EP4626305A1

EP4626305A1 - Non-invasive apparatus, system and methods for monitoring blood flow and coagulation

Info

Publication number: EP4626305A1
Application number: EP23898914.9A
Authority: EP
Inventors: Erika A. DILLARD; Tamara C. Baynham
Original assignee: Popcheck Technologies Inc
Current assignee: Popcheck Technologies Inc
Priority date: 2022-11-30
Filing date: 2023-11-30
Publication date: 2025-10-08
Also published as: KR20250114123A; WO2024118949A1; CN120603532A; AU2023406122A1; WO2024118949A4

Abstract

Wearable devices including an external patch adhered to the skin or a sleeve are used to stimulate and recognize changes in blood flow and coagulation. Non-invasive, transcutaneous electrical muscle stimulators (EMSs) are embedded in the patch or sleeve and provide sequential stimulation to the perivascular tissue to promote blood flow. The stimulation system includes a series of electrodes positioned on the skin via the patch or sleeve and an external programmable generator having a wireless connection during stimulation. Biosensors in the patch or sleeve and other biosensors applied to the body periodically check for abnormal biomarker patterns. These patterns may be used by artificial intelligence (AI)/machine learning (ML) systems as an early indicator or by prediction methods to predict changes in blood flow and coagulation as caused by venous clots. When abnormal biomarker patterns are detected, an alert may be sent and therapeutic response initiated.

Description

NON-INVASIVE APPARATUS, SYSTEM AND METHODS FOR MONITORING BLOOD FLOW AND COAGULATION

TECHNICAL FIELD

[0001] The present disclosure relates generally to apparatus, systems and methods for monitoring blood flow and coagulation and, more specifically, to devices and methods for surveillance of blood clots and related conditions by predicting the probability of developing a blood clot in the venous system from non-invasive biosensor measurements.

BACKGROUND

[0002] A thrombotic event occurs when a blood clot (thrombus) blocks the flow of blood within the circulatory or vascular system. Depending on the vessel obstructed, this blockage can decrease the supply of oxygen and vital nutrients carried in the blood to tissues in the body resulting in cell damage and potentially cell death if blood flow is not restored quickly. This blockage can also prevent the outflow of deoxygenated blood and metabolic waste products away from tissues. Obstructing the drainage of blood and waste from tissues can increase pressure in the blood vessel causing leakage of fluid in the surrounding perivascular space resulting in pain and swelling. Veins are the blood vessels that carry blood from tissues back to the heart. Therefore, blood clots that form in and travel through the venous system are referred to as venous thromboembolism (VTE). VTE that form within larger veins of the body are referred to as deep vein thrombosis (DVT) and can develop in the arms, legs, pelvis, torso and even brain (cerebral venous sinus thrombosis). Additionally, clots can break away from DVTs and travel or embolize to other parts of the body causing obstruction of blood flow in other locations. Clots that embolize and settle in the lungs are called pulmonary embolism (PE) and can be deadly.

[0003] Common features of DVT include pain, discoloration such as redness or bruising of the skin, swelling or prominence of the superficial veins. Clues of PE might include shortness of breath or rapid breathing, palpitations, dizziness, sweating or sharp chest and rib pain that worsens with inspiration. Unfortunately, many of the signs and symptoms for VTE are non-specific and can be misinterpreted as being caused by less urgent etiologies like a muscle strain as with a lower extremity DVT or anxiety for PE. This can lead to worse clinical outcomes due to delays in diagnosis.

[0004] Collectively, 1 million people per year suffer from VTE contributing to as many as 300,000 deaths in the United States alone. These values are likely underestimated as annual surveillance for VTE in the United States is not performed. Despite its prevalence and mortality, VTE is considered a preventable disease in the majority of cases. In fact, it is the leading cause of preventable death in hospitalized patients in the United States and worldwide. For those who experience VTE, long term complications and recurrence are a concern. One in three will develop another clot which often requires life-long anticoagulant therapy. Another third will develop post- thrombotic syndrome, a major source of impairment and decline in quality of life following the initial DVT diagnosis.

[0005] The pathogenesis of thrombosis has been well described by Virchow’s triad, a principle delineating three major factors that, when present, create an environment conducive for thrombosis: venous stasis, endothelial injury of the blood vessel, and hypercoagulability. Risk factors for VTE are traditionally classified by whether they are identifiable (provoked) or not (unprovoked). Risk factors leading to provoked VTE can be inherited or acquired. Genetic factors include inherited deficiencies of antithrombin, protein C, and protein S. Acquired factors include medical conditions such as infection and cancer, immobility, surgeries such as joint replacements and particularly those in response to a trauma or cancer and medications such as hormone replacement therapies. Hospitalization has a distinct association with the occurrence of VTE and are referred to as healthcare-associated or HA- VTE when they develop during or soon following hospitalization. However classified, these risk factors can incite any or all elements of Virchow's triad. Recognizing these associations can improve management and foster the development of novel tools from VTE prevention to treatment.

[0006] Conventionally, persons suspected of having VTE are assessed using pretest probability (PTP) scoring systems such as the Wells DVT and PE score and Geneva score. These scoring batteries take into account specific criteria including demographics, pre-existing conditions and active symptoms like calf pain and swelling as in the case of suspected DVT. These traditional scoring systems do not account for social determinants of health (SDoH) which are defined by the World Health Organization (WHO) as the “circumstances in which people are born, grow, live, work, and age and the systems put in place to deal with illness.” SDoH have been directly tied to risk and outcomes of cardiovascular disease including VTE. Without inclusion of SDoH in risk assessment and outcome predictions, the application of predictive models utilizing traditional data sets and measures will not accurately reflect true risk and outcome probabilities and even further will continue to deepen healthcare disparities and intensify disease burden experienced by marginalized and underserved populations. In effect, results of PTP cannot safely rule VTE in or out.

