Patient monitoring using implanted biosensors

Abstract

A computer system provides patient monitoring and treatment using an implanted biosensor. Data associated with a target joint of a patient is collected via one or more implanted biosensors. A plurality of feature values are extracted from the data. A trained classification model is used to process the plurality of feature values to select a recommendation. The recommendation is provided to mitigate a hemophilia-related injury to the target joint. Embodiments of the invention also include methods and program products for providing patient monitoring and treatment using an implanted biosensor in substantially the same manner as described above.

Description

Technical Field

[0001] Embodiments of the present invention relate to implantable biosensors, and more specifically to using implanted biosensors to monitor a patient’s joint health and provide health care to the patient. Background Technology

[0002] Hemophilia, a genetic disorder in which the blood cannot clot properly, is one of the most deadly and well-known congenital diseases. Hemophilia occurs because the liver cannot produce essential proteins and includes two distinct types: hemophilia A, which occurs when the body cannot synthesize clotting factor VIII, and hemophilia B, which occurs when the body cannot synthesize clotting factor IX. Due to recurrent bleeding in various joints (such as the elbow or knee), these joints become weaker over time; typically, a specific joint (called the "target" joint) becomes the weak point. If the target joint is not properly treated from a young age, it can rapidly deform and eventually degenerate to the point of permanent damage. Degeneration of the target joint leads to reduced mobility and pain, adversely affecting the patient's quality of life. Therefore, treatment for hemophilia can specifically seek to protect the target joint. Summary of the Invention

[0003] According to one embodiment of the present invention, a computer system provides patient monitoring and treatment using implanted biosensors. Data associated with a patient's target joint is collected via one or more implanted biosensors. Multiple feature values ​​are extracted from the data. The multiple feature values ​​are processed using a trained classification model to select a recommendation. This recommendation is provided to mitigate hemophilia-related damage to the target joint. Embodiments of the invention also include methods and procedures for patient monitoring and treatment using implanted biosensors in substantially the same manner as described above. Attached Figure Description

[0004] In general, the same reference numerals in the various figures are used to designate the same parts.

[0005] Figure 1 This is a block diagram depicting a computational environment for patient monitoring and treatment using implanted biosensors according to an embodiment of the present invention;

[0006] Figure 2 This is a diagram depicting a user environment according to an embodiment of the present invention;

[0007] Figure 3 An implantable biosensor according to an embodiment of the present invention is described;

[0008] Figure 4 An implantable biosensor according to an embodiment of the present invention is described;

[0009] Figure 5 This is a block diagram depicting a workflow for patient monitoring and treatment according to an embodiment of the present invention;

[0010] Figure 6 This is a flowchart depicting a method for monitoring and treating a patient using an implanted biosensor according to an embodiment of the present invention; and

[0011] Figure 7 This is a block diagram depicting a computing device according to an embodiment of the present invention. Detailed Implementation

[0012] This invention relates to implantable biosensors, and more specifically to using implanted biosensors to monitor a patient's joint health and provide health care to the patient. Specifically, the biosensors can be used to monitor the joints of hemophilia patients, whose joints may be particularly vulnerable to degenerative diseases due to the nature of hemophilia. Typically, while all joints may suffer damage over time due to bleeding, the target joint often presents itself as a weak point and is therefore susceptible to recurrent bleeding.

[0013] Because untreated or poorly managed target joints can essentially lead to a positive feedback loop of increased deterioration, healthcare providers aim to identify and treat target joints as early as possible. However, identifying target joints can be difficult, especially in children, and because hemophilia is a congenital disease, most people with hemophilia suffer permanent damage to a joint (usually the knee or elbow). In severe cases, even those with hemophilia receiving conventional treatment (e.g., clotting factor replacement therapy) often lose use of their knees in their early twenties and require total knee replacement surgery. However, these surgeries only restore a portion of the knee's natural capacity, and the patient will have to live with an artificial knee or undergo additional knee replacement surgery over time.

