seen in Figure 10. This circuit shows the signal, charging the capacitor in the time interval in which the demultiplexer is with its output directed to it, and maintains the load when the demultiplexer puts the output directed to it in a high impedance state, because both on the demultiplexer side and on the operational amplifier side, in this situation, the circuit presents a very high impedance, in the order of 1012 Q. The timing of the demultiplexer circuit is also shown in Figure 8, and the synchronization is performed with the LEDs excitation circuit.

[0087] The signals already demultiplexed are then filtered to eliminate noise and separate the alternating and continuous components of each signal. In the case of the alternating component, an additional filtering is performed to eliminate the continuous component, as the alternating component represents, in most cases, a variation of 1% to 3% of the total signal received by the photodetector (3) and, if not if this filtering is performed before going through the amplification circuit, the alternating component would occupy a small range of the analog-digital converter, which would cause large inaccuracies in the calculation of R. In the case of the continuous component, an additional filtering is performed to determine eliminate the alternating component of the signal.

[0088] The low-pass filters (PB filters indicated in Figure 6) are common filters of the second order Butterworth type with a cut-off frequency of 5 Hz, with the additional filtering of the signal, for total elimination of the AC component, made by the microcontroller . This filter is applied to 805 nm LED, 1350 nm LED and ambient light signals. The bandpass filters (PF filters indicated in Figure 6) were implemented with a low-pass filter followed by a high-pass filter, both of the second order Butterworth type, with the low-pass filter having a cutoff frequency of 40Hz and the high-pass filter a cutoff frequency of 0.05Hz. This topology was chosen for the bandpass filter due to the difficulty of configuring components using a single circuit.

[0089] After the signals are filtered, they are again multiplexed in order to go through a programmable gain amplifier in which each signal can be amplified with a different gain, and finally reach the input of the microcontroller's analog-digital converter (again it was chosen the integrated component CD4051B from PHILIPS CORPORATION for this circuit, also controlled by the microcontroller, but other similar components already existing or to be developed may be used).

[0090] Figure 11 presents the time diagram of the multiplexer circuit, and in this diagram each phase has a time of approximately 520 μs. Each signal is active at the output of the multiplexer for three phases, and the letter “c” inside each signal represents the phase in which the reading of the analog-digital converter is performed. The delay relative to the appearance of the signal in the converter and the start of the conversion is due to the response time of the multiplexing circuits and the programmable gain. Also in Figure 11, MUX PF V represents the output of the signal relative to the 805 nm LED filtered by its band-pass filter; MUX PB V represents the output of the signal relative to the 805 nm LED filtered by its low-pass filter; MUX PF IV represents the output of the signal relative to the 1350 nm LED filtered by its band-pass filter; MUX PB IV represents the output of the signal relative to the 1350 nm LED filtered by its low-pass filter, and MUX PB AMB represents the output of the signal relative to ambient light filtered by its low-pass filter.

[0091] In phase 1, the multiplexer input relative to the bandpass filter of the 805 nm LED signal is enabled, the programmable gain amplifier is programmed with the value corresponding to this signal and awaits the necessary response time of the analog circuit in front of the multiplexer. In phase 2, the same conditions as in phase 1 are maintained and the microcontroller's analog-digital converter is triggered. In phase 3, the same conditions as in phase 2 are maintained and the analog-digital converter is read, which contains the data related to the alternating component of the 805 nm LED.

[0092] In phase 4, the multiplexer input relative to the low-pass filter of the 805 nm LED signal is enabled, the programmable gain amplifier is programmed with the value corresponding to this signal and awaits the necessary response time of the analog circuit in front of the multiplexer. In phase 5, the same conditions as in phase 4 are maintained and the analog-digital converter of the microcontroller is triggered. In phase 6, the same conditions as in phase 5 are maintained and the analogue-digital converter is read, which contains the data relating to the continuous component of the 805 nm LED.

[0093] In phase 7 the multiplexer input relative to the bandpass filter of the 1350 nm LED signal is enabled, the programmable gain amplifier is programmed with the value corresponding to this signal and awaits the necessary response time of the analog circuit in front of the multiplexer. In phase 8, the same conditions as in phase 7 are maintained and the analog-digital converter of the microcontroller is triggered. In phase 9 the same conditions as in phase 8 are maintained and the analogue-digital converter is read, which contains the data related to the alternating component of the 1350 nm LED.

[0094] In phase 10, the multiplexer input relative to the low-pass filter of the 1350 nm LED signal is enabled, the programmable gain amplifier is programmed with the value corresponding to this signal and await the necessary response time of the analog circuit in front of the multiplexer. In phase 11, the same conditions as in phase 10 are maintained and the analog-digital converter of the microcontroller is triggered. In phase 12, the same conditions as in phase 11 are maintained and the analogue-digital converter is read, which contains the data relating to the continuous component of the 1350 nm LED.