[0007] When scoring from traditional PTP systems results in a low likelihood for VTE, a serum D-dimer test is performed to help exclude a VTE diagnosis. Alternatively, if there is a high likelihood for VTE based on scoring, a diagnosis can be confirmed with relevant imaging studies such as an Ultrasound or CT angiogram. Currently, specific standards for imaging studies to confirm or exclude a VTE diagnosis do not exist beyond clinical judgement. Such inconsistencies in management provide another opportunity for delays in care.

[0008] VTE treatment classically involves a regimen of anticoagulants (ACs) or blood thinning agents. Many ACs, such as warfarin, require frequent testing to ensure they are within the therapeutic window which can often take weeks to achieve. Other blood thinning agents, like direct oral anticoagulants (DOACs) do not typically require serum studies to ensure they are within the therapeutic window; however, variations in patient response do exist. Consequently, inappropriate dosing may not be detected, resulting in worse outcomes from bleeding or, conversely, thrombotic events when dosing is subtherapeutic. In addition, physical therapy, when implemented along with pharmacotherapy, has become an important adjunct to the treatment regimen and for prevention of long term VTE related complications.

[0009] With the complexities and variability in management and access to technologies from prevention to treatment, techniques that can safely and reliably assist with management in diverse settings along the spectrum of care are warranted.

SUMMARY

[0010] Various examples are now described to introduce a selection of concepts in a simplified form that are further described below in the Detailed Description. This Summary is not intended to be used to limit the scope of the claimed subject matter.

[0011] In sample configurations, a patch is adhered to the back of the leg or other anatomical position or a sleeve is applied where venous thromboembolism (VTE) may develop. Non- invasive, transcutaneous nerve electrical muscle stimulators (EMSs) are embedded in the patch or sleeve that is applied to a patient during conditions when changes in blood flow and coagulation arc expected, e.g., post-operatively, to prevent blood clots through external stimulation of the calf muscles or other local muscles. The stimulation system includes a series of electrodes positioned on the skin of the patient via the patch or sleeve and an external programmable stimulator having a wireless connection during stimulation and that is battery powered and rechargeable to facilitate mobility. Biosensors in the patch or sleeve as well as other biosensors applied to the body are periodically checked for abnormal biomarker patterns that may be used by artificial intelligence (AI)/machine learning (ML) systems to predict blood clot formation. When such abnormal biomarker patterns are detected, the healthcare professional is notified that further management of potential VTE in the patient is required.

[0012] In sample configurations, at least one of a computer application, a remote computer, or a cloud server may process quantitative biosensor data collected by the biosensors and collected qualitative data determinative of a patient’s health in a predictive algorithm that uses machine learning techniques to predict the probability of development of a blood clot in the patient and, based on the prediction of probability of development of a blood clot generated by the predictive algorithm, calculate a risk score. The computer application, the remote computer, and/or the cloud server may generate an alert representing the status of the patient and/or initiate a therapeutic response to the development of the blood clot based on a predetermined change in the risk score. The alert may be issued using different modalities to initiate different actions in accordance with whether the recipient of the alert is a healthcare professional, the patient, or a caregiver. The different modalities may include at least one of a voice or audio message, e-mail, SMS text utilizing cellular data, a chat application, an EMR alert, a telemedicine system, or an artificial intelligence generated message through a software application. In addition, the system may include a device gateway that receives and transmits data from the biosensors and third party data sources to at least one of the computer application, the remote computer, or the cloud server.

[0013] A method of monitoring blood flow and coagulation to predict development of blood clots in the body of a patient is also provided. The method includes positioning on a region of interest on the surface of the body of a patient a device configured to measure physiological biomarkers, create localized blood flow, and transfer data to a remote computer server; creating localized blood flow using the device; measuring biomarkers indicative of changes in blood flow and coagulation; and implementing a predictive algorithm on the computer or cloud server that uses machine learning techniques to predict from at least the biomarker measurements the probability of development of a blood clot in the patient. Based on the prediction of probability of development of a blood clot generated by the predictive algorithm, the method further includes calculating a risk score and at least one of generating an alert representing the status of the patient or initiating a therapeutic response to the development of the blood clot.

[0014]The method may further include transferring data from the device in an out-patient setting or a communication device in an in-patient setting to a computer or a cloud server, wherein the predictive algorithm is implemented on the computer or the cloud server. Also, measuring biomarkers indicative of changes in blood flow and coagulation may further include initiating settings of biomarker measurement devices using an initial patient biomarker profile based on pre-assessed patient specific information highlighting an existing disease state or physiological conditions that affect or are affected by blood flow and coagulation and may alter risk for development of a blood clot. For example, the pre-assessed patient specific information may be based on a particular diagnosis, procedure, or blood flow and coagulation altering therapy prescribed to the patient and/or may include socioeconomic factors or social determinants of health that influence incidence of blood clots or contribute to worse outcomes in patients diagnosed with a blood clot.

[0015] This summary section is provided to introduce aspects of the inventive subject matter in a simplified form, with further explanation of the inventive subject matter following in the text of the detailed description. The particular' combination and order of elements listed in this summary section is not intended to provide limitation to the elements of the claimed subject matter. Rather, it will be understood that this section provides summarized examples of some of the embodiments described in the Detailed Description below.

BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The foregoing and other beneficial features and advantages of the invention will become apparent from the following detailed description in connection with the attached figures, of which:

[0017] FIG. 1A is a view of the back view of a patient’s leg. showing the positioning of the patient’s popliteal vein.

[0018] FIG. IB is an exploded view of a system comprised of a patch and a stimulation device adapted to acquire data from biosensors placed on the skin of a patient in a sample configuration.

|0019| FIGS. 2A-2C are diagrams of additional configurations of the system illustrated in Figure 1 with application to other anatomical placements using a sleeve for placement on a patient’ s arm (FIG. 2 A) or a patient’s leg (FIGS. 2B and 2C).

[0020] FIG. 3 is a diagram of the electronics used to acquire patient data from biosensors and tr ansfer patient data to a computer application in a sample configuration.

[0021] FIG. 4 is a diagram illustrating examples of patient journey use cases in which the system and methods herein described may be applied.