[0014] Because surgery on hemophiliac patients can be particularly dangerous, and because joint replacements cannot fully reproduce the joint's natural capabilities, healthcare providers aim to prevent the need for joint surgery, or at least prolong it for as long as possible. Therefore, embodiments of the present invention utilize implanted biosensors to monitor the state of the target joint in order to significantly minimize the amount of time without detected internal bleeding, thereby enabling rapid treatment and thus extending the lifespan of the natural joint. Specifically, embodiments of the present invention use biochemical sensors to monitor markers in the blood, mechanical sensors to measure joint strain and other mechanical properties, and / or synovial fluid sensors to measure synovial fluid-related parameters of the joint. Machine learning techniques can be employed to perform data cleaning, and the trained classification algorithm can then provide recommendations related to the patient's target joint.

[0015] Therefore, embodiments of the present invention provide practical applications in patient monitoring and treatment by more rapidly identifying internal bleeding in target joints of hemophilia patients, thereby significantly extending the use of patients' natural joints. Embodiments of the present invention improve bleeding detection by providing a novel machine learning-based method to reduce noise in collected data. Conventional methods struggle to analyze data collected from patients due to noise introduced from internal and external sources, including patient movement, the patient's environment (due to external forces acting on the patient, ambient temperature levels, etc.), and any other sources. In contrast, embodiments of the present invention remove noise from the data by employing machine learning techniques to establish thresholds for each type of biosensor, thereby performing data cleaning operations on the data derived from each type of biosensor.

[0016] Therefore, the embodiments presented herein offer improved accuracy compared to conventional techniques that struggle to account for patient movement (which introduces noise into the collected data). Furthermore, the embodiments of the invention employ a machine learning model specifically trained to perform multi-class classification to generate recommendations, including orthopedic and hemolytic interventions that can prolong the natural lifespan of joints. Thus, the embodiments of the invention improve the field of hemophilia-related bleeding detection by identifying internal bleeding more quickly and accurately and providing actionable recommendations to mitigate detected bleeding. Moreover, patient outcomes can be used to update the classification model, thereby providing increasingly accurate solutions over time.

[0017] It should be noted that references to features, advantages, or similar language throughout this specification do not imply that all features and advantages achievable with the embodiments disclosed herein should be, or are, included in any single embodiment of the invention. Rather, language relating to features and advantages should be understood to mean that a particular feature, advantage, or characteristic described in connection with an embodiment is included in at least one embodiment of the invention. Therefore, the discussion of features, advantages, and similar language throughout this specification may, but does not necessarily, refer to the same embodiment.

[0018] Furthermore, the features, advantages, and characteristics of the invention described herein can be combined in one or more embodiments in any suitable manner. Those skilled in the art will recognize that the invention can be practiced without one or more specific features or advantages of a particular embodiment. In other instances, additional features and advantages that may not be present in all embodiments of the invention may be recognized in certain embodiments.

[0019] These features and advantages will become more apparent from the following drawings, description and appended claims, or may be learned through practice of the embodiments of the invention set forth below.

[0020] Embodiments of the present invention will now be described in detail with reference to the accompanying drawings. Figure 1 This is a block diagram depicting a computing environment 100 for patient monitoring and treatment using implanted biosensors according to an embodiment of the present invention. As depicted, the computing environment 100 includes a data collection device 105, a biosensor 110, a treatment server 115, and a network 145. It should be understood that the functional divisions among the components of the computing environment 100 have been selected for the purpose of explaining embodiments of the invention, and these functional divisions should not be construed as limiting examples.

[0021] Data collection device 105 includes a network interface (I / F) 106 and at least one processor 107. Data collection device 105 may include a laptop computer, tablet computer, netbook computer, personal computer (PC), desktop computer, personal digital assistant (PDA), smartphone, thin client, or any programmable electronic device capable of executing computer-readable program instructions. Network interface 106 enables components of data collection device 105 to send and receive data via a network such as network 145. Generally, data collection device 105 receives or acquires data from an implanted biosensor such as biosensor 110 and transmits the data to a server (e.g., treatment server 115) for further processing. In some embodiments, data collection device 105 is a microcontroller. Data collection device 105 may include internal and external hardware components, such as those described above. Figure 7 A more detailed description and depiction.