[0095] In phase 13, the multiplexer input relative to the low-pass filter of the ambient light interference signal is enabled, the programmable gain amplifier is programmed with the value corresponding to this signal and await the necessary response time of the analog circuit in front of the multiplexer. In phase 14, the same conditions as in phase 13 are maintained and the analog-digital converter of the microcontroller is triggered. In step 15, the same conditions as in step 14 are maintained and the analogue-digital converter is read, which contains the data relating to the continuous component of ambient light interference. In phase 16, the signal coming from the INOP line is read by the digital analog converter of the microcontroller, which serves to detect the presence or not of the blood glucose sensor.

[0096] For the programmable gain amplifier (14), a circuit with a 12-bit multiplier digital-analog converter with serial input DAC8043 (ANALOG DEVICES, INC) was chosen, as shown in Figure 12. With this circuit it is possible to adjust gains from 1 to 4095, which allows us a wide range of gains with a very simple circuit.

[0097] The SRI_DAC, CLK_DAC and LOAD_DAC control lines are controlled by the microcontroller and are related to the loading of the converter data. The gains are swapped for each signal and these are in phase with the signals shown in Figure 11.

[0098] For the isolation circuits, an already available module was used, called S.FNT.0002-4, which has an isolated source within the prescriptions of the NBR IEC 60601 Technical Standards series. Supplying the module with +5V, it supplies two +5V sources (one for the analog circuits and one for the digital circuits) and another -5V source for the analog circuits, all already isolated. Within this module, the isolation of serial communication signals is also carried out, using opto-couplers (in the built prototype, the H11N1-M from FAIRCHILD SEMICONDUCTOR CORPORATION was used). At the output of the serial board, and using the same +5V source used to power the S.FNT.0002-4 module, a circuit was included to convert the voltage levels of the serial communication output to the RS-232C standard (it was used the MAX232C integrated circuit from MAXIM INTEGRATED PRODUCTS, INC, which can be replaced by one with similar characteristics).

[0099] Evidently, it is possible to conceive several Measurement Systems similar to the one described here, of the same basic conception, but differing in aspects related to shapes, dimensions, materials and other criteria of secondary importance, without thereby departing from the scope required protection. The same happens with the circuits and components, which were presented as an example, among many possible ones, of implementation and realization of the instrument. Such sets, however, will remain inextricably linked to the spirit and scope of this patent.

[0100] Thus, as a corollary of the above and illustrated, it is concluded that the invention presents logistical and functional characteristics completely unprecedented and, therefore, is provided with the necessary requirements to obtain the patent privilege of invention claimed here.

Claims (8)

1) "NON-invasive MEASUREMENT INSTRUMENT FOR CONTINUOUS MONITORING OF GLUCOSE CONCENTRATION IN ARTERIAL BLOOD" comprised of an instrument (1) consisting of a light-emitting LED (2), photodetector (3), non-invasive optical sensor (4) and electronic circuit (5), characterized by the fact that it uses the pulsatile characteristic of blood to distinguish the absorbance of arterial blood from other absorbers, using two different wavelengths, being waves of 805nm and between 1300 and 2400nm, to perform the measurement. 2) "NON-invasive MEASUREMENT INSTRUMENT FOR CONTINUOUS MONITORING OF GLUCOSE CONCENTRATION IN ARTERIAL BLOOD" according to claim 1, characterized in that it performs the measurement of blood glucose exclusively in arterial blood. 3) "NON-invasive MEASUREMENT INSTRUMENT FOR CONTINUOUS MONITORING OF GLUCOSE CONCENTRATION IN ARTERIAL BLOOD" according to claim 1, characterized in that it uses the measurement of absorbance or transmittance or reflectance of light in living tissues. 4) "NON-invasive MEASUREMENT INSTRUMENT FOR CONTINUOUS MONITORING OF GLUCOSE CONCENTRATION IN ARTERIAL BLOOD" according to claim 1, characterized in that it can be used in humans or animals. 5) "NON-INVASIVE MEASURING INSTRUMENT FOR CONTINUOUS MONITORING OF GLUCOSE CONCENTRATION IN ARTERIAL BLOOD" according to claim 1, characterized in that it can be constructed in the form of a ring, earring, bracelet or any other format that allows its use personal and viewable data. 6) "NON-invasive MEASUREMENT INSTRUMENT FOR CONTINUOUS MONITORING OF GLUCOSE CONCENTRATION IN ARTERIAL BLOOD" according to claim 1, characterized in that it shows the measurement value in the instrument itself. 7) "NON-invasive MEASUREMENT INSTRUMENT FOR CONTINUOUS MONITORING OF GLUCOSE CONCENTRATION IN ARTERIAL BLOOD" according to claim 1, characterized in that it has a data transmission system by cables, optical fibers, radio transmission waves or any other suitable means, to a receiving unit equipped with a specific program for the dissemination, manipulation and/or control of information. 8) "NON-INVASIVE MEASURING INSTRUMENT FOR CONTINUOUS MONITORING OF GLUCOSE CONCENTRATION IN ARTERIAL BLOOD" according to claim 1, characterized in that it is portable and enables self-monitoring of blood glucose in a non-invasive and continuous way.

Images