[0022] FIG. 5 is a diagram representing the transfer of the acquired patient data to the cloud for application of a predictive model in a sample configuration.

[0023] FIG. 6 is a flow chart depicting a method for predicting the likelihood of a patient developing a blood clot in the blood vessel of interest in a sample configuration.

[0024] FIG. 7 is a diagram representing the plurality of data sources within the system including equipment that can transmit data via a device gateway and 3^rd party sources such as public databases with bi-directional transfer of data to the cloud in a sample configuration.

[0025] FIG. 8 is a detailed flow chart depicting the method for prediction of a patient’s risk of developing a blood clot and either repeating measurements at timed intervals if a risk score is unchanged or is not concerning or sending an alert to a respondent such as a healthcare professional when a critical threshold or pattern is detected in order to initiate next steps in patient management in a sample configuration.

[0026] FIG. 9 is a hierarchical depiction of the alert process as it relates to respondents and potential actions that might be taken in patient management by individuals or by the system itself in a sample configuration.

DETAILED DESCRIPTION

[0027] A detailed description of illustrative embodiments will now be described with reference to FIGS. 1-9. Although this description provides a detailed description of possible implementations, it should be noted that these details are intended to be exemplary and in no way delimit the scope of the inventive subject matter.

[0028] FIG. 1A is a view of the back view of a patient’s leg, showing the positioning of the patient’s popliteal vein. FIG. IB is an exploded view of a system 100 comprised of a multi-layered adhesive patch 110 and a detachable electronic/stimulation device 120 adapted to acquire data from biosensors 130 placed on the body of a patient in a sample configuration. As illustrated, the multilayered adhesive patch 110 includes at minimum an outer layer 140 having a flexible fabric covering and an inner layer 150 having electronic connections between a plurality of biosensors 130 for sensing physiological signals from the skin of a patient as well as electrodes 160 for sensing bioimpedance and delivering electrical stimulation. The electronic connections include a connection to the detachable electronic/stimulation device 120 that houses the electronics used to acquire patient data recorded with the plurality of biosensors 130 and electrodes 160 and to transfer the acquired data to the cloud and/or a computer server to analyze the patient data and implement a machine learning algorithm configured to predict the probability of developing a blood clot in a blood vessel of a patient based on the acquired data. It should be understood that the electronic/stimulation device 120 may be integrated within the patch. In another configuration, the electronic/stimulation device 120 may be integrated and powered externally using radio-frequency (RF), by other wireless power means, or corded (plug-in power).

[0029] The system in FIG. IB is configured to align the biosensors 130 and electrodes 160 along a blood vessel of interest. For example, the multi-layered adhesive patch 110 may be placed along the popliteal vein in the lower extremity of a patient (as seen from the back view in FIG. 1A) for predicting the probability of developing a blood clot in a sample configuration. It should be understood that the system may be realized in a variety of configurations.

[0030] For example, configurations of the system 100 shown in FIGS. 2A-2C may include an arm configuration 200 (FIG. 2 A) and leg configur ations 210 (FIG. 2B-2C) that incorporate a multi-layered sleeve 220 that would be placed along the blood vessel of interest (e.g., popliteal vein of the leg). The multi-layered sleeve 220 includes at minimum an outer layer having a flexible fabric covering and an inner layer having electronic connections between a plurality of biosensors 130 for sensing physiological signals from the skin of a patient as well as electrodes 160 for sensing bioimpedance and delivering electrical stimulation. The multi-layered sleeve 220 may be configured to accommodate different sizes. The multi-layered sleeve 220 may be closed with Velcro or other means around the limb. The electronic connections include a connection to the electronic/stimulation device 120 that houses the electronics used to acquire patient data recorded with the plurality of biosensors 130 and electrodes 160 and to transfer the acquired data to the cloud and/or a computer server to analyze the patient data and implement a machine learning algorithm configured to predict the probability of developing a blood clot in a blood vessel of a patient based on the acquired data.

|0031] The example configuration of the system 100 illustrated in FIG. 2C further incorporates a mercury strain gauge 240 for plethysmography and includes one or more pressure cuffs 230 and 250 spaced apart along a vascular network to apply occlusive pressure to target blood vessel(s) with or without a multi-layered sleeve 220. It is also understood that a targeted occlusion of a blood vessel may be accomplished without the use of a cuff. For example, it is possible to provide targeted pressure by other means such as smart compression garments composed of shape memory alloy or electromechanical percussive massage. The components described herein would be connected to the device 120 that houses the electronics used to acquire patient data measured with the mercury strain gauge 240 and to transfer the acquired data to the cloud and/or a computer server to analyze the patient data and implement a machine lear ning algorithm configured to predict the probability of developing a blood clot in a blood vessel of a patient based on the acquired data.

[0032] FIG. 3 is a diagram of the electronics disposed in detachable electronic/stimulation device 120 and used to acquire patient data from the plurality of biosensors 130 and electrodes 160 and to transfer the acquired patient data to an application on a local computing device or to the cloud and/or to a remote computer server in a sample configuration. It is to be understood that the electronic/stimulation device 120 may be integrated into the system 100 or have other configurations. As illustrated, a plurality of biosensors 130 and bioimpedance electrodes 160 are connected to electronics of a microcontroller 300 powered by battery 315 and including signal conditioning circuit(s) 320, impedance measurement circuit 330, data acquisition circuits 340, and a timing control circuit 350. The biosensors 130 include, but are not limited to, photoplethysmography (PPG), infrared thermopile, electrodermal activity (EDA) sensor, galvanic skin response (GSR) sensor, and/or strain gauge plethysmography. The electrodes 160 may be used to acquire the voltage difference produced by the current injected by current injection electrodes 360. The current is generated by a signal generator circuit 370. The current delivered may be configured to provide constant current stimulation. The constant current stimulation may be configured to reduce or eliminate perception by the patient by delivering a frequency between 20 kHz and 100 kHz. The constant current may be configured to have an amplitude greater than 1 mA to provide sufficient signal-to-noise ratio to detect changes in impedance.