[0022] Network interface 106 enables components of data collection device 105 to send and receive data via a network such as network 145. In some embodiments, network interface 106 includes multiple interfaces, such as wired and / or wireless interfaces. Network interface 106 may include a near field communication (NFC) interface, a body area network (BAN) interface, or a similar interface to support wireless exchange of data between data collection device 105 and biosensor 110, which may be implanted in a target joint of a patient. In one embodiment, data collection device 105 uses a first short-range wireless network interface to obtain data from biosensor 110 and uses a second wired or wireless network interface to provide data to treatment server 115.

[0023] In some embodiments, processor 107 preprocesses and / or otherwise prepares data from biosensor 110 for transmission for further processing (e.g., by organizing data, compressing data, etc.). Processor 107 may organize data according to sensor type and / or may add timestamps to data collected by biosensor 110 to generate time-series data.

[0024] Biosensor 110 may include any transducer for converting physical, chemical, and / or biological phenomena into electrical signals. In various embodiments, biosensor 110 may include temperature sensors, pressure sensors, stress sensors, strain sensors, acceleration sensors, rotation sensors, biochemical sensors, etc. In some embodiments, biosensor 110 includes a surface to which specific ions or organic molecules (including macromolecules and monomers), such as antibodies, carbohydrates, calcium, etc., can bind. When a specific ion or organic molecule of interest binds to the surface of the biosensor, the conductivity of the surface changes, and the resulting voltage change can be used to determine the in vivo concentration of the specific ion or organic molecule. Biosensor 110 may be implanted via microsurgical techniques (e.g., injection into a target joint) and may include technologies such as microelectromechanical systems (MEMS) sensors, two-dimensional nanomaterial sensors (e.g., graphene-based sensors), etc. In some embodiments, biosensor 110 may use power obtained from data collection device 105 to transmit data, which may be provided via wireless power transfer technology. Figures 3-5 The biosensor 110 is described in more detail.

[0025] Treatment server 115 includes a network interface (I / F) 116, at least one processor 117, memory 120, and database 140. Memory 120 may include a data cleaning module 125, a preprocessing module 130, and a treatment planning module 135. Treatment server 115 may include a laptop computer, tablet computer, netbook computer, personal computer (PC), desktop computer, personal digital assistant (PDA), smartphone, thin client, rack server, or any programmable electronic device capable of executing computer-readable program instructions. Network interface 116 enables components of treatment server 115 to send and receive data via a network such as network 145. Typically, treatment server 115 processes and analyzes data collected from biosensors 110 to identify internal bleeding in target joints of hemophilia patients. Additionally or alternatively, treatment server 115 classifies bleeding events to recommend various treatment plans for patients. Treatment server 115 may include internal and external hardware components, such as those related to… Figure 7 A more detailed description and depiction.

[0026] The data cleaning module 125, preprocessing module 130, and treatment planning module 135 may include one or more modules or units that perform the various functions described in the embodiments of the invention below. The data cleaning module 125, preprocessing module 130, and treatment planning module 135 may be implemented by any combination of any number of software and / or hardware modules or units, and may reside in the memory 120 of the treatment server 115 for execution by a processor such as processor 117.

[0027] The data cleaning module 125 can receive data collected from the biosensors 110 via the data collection device 105 and can perform data cleaning and / or noise reduction operations on the data to exclude any data that does not meet certain predefined or determined criteria. In some embodiments, the data cleaning module 125 compares the data from each type of biosensor 110 with a predetermined threshold, and if the data violates the threshold, the data can be discarded. Thus, for example, a large value of a pressure reading can be discarded because such a value can be caused by rapid movement of the patient and therefore does not indicate typical pressure on the joint.