[0033] In sample configurations, stimulation electrodes 310 may be used to excite muscle fibers or nerves to elicit blood flow in response to signals generated by the signal generator circuit 370. This blood flow may be used to facilitate acquisition of data from the physiological sensors 130. Additionally, the blood flow can be used to help in the prevention of blood clotting. It is understood that other means to create blood flow may be incorporated that may include and not be limited to pneumatic compression, ultrasound, massage (or local compression), vibration, and passive and/or voluntary movement.

[0034] The microcontroller 300 includes a timing control circuit 350 that controls the timing of signal generator 370, data acquisition circuits 340, signal conditioning circuits 320, as well as the impedance measurement circuit 330. The impedance measurement will be timed based on the injection of current produced by the signal generator 370 into the current injection electrodes 360. The data acquisition circuit 340 measures the voltage at the impedance measurement electrodes 160 and transfers the acquired voltage to the memory 380. Data acquisition from the physiological sensors 130 is also controlled by the timing control circuit 350. Physiological sensor data from the physiological sensors 130 is input into the appropriate signal conditioning circuits 320. The signal conditioning circuits 320 may include but are not limited to filtering, strain gauge signal conditioning (isolation, bridge balance, filtering, excitation voltage), thermopile signal conditioning (isolation, linearization, etc.), as well as other signal conditioning as may be needed to facilitate acquisition of physiological sensor data from the physiological sensors 130. Once the physiological sensor data has been acquired and conditioned via the applicable signal conditioning circuits 320, the physiological sensor data is saved in memory 380.

[0035] A communication circuit 390 facilitates the transfer of data from memory 380 to a computer application 500 implemented on one or more smart devices such as a phone or tablet and adapted to receive and process the acquired physiological data and voltage data. The wireless communication circuit 390 may facilitate communication using BLUETOOTH®, WI-FI®, Zigbee®, cellular data, or radio frequency (RF) circuitry to the local smart phone or other electronic device that processes the computer application 500. The acquired patient data then may be transferred from the computer application 500 to the cloud or another server via wired or wireless Internet connectivity (FIG. 5). Conversely, the communications circuit 390 may be adapted to transfer the acquired physiological data and voltage data directly from the detachable electronic/stimulation device 120 to a remote server device for processing.

[0036] As illustrated in FIG. 4, different predetermined settings expressed in patient journey use cases 400 may be selected for the system. The predetermined settings may be based on preassessed patient specific information highlighting existing disease states or physiological conditions that affect or are affected by blood flow and coagulation and therefore may alter the risk for development of clots. This would include but is not limited to a particular diagnosis, procedure, or blood flow and coagulation altering therapies. Examples include surgery use case 410 such as hip replacements, cancer, infections, inflammatory diseases, sepsis, vascular insufficiencies, dehydration, immobilization, pregnancy, etc. Other pre-determined settings for the system and methods described might involve monitoring pharmacological therapy use cases 420 such as anticoagulants, antiplatelets, or other blood thinning drugs, and non-pharmacological therapy use cases 430 such as exercise regimens or use of preventative mechanical devices such as pneumatic compressors, and the like, or surgical therapies such as revascularization procedures (stent, bypass) or monitoring the patency of conduits for infusions or hemodialysis. Additional pre-determined settings for the system and methods described might include socioeconomic factors or social determinants of health that influence incidence of blood clots and/or outcomes in patients diagnosed with blood clots such as social support, education, health literacy, food insecurity, access to and payment method available for medical care, residential environment, etc. Social determinants of health may have quantitative measures such as Social Vulnerability Index (SVI) or Area Deprivation Index (ADI). Qualitative data may include information from the patients such as how the patient might describe their relationship with their medical provider (communication, bias, etc.) or congruency in patient expectations in outcomes for a particular therapy (e.g., blood thinners) with the medical provider’s expectations. In each case, initial patient biomarker profiles are obtained and serve as baselines for comparative analysis. It is to be understood that the system 100 may be used in conjunction with preventative devices. For example, the system 100 may be positioned under a pneumatic compressor and operated along with the operation of the pneumatic compressor.

[0037] FIG. 5 is a diagram representing the transfer of the acquired patient data to the cloud 510 for application of a predictive model 520 to assess the probability of developing a blood clot in a sample configuration. The cloud 510 includes circuitry for ingesting the patient data (530), analyzing the patient data (540), storing the analyzed patient data (550), and reporting the outcome of the data analysis to a healthcare professional (560). The system 100 of FIG. 1 may be used to assess the patient risk of developing a blood clot in either an outpatient (ambulatory care, assisted living facilities, single family home, etc.) (570) or inpatient (acute care hospitals, rehab centers, nursing homes, etc.) setting (580).

[0038] In the outpatient setting 570, for example, the patient 585 may be home and have the system 100 of FIG. 1 positioned on the body in proximity to the blood vessel of interest. Data from the system 100 is transferred to the application 500, which may operate on any mobile platform including Android, iOS, windows, etc. and be configured on a single or a plurality of displays including a smart watch, mobile phone, tablet, computer, television monitor, etc. that has the capability to communicate via BLUETOOTH®, WI-FI®, Zigbee®, cellular data, or other wireless means with the system 100. Additionally, the computer application 500 may be configured to transfer the acquired data to the cloud 510.

[0039] In the in-patient setting 580, the patient 585 may be in the hospital and have the system 100 positioned on the body in proximity to the blood vessel of interest. The system 100 may transfer the acquired patient data directly from the system 100 to the cloud 510 or there may be another communication device in proximity to the patient 585 used to transfer data from the system 100 to the cloud 510.

[0040] Once the acquired patient data is transferred to a cloud server of cloud 510, the data is ingested at 530, analyzed at 540, and the predictive model 520 is implemented. Based on the prediction by the predictive model 520, a risk score is calculated and reported to the healthcare professional 590 by reporting software 560 via the healthcare professional interface 595. The healthcare professional interface 595 may be displayed on a mobile phone, tablet, computer, patient status board, in-room monitor, etc. connected or otherwise to an electronic medical record (EMR) or health information systems (HIS) server. All data in the cloud 510, including the analysis data (540) and model data (520), is stored in data storage 550 and used to improve the predictive model 520 over time as more patient data is transferred to the cloud 510.

[0041] FIG. 6 is a flow chart depicting a method 600 for predicting the likelihood of a patient developing a blood clot in a sample configuration. As illustrated, the method 600 starts by reading in data from one or more datasets 610 including data from physiological sensors 130, electrodes 160, and other patient data (including demographic information, medical history, past or ongoing therapeutic interventions, serum biomarker values (e.g., D-dimer, PT/PTT, INR, etc.), pain scale and other patient reported symptoms, supplemental biometric data, etc. and social determinants of health. The datasets 610 may also include risk assessment and scoring batteries including related variations (e.g., Caprini Score, Wells Criteria, PERC, etc.). Preprocessing of the data read from the dataset 610 is performed at 620 including any signal conditioning and analysis required.

[0042] The patient input also may include subjective information such as a pain indication (Y/N and if yes, so scale 1-10), swelling (Y/N), discoloration, i.e., redness and/or bruising (Y/N), warmth (Y/N), Chest pain (Y/N), cough (Y/N), shortness of breath (Y/N), etc. Additional data that can be entered for analysis but is collected prior to and not during the monitoring period may include, for example, patient demographics (sex, race, age, etc.) and patient medical history (pre-existing conditions, past surgeries, medications, etc.).

[0043] The processed dataset may be split into training dataset 630, validation dataset 640, and test dataset 650. The training dataset 630 may be used to train a classifier 632 such as a classification machine learning algorithm of a machine learning device. Classification algorithms implemented by the classifier 632 may include logistic regression, k-nearest neighbors, decision tree, random forest, and/or support vector machines that classifies the input data to correspond to VTE conditions or not. The resulting validation dataset 640 may be used to evaluate the prediction accuracy 642 of the classification 644 on the “trained” model 520. The trained model 520 is tweaked at 660 according to the results of the prediction accuracy 642 based on the validation dataset 640. This iterative process may be repeated.

[0044] The best model 670 based on the validation dataset 640 is selected and the results are then confirmed on the test dataset 650 at 672. The resulting classification 674 may be used to calculate a risk score 676 corresponding to the probability that a patient may develop a blood clot in the blood vessel of interest.

100451 Based on the calculated risk score 676, the system output reporting will vary depending on the recipient (e.g., patient 585 versus healthcare professional 590) and whether or not the risk score 676 corresponding to biomarker patterns has changed. If the risk score 676 is unchanged, the patient 585 and healthcare professional 590 will receive similar messages through the patient interface 500 and the healthcare professional interface 595, respectively, for confirmation of a successful biomarker check and subsequent data storage. On the other hand, if the risk score 676 is elevated a predetermined amount in a manner corresponding to measurable abnormalities in biomarker patterns, then an alert will prompt the patient 685 through the patient interface 500 to complete a device check to ensure that the leads and sensors are correctly positioned. If the device check results in a concern for device function or positioning, troubleshooting instructions will be provided to the patient 585. If there is no device issue, additional information may be requested from the patient via the patient interface 500 concerning current signs and symptoms. A device alert may also be sent for completion of data analysis of data sent and stored, and a change in risk status may be reported. Simultaneously, the healthcare professional 590 may receive through the healthcare professional interface 595 a device alert sent for change in risk status including providing access for the healthcare provider to stored patient data 550.

[0046] FIG. 7 is a diagram representing the plurality of data sources within the system 700 including wearables or contactless sensors 710 including wearable device 100 (FIG. 1A) that can transmit data 610 via a device gateway 720 and 3^rd party sources such as public databases 790 that provide bi-directional transfer of data to the cloud 510. In turn, the cloud 510 communicates with a plurality of data receiving devices including a patient status board 730, an in-room monitor 740, an EMR/HIS server 750, a display such as a smart television or computer 755 or a tablet 760, a smart watch 765 or cellular device 770, and the like that are capable of displaying and communicating raw and processed data in various forms. Datasets 610 can be acquired from one or multiple sources at various time points along one or more patient care journeys and over varying patient conditions such as before and after application of a therapeutic regimen. Information from one or more datasets can be acquired prior to device use and can serve as patient specific baselines. Patient specific values and/or those from population specific datasets can be used as baselines to guide device monitoring and detection of abnormal biomarker patterns.

[0047] Datasets can contain both quantitative (temperature, heart rate, social vulnerability index) and qualitative (patient care perceptions and expectations) data via patient surveys 780 as well as both subjective (patient reported symptoms) and objective (observed signs) information. Data beyond biomarker patterns related to VTE development can be acquired, stored and analyzed during the application of the system to provide insight into related outcome measures such as length of stay during an inpatient admission, readmission rate, mortality, development of long-term complications, exacerbation of pre-existing medical conditions or development of new medical conditions, medication adherence, etc. These outcome measures may contribute to the development of new protocols or strategies to improve patient safety, quality of patient care, costs of care and utilization of healthcare resources.