[0028] In some embodiments, the data cleaning module 125 utilizes a machine learning model to determine dynamic thresholds for each type of biosensor data. The machine learning model can be trained using conventional or other machine learning techniques; in some embodiments, the machine learning model is a neural network trained to output thresholds based on inputs describing the patient's target joint and other relevant factors. Specifically, the data cleaning module 125 may employ a machine learning model that outputs thresholds for data cleaning based on inputs such as the patient's hemophilia status (e.g., mild or severe), a specific joint (e.g., elbow, knee, ankle, etc.), the target joint's flexibility (e.g., normal, hypermobility, etc.), and the patient's joint condition (e.g., better than six months ago, worse than six months ago, etc.). The machine learning model can be trained using a labeled training dataset comprising example thresholds labeled with one or more input parameter values ​​(e.g., patient's condition, affected target joint, target joint flexibility, target joint condition). Thus, the machine learning model can learn thresholds for a variety of inputs, including combinations of inputs for which no example input combinations are provided.

[0029] Using predetermined and / or machine learning thresholds, the data cleaning module 125 cleans the raw data obtained from the biosensor 110. After cleaning, the results are provided to the preprocessing module 130 for further processing.

[0030] The preprocessing module 130 performs preprocessing operations on the cleaned data to extract feature values ​​from the biosensor data. The extracted feature values ​​may depend on the type of biosensor from which each dataset is extracted.

[0031] In some embodiments, the preprocessing module 130 processes the cleaned data to determine the rate of change of osteoarthritis stage, including daily change, weekly change, monthly change, and standard deviation. Osteoarthritis stage may be determined based on joint temperature data, synovial fluid pressure data, hydroxyapatite levels, osteocalcin levels, synovial volume, user-reported data, healthcare provider-reported data, etc.

[0032] In some embodiments, the preprocessing module 130 determines the rate of temperature change in the target joint, such as daily change, weekly change, monthly change, and standard deviation. Temperature data can be obtained using biosensors that measure temperature using conventional or other techniques.

[0033] In some embodiments, the preprocessing module 130 determines the rate of change of cartilage distribution, including daily variation, weekly variation, monthly variation, and standard deviation. Cartilage stress distribution can be determined based on mechanical parameters such as strain, stress, friction, pressure, etc.

[0034] In some embodiments, the preprocessing module 130 determines the rate of change of synovial fluid pressure or volume, including daily, weekly, monthly, and standard deviation. The synovial fluid pressure can be determined using a biosensor 110 implanted within the synovium of the target joint, and the volume can be determined based on comparisons of the pressure data with expected pressure of the synovium at a specific joint, patient body measurements, etc.

[0035] In some embodiments, the preprocessing module 130 determines the rate of change of antihemophilic factor (AHF) levels, including daily changes, weekly changes, monthly changes, and standard deviations. AHF levels can be determined using a biosensor 110, which performs measurements to determine AHF concentration directly or indirectly based on the presence of metabolites or other organic molecules.

[0036] In some embodiments, the preprocessing module 130 determines the rate of change of Bethesda units (BU), including daily change, weekly change, monthly change, and standard deviation. The biosensor 110 can be used to detect BU to perform a Bethesda assay to detect the presence of factor VIII inhibitors (i.e., anti-factor VIII antibodies).

[0037] In some embodiments, the preprocessing module 130 determines the rate of change of hemoglobin levels, including daily changes, weekly changes, monthly changes, and standard deviation. Hemoglobin levels can be determined using an optofluid sensor to measure the amount of hemoglobin in a fluid.

[0038] In some embodiments, the preprocessing module 130 determines the rate of change of blood levels in the synovial fluid, including daily, weekly, monthly, and standard deviation. Blood levels in the synovial fluid can be detected using a microneedle-based biosensor 110 based on capillary action measurements.

[0039] When the preprocessing module 130 extracts feature values ​​from the cleaned biosensor data, the preprocessing module 130 can provide the feature values ​​to the treatment planning module 135.

[0040] The treatment planning module 135 may employ one or more trained machine learning models to analyze extracted feature values ​​to assess the condition of the target joint and recommend treatment for the target joint. In some embodiments, the treatment planning module 135 employs conventional or other machine learning techniques to map feature values ​​to specific recommendations to mitigate damage to the target joint. The machine learning model may include conventional or other model types, such as neural networks, support vector machines, hidden Markov models, adversarial network models, etc. In some embodiments, the machine learning model is a gradient boosting-based model that performs multi-class classification to provide recommendations based on input feature values.