[0048] FIG. 8 is a detailed flow chart depicting the method 800 for prediction of a patient’s risk of developing a blood clot and initiating an alert for further management which may include but is not limited to a healthcare professional ordering additional diagnostic studies or initiating treatment to prevent blood clot formation in a sample configuration. In the example of FIG. 8, the system 100 has been placed on the limb of a patient with the sensors 130 positioned in proximity to the blood vessel of interest. Localized blood flow may be created at 810 by electrical stimulation of the muscle and/or nerve by the stimulation electrodes 310 and/or current injection electrodes 360 as depicted in FIG. 3. However, it is understood that blood flow may be created using other methods such as pneumatic compression, ultrasound, massage (localized compression), vibration, and/or passive or voluntary movement. In another embodiment, a risk increase might initiate a response to prompt the system to automatically initiate a therapeutic regimen delivered by the apparatus 100 or through communication with other therapeutic devices within the system. [0049] Changes in physiological biomarkers may be measured during the localized blood flow at 820. For example, physiological, non-invasive, biosensors 130 as shown in FIG. 1 may measure flow gradient, venous compliance, valve function, temperature gradient, oxygen gradient, and the like. In another configuration, the biosensors 130 may also measure molecular biomarkers such D- dimer test results, fibrin degradation products (FDPs) or other biomarkers of fibrinolysis, von Willebrand factor (vWF), P-selectin protein, intercellular adhesion molecule (ICAM-1), thrombomodulin (THBD) protein, endothelial protein receptor (EPCR), tissue factor pathway inhibitor (TFPI), forkhead box protein C2 (FOXC2), prospero homeobox protein 1 (PROXI), and the like. In another alternative configuration, previously collected serum biomarkers such as these may be provided as an additional dataset 610. The biomarkers measured by the biosensors 130 are analyzed to identify any patterns at 830. Means of measurement of the biomarkers may include electrical impedance measurements, bio-capacitance, thermal imaging, ultrasound imaging, auscultation, photoplethysmography, strain gauge plethysmography, and the like.

[0050] Impedance change that could be indicative of blood clot formation may be between 1 -10% change in impedance, with 1-4% being the optimal range. On the other hand, an increase in local skin temperature may be indicative of a pathological process, such as blood clot formation. For example, a temperature change of greater than 0.2°C with respect to the contralateral limb or the limb outside the measurement area with a range of 0.4°C to 2.5°C may be measured (see, e.g., Shaydakov, et al., “Effectiveness of infrared thermography in the diagnosis of deep vein thrombosis: an evidencebased review,” J Vase Diag Interven, 2017, Vol 5, pp. 7-14).

[0051] Once biomarkers are measured at 820 and the acquired data transferred to the cloud 510, the data is analyzed at 830 along with any additional datasets 610 acquired from other sources and the predictive model is implemented at 840. A risk score is calculated at 850 based on the output of the predictive model 520 (FIG. 5). If no critical threshold or abnormal pattern corresponding with an elevated risk is detected at 860, biomarker measurements may be repeated at timed intervals 880 upon initiation of localized blood flow at 810. For example, the timed intervals may include a range of 5 minutes to one hour. On the other hand, if the calculated risk score is indicative of blood clot formation, an alert at 870 is sent to a respondent such as a healthcare professional to initiate next steps in management.

[0052] FIG. 9 describes the alert process 900 by which the alert 870 may be received by various modalities 910 including as a voice or audio message, by e-mail, by SMS utilizing cellular data, by a chat application, as an EMR alert, via a telemedicine system, and as an Al generated message through a software platform or application. Respondents 920 of the alert can be an individual such as a healthcare professional, patient and/or their caregiver or the device within the system. [0053] In response to the alert, a healthcare professional 590 might perform one or more actions 930 including ordering diagnostic studies such as ultrasound imaging, initiate therapy, schedule an in-person or virtual visit, consult additional professionals to the care team such as a vascular- medicine specialist or a social worker to address identified socioeconomic factors, or send the patient directly to the emergency room for further work up or admission. Patient or caregiver 585 prompted actions 940 might include seeking emergency care, calling their healthcare professional, emphasizing or de-emphasizing existing strategies such as physical therapy exercises or making no changes and continuing current regimen. In addition, the device 100 might respond with actions 950 including connecting the other respondents to a 3^rd party such as an insurance company, community organization to connect the patient with community-based resources, or initiate an emergency response if patient’s risk is concerning for poor outcomes. The device 100 might also automatically provide therapy by eliciting a programmed electrical stimulation regimen or communicate with other connected devices within the system that are programmed to deliver therapy such as a drug injection or infusion device.

[0054] Thus, it will be appreciated that the methods described herein implement a method of predicting a blood clot in the body of a patient by implementing steps including: positioning on a region of interest on the surface of the body of a patient a device configured to measure physiological biomarkers, create localized blood flow, and transfer data to a remote computer server; creating localized blood flow using the device; measuring biomarkers indicative of changes in blood flow and coagulation, which may be influenced by predetermined settings reflecting patient journey use cases based on pre-assessed patient specific data; combining measured biomarker data with data from other connected devices including third party databases and servers and survey mediated patient reported data; transferring data from the biomarker measurements to a computer; transferring data from the computer application in an out-patient setting or a communication device in an in-patient setting to a computer or cloud server; implementing a predictive algorithm on the computer or cloud server that uses machine learning techniques to predict from at least the biomarker measurements the probability of blood clot formation in the patient; based on the prediction of probability of blood clot formation generated by the predictive algorithm, calculating a risk score; and at least one of generating an alert representing the status of the patient or initiating a therapeutic response to development of a blood clot.

[0055] An alert may not be generated if risk status is unchanged or not concerning. If an alert is issued, change in risk status of the patient will be communicated and a response action taken depending on the respondent and the severity of the status change.

[0056] The method may be repeated at timed intervals ranging from, e.g., between 5 minutes and 1 hour.

[0057] The method for creating localized blood flow may include at least one of but not limited to the following: electrical muscle stimulation; electrical nerve stimulation; pneumatic compression; ultrasound; massage (localized compression); or vibration.

[0058] The method for measuring biomarkers may include at least one of but not limited to the following: electrical impedance measurements; bio-capacitance measurements; thermal imaging; ultrasound imaging; auscultation; photoplethysmography; impedance plethysmography; or strain gauge plethysmography.

[0059] The method may further include implementing a response to developing a blood clot in the body of the patient thr ough at least one of but not limited to the following: electrical neuromuscular' stimulation; electrical nerve stimulation; pneumatic compression; ultrasound; massage; vibration; or phototherapy.

|0060| The response may further include initiating therapy by, e.g., initiating drug therapy(ies) via communication with a drug delivery device, consulting a specialist, employing community resources, suggesting emergency services, and the like.