[0041] Machine learning models can be trained to classify specific feature values ​​or combinations of feature values ​​to the condition of a target joint and / or a treatment recommendation for that target joint. Specifically, a supervised machine learning model can be trained using training data comprising feature values ​​of one or more extracted features, each feature value being provided with a label indicating whether the feature value is associated with internal bleeding, and if so, the amount of internal bleeding. The amount or condition of the target joint can be associated with a specific treatment recommendation, and those associations can be learned using labeled training data, which includes examples of joint conditions and corresponding treatment options. Thus, the machine learning model can be trained to output a mapping of feature values ​​to joint conditions, which are then used to recommend treatments.

[0042] A machine learning model can be trained using training data containing examples of past mappings of treatment values ​​to known joint conditions and / or treatments used to provide care to previous patients and labeled with positive or negative patient outcomes (e.g., indicators of whether the recommendation satisfactorily alleviated the problem in the patient's target joint). Therefore, the treatment planning module 135 can learn from successful and / or unsuccessful previous mappings to more accurately map feature values ​​to specific target joint conditions and / or recommendations.

[0043] Database 140 may include any non-volatile storage medium known in the art. For example, database 140 may be implemented using a tape library, an optical library, one or more individual hard drives, or multiple hard drives in a redundant array of independent disks (RAID). Similarly, the data in database 140 may follow any suitable storage architecture known in the art, such as files, relational databases, object-oriented databases, and / or one or more tables. In some embodiments, database 140 may store data including trained machine learning models, training data, raw biosensor data, cleaned biosensor data, extracted feature values, historical mappings of feature values ​​to target joint conditions, and / or historical mappings of target joint conditions to recommendations.

[0044] Network 145 may include a local area network (LAN), a wide area network (WAN) (such as the Internet), or a combination of both, and may include wired, wireless, or fiber optic connections. Typically, according to embodiments of the invention, network 145 may be any combination of connections and protocols known in the art that will support communication between data collection device 105 and treatment server 115 via their respective network interfaces.

[0045] Figure 2 This is a diagram depicting a user environment 200 according to an embodiment of the present invention. As depicted, the user environment 200 includes a user 205 (e.g., a hemophilia patient), a data collection device 105, and one or more biosensors 110. The biosensors 110 may be implanted in a target joint, which in the depicted embodiment is the left knee of the user 205. In some embodiments, the biosensors 110 may be located in or around the target joint, including one or more biosensors 110 within the synovium, one or more biosensors 110 positioned to monitor circulatory system components of the target joint, and one or more biosensors 110 positioned to obtain mechanical measurements (such as stress and strain) of the target joint.

[0046] Data collection device 105 can wirelessly acquire data from biosensor 110. The data acquired by data collection device 105 can then be provided to a server, such as processing server 115, for further processing. In some embodiments, data collection device 105 is a wearable device or can be carried in a user's pocket, etc. Data collection device 105 can acquire data from biosensor 110 continuously, according to a predetermined schedule, or on a self-organizing basis (e.g., when new data becomes available). Similarly, data collection device 105 can provide data to treatment server 115 continuously, self-organizingly, or according to a predetermined schedule.

[0047] Figure 3 An implantable biosensor 300 according to an embodiment of the present invention is depicted. The implantable biosensor 300 may be included in the biosensor 110, as described above. Figure 1 More detailed description and depiction are provided below. As depicted, the implantable biosensor 300 includes a synovial pressure sensor 310 and an external body 320. The synovial pressure sensor 310 may include any transducer for converting pressure into an electrical signal, such as a piezoelectric transducer. The external body 320 may be coated in a material to provide a sensor array for detecting antibodies. When antibodies adhere to the coating of the external body 320, the conductivity or other properties of the external body 320 may be altered in a manner related to the antibody concentration in the fluid in which the implantable biosensor 300 is inserted.