[0061] The corresponding system of monitoring for blood clots may include a patch or a sleeve having an outer layer made from flexible, waterproof material and an inner layer having electrical connections and a plurality of physiological sensors configured to measure biomarkers and a plurality of electrodes configured to measure bio-impedance and to deliver electrical stimulation. The system also includes a detachable electronic/stimulation device having a microcontroller that connects to the electrical connections and includes timing control circuitry, data acquisition circuitry to acquire data from the plurality of physiological sensors and bio-impedance electrodes, signal conditioning circuitry configured to process data from physiological sensors, and impedance measurement circuitry configured to process data from the bio-impedance electrodes, as well as signal generator circuitry configured to provide electrical stimulation to create localized blood flow and current injection for impedance measurement. Communication circuitry transfers data to and from the detachable electronic/stimulation device. A memory is also provided to store acquired data and program instructions, and a battery is provided to power the device to enable ambulatory movement. [0062] The physiological sensors may be set to pre-determined settings based on patient specific information that includes at least one of but not limited to the following: existing disease states or diagnoses; physiological conditions that affect or are affected by blood flow and coagulation; recent history of surgery or procedures; use of blood flow and coagulation altering therapies; patient demographics; and social determinants of health. The physiological sensors may include photoplethysmography sensors, strain gauge plethysmography sensors, impedance plethysmography sensors, electrodes, ultrasound sensors, bio-capacitance sensors, infrared thermopile sensors, electrodermal activity sensors, galvanic skin response sensors, and the like.

[0063] The signal conditioning circuitry of the microcontroller may include strain gauge signal conditioning circuitry that includes isolation, bridge balance, filtering and an excitation voltage measurement; thermopile signal conditioning circuitry that includes isolation and linearization; as well as other signal conditioning circuitry needed to facilitate acquisition of physiological data.

[0064] The communication circuitry that facilitates transfer of acquired data to the application or another device may include a wireless circuit that facilitates communication using BLUETOOTH®, WI-FI®, Zigbee®, cellular data, or radio frequency circuitry to transfer data. A computer application may be configured on a smart watch, mobile phone, or tablet configured to communicate using BLUETOOTH®, WI- FI®, Zigbee®, cellular data, or radio frequency circuitry with the detachable electronic/stimulation device and a remote computer or cloud server. The acquired data may be transferred to a device located in an in-patient setting, such as a hospital or clinic, and/or configured to communicate the data to the computer application, remote computer, or cloud server or other wireless or contactless sensing device within a network including the communication circuitry.

[0065] The remote computer or cloud server may be configured to analyze patient data, implement a predictive model, calculate a risk score, and report the risk score and the like to the healthcare professional via a healthcare professional interface such as a mobile phone, a tablet, a computer, or other display interfaces for a healthcare professional of the type illustrated by way of example in FIG. 7.

CONCLUSION

[0066] While various implementations have been described above, it should be understood that they have been presented by way of example only, and not limitation. For example, any of the elements associated with the systems and methods described above may employ any of the desired functionality set forth hereinabove. Thus, the breadth and scope of a preferred implementation should not be limited by any of the above-described sample implementations. [0067] As discussed herein, the logic, commands, or instructions that implement aspects of the methods described herein may be provided in a computing system including any number of form factors for the computing system such as desktop or notebook personal computers, mobile devices such as tablets, netbooks, and smartphones, client terminals and server-hosted machine instances, and the like. Another embodiment discussed herein includes the incorporation of the techniques discussed herein into other forms, including into other forms of programmed logic, hardware configurations, or specialized components or modules, including an apparatus with respective means to perform the functions of such techniques. The respective algorithms used to implement the functions of such techniques may include a sequence of some or all the electronic operations described herein, or other aspects depicted in the accompanying drawings and detailed description below. Such systems and computer-readable media including instructions for implementing the methods described herein also constitute sample embodiments.

[0068] The processing functions described herein (e.g., with respect to FIGS. 3-9) may be implemented in software in one embodiment. The software may consist of computer executable instructions stored on computer readable media or computer readable storage device such as one or more non-transitory memories or other type of har dware-based storage devices, either local or networked. Further, such functions correspond to modules, which may be software, hardware, firmware, or any combination thereof. Multiple functions may be performed in one or more modules as desired, and the embodiments described are merely examples. The software may be executed on a digital signal processor, ASIC, microprocessor, or other type of processor operating on a computer system, such as a personal computer, server, or other computer system, turning such computer system into a specifically programmed machine.

[0069] Examples, as described herein, may include, or may operate on, processors, logic, or a number of components, modules, or mechanisms (herein “modules”). Modules are tangible entities (e.g., hardware) capable of performing specified operations and may be configured or arranged in a certain manner. In an example, circuits may be arranged (e.g., internally or with respect to external entities such as other circuits) in a specified manner as a module. In an example, the whole or part of one or more computer systems (e.g., a standalone, client or server computer system) or one or more hardware processors may be configured by firmware or software (e.g., instructions, an application portion, or an application) as a module that operates to perform specified operations. In an example, the software may reside on a machine -readable medium. The software, when executed by the underlying hardware of the module, causes the hardware to perform the specified operations.

[0070] Accordingly, the term “module” is understood to encompass a tangible hardware and/or software entity, be that an entity that is physically constructed, specifically configured (e.g., hardwired), or temporarily (e.g., transitorily) configured (e.g., programmed) to operate in a specified manner or to perform part or all of any operation described herein. Considering examples in which modules are temporarily configured, each of the modules need not be instantiated at any one moment in time. For example, where the modules comprise a general-pmpose hardware processor configur ed using software, the general-purpose hardware processor may be configured as respective different modules at different times. Software may accordingly configure a hardware processor, for example, to constitute a particular module at one instance of time and to constitute a different module at a different instance of time.

[0071] Those skilled in the art will appreciate that while the disclosure contained herein pertains to techniques for measuring blood flow and coagulation related to the development of venous blood clots that the techniques described herein may be applied to other vascular conditions.

Accordingly, these and other such applications are included within the scope of the following claims.

Claims

WHAT IS CLAIMED:

1. A method of monitoring blood flow and coagulation to predict development of blood clots in the body of a patient, comprising: positioning on a region of interest on the surface of the body of a patient a device configured to measure physiological biomarkers, create localized blood flow, and transfer data to a remote computer server; creating localized blood flow using the device; measuring biomarkers indicative of changes in blood flow and coagulation; transferring data from the biomarker measurements to a computer; implementing a predictive algorithm on the computer that uses machine learning techniques to predict from at least the biomarker measurements the probability of development of a blood clot in the patient; based on the prediction of probability of development of a blood clot generated by the predictive algorithm, calculating a risk score; and at least one of generating an alert representing the status of the patient or initiating a therapeutic response to the development of the blood clot based on the risk score.

2. The method of claim 1, further comprising repeating the blood flow creating, biomarker measuring, data transferring, predictive algorithm implementing, risk score calculating, and alert generating or therapy starting steps at timed intervals from between 5 minutes to 1 hour.

3. The method of claim 1, wherein creating localized blood flow includes at least one of: electrical muscle stimulation; electrical nerve stimulation; pneumatic compression; ultrasound application; massage (localized compression); or vibration.

4. The method of claim 1 , wherein measuring hiomarkers may include at least one of: electrical impedance measurements; bio-capacitancc measurements; thermal imaging; ultrasound imaging; auscultation; photoplethysmography ; impedance plethysmography, or strain gauge plethysmography.

5. The method of claim 1, wherein initiating a therapeutic response based on biomarker measurements and resultant risk score comprises at least one of: electrical neuromuscular stimulation; electrical nerve stimulation; pneumatic compression; ultrasound; massage; vibration; or phototherapy.

6. The method of claim 1, further comprising transferring data from the computer application in an out-patient setting or from a communication device in an in-patient setting to the computer or a cloud server, wherein the predictive algorithm is implemented on the computer or the cloud server.

7. The method of claim 1, wherein measuring biomarkers indicative of changes in blood flow and coagulation comprises initiating settings of biomarker measurement devices using an initial patient biomarker profile based on pre-assessed patient specific information highlighting an existing disease state or physiological conditions that affect or are affected by blood flow and coagulation and may alter risk for development of a blood clot.

8. The method of claim 7, wherein the pre-assessed patient specific information is based on a particular diagnosis, procedure, or blood flow and coagulation altering therapy prescribed to the patient.

9. The method of claim 7, wherein the pre-assessed patient specific information includes socioeconomic factors or social determinants of health that influence incidence of blood clots or contribute to worse outcomes in patients diagnosed with a blood clot.

10. A system for monitoring blood flow and coagulation to predict development of blood clots in the body of a patient, comprising: a patch or sleeve adapted for placement on the patient, the patch or sleeve having an outer layer made from flexible, waterproof material, an inner layer having electrical connections, a plurality of physiological sensors configured to measure biomarkers, and a plur ality of electrodes configur ed to measure bio-impedance and to deliver electrical stimulation; a detachable electronic/stimulation device having a microcontroller that connects to the electrical connections and includes timing control circuitry, data acquisition circuitry to acquire data from the plurality of physiological sensors and electrodes, signal conditioning circuitry configured to process data from the physiological sensors, and impedance measurement circuitry configured to process data from the electrodes; signal generator circuitry configured to provide electrical stimulation to create localized blood flow and current injection for impedance measurement; and communication circuitry that transfers data to and from the detachable electronic/stimulation device.

11. The system of claim 10, wherein the sleeve further comprises a strain gauge for plethysmography and at least one occlusion cuff configured to apply occlusive pressure to a target blood vessel.

12. The system of claim 10, wherein the signal generator circuitry provides current injection for impedance measurement as a constant current signal that has a frequency of between 20 kHz and 100 kHz and an amplitude of greater than 1 mA.

13. The system of claim 10, further comprising a memory to store acquired data and program instructions, and a battery to power at least one of the patch or sleeve, the detachable electronic/stimulation device, the signal generator, or the communication circuitry.

14. The system of claim 10, wherein the physiological sensors include at least one of photoplethysmography sensors, strain gauge plethysmography sensors, impedance plethysmography sensors, impedance electrodes, ultrasound sensors, bio-capacitance sensors, infrared thermopile sensors, electrodermal activity sensors, or galvanic skin response sensors.

15. The system of claim 10, wherein the signal conditioning circuitry of the microcontroller includes at least one of strain gauge signal conditioning circuitry that includes isolation, bridge balance, filtering, and an excitation voltage measurement, or thermopile signal conditioning circuitry that includes isolation and linearization.

16. The system of claim 10, further comprising a computer application configured on at least one of a small watch, a mobile phone, or a tablet configured to communicate using BLUETOOTH®, WI-FI®, Zigbee®, cellular or radio frequency circuitry with the communication circuitry, and a remote computer or cloud server configured to communicate with the computer application.

17. The system of claim 16, wherein at least one of the computer application, remote computer, or cloud server processes data transferred by the communication circuitry in a predictive algorithm that uses machine learning techniques to predict the probability of development of a blood clot in the patient and, based on the prediction of probability of development of a blood clot generated by the predictive algorithm, calculates a risk score.

18. The system of claim 17, wherein at least one of the computer application, the remote computer, or the cloud server generates an alert representing the status of the patient, initiates a therapeutic response to the development of the blood clot based on a predetermined change in the risk score, or both.

19. The system of claim 18, wherein the alert is issued using different modalities to initiate different actions in accordance with whether the recipient of the alert is a healthcare professional, the patient, or a caregiver, the different modalities including at least one of a voice or audio message, e-mail, SMS text utilizing cellular data, a chat application, an EMR alert, a telemedicine system, or an artificial intelligence generated message through a software application.

20. The system of claim 16, further comprising a device gateway that receives and transmits data from the communication circuitry and third party data sources to at least one of the computer application, the remote computer, or the cloud server.