[0048] Figure 4 An implantable biosensor according to an embodiment of the present invention is depicted. The implantable biosensor 400 may be included in the biosensor 110, as described above. Figure 1 A more detailed description and depiction follows. The implantable biosensor 400 may correspond to an internal view of an embodiment of the implantable biosensor 300, such as... Figure 3 The depicted and described, or implantable, biosensor 400 may be another biosensor embodiment.

[0049] As depicted, the implantable biosensor 400 includes a blood-related element 410, a musculoskeletal-related element 420, a transducer 430, and a battery 440. The blood-related element 410 may include one or more on-chip laboratory or microscale assays for collecting biosensor data associated with blood-related parameters, and the musculoskeletal-related element 420 may include one or more biosensors for collecting mechanical data, such as stress and strain. The transducer 430 may include one or more transducer assemblies for converting biological signals acquired by other elements (e.g., the blood-related element 410 and / or the musculoskeletal-related element 420) into electrical signals. The battery 440 may provide power to the implantable biosensor 400 to power its data collection components and / or transmitter (not shown).

[0050] Figure 5 This is a block diagram depicting a workflow 500 for patient monitoring and treatment according to an embodiment of the present invention. As depicted, workflow 500 includes a data acquisition phase 505, a data transmission and cleaning phase 525, a feature extraction phase 545, a feature processing phase 560, and a treatment phase 570.

[0051] Data acquisition phase 505 includes the acquisition of circulatory data 510, mechanical data 515, and synovial fluid data 520. As depicted, circulatory data 510 may include AHF data (including coagulation factor VIII, coagulation factor IX, or both), inhibitor data, hemoglobin data, and hydroxyapatite data. AHF data can be used to determine the level of AHF degradation in a patient's blood, which is associated with an increased chance of internal bleeding. Since hemophiliac patients cannot produce their own AHF, the presence of AHF is due to the AHF treatment that the patient must receive regularly. AHF levels can be determined using a biosensor that performs a one-stage activated partial thromboplastin time (aPTT)-based assay, a two-stage chromogenic assay, or a similar assay.

[0052] Inhibitor data can be acquired via a biosensor that performs a Bethesda assay to determine Bethesda units (BU) in a liquid. The Bethesda assay measures anti-factor VIII antibodies (e.g., IgG4 and / or IgG1 antibodies) produced by the body and inhibiting the regular activity of AHF in the blood. Inhibitors may include tissue factor pathway inhibitors (TFPIs) and / or antithrombins. Inhibitor data can be collected via the biosensor by performing a Nijmegen-Bethesda assay and / or by performing a fluorescence immunoassay (FIA) using a coating on the biosensor to detect the presence of inhibitors.

[0053] Hemoglobin data can be obtained using biosensors that include optofluidic sensors composed of nanofilters, which perform evanescent wave absorption measurements of hemoglobin.

[0054] Hydroxyapatite data can be obtained using biosensors that detect biochemical markers, such as the C-reactive protein biomarker CRPM, which can warn of joint inflammation and / or bone erosion. Therefore, the sensing surface of the biosensor can be coated with detection antibodies, such as anti-CRPM antibodies, and changes in voltage can be correlated with biomarker concentration. Additionally or alternatively, antibodies conjugated with osteocalcin (an organic molecule released when bone degrades) can similarly identify hydroxyapatite levels.

[0055] Mechanical data 515 may include strain or friction data, joint instability data, pressure data, and temperature data. Pressure data may be collected via a capacitive pressure sensor, which may include a polyethylene-coated needle or a similar needle containing a pressure sensor. Strain data may be collected using a piezoresistive strain sensor. Cartilage stress distribution data may be collected using a force plate sensor, and temperature data may be collected using a graphene-coated infrared thermopile sensor or other temperature acquisition microelectromechanical systems (MEMS).

[0056] Synovial fluid data can be collected using biosensors that evaluate capillary action in one or more microneedles 520. Capillary action can indicate synovial fluid properties, such as viscosity, which indicates blood density and / or presence. Additionally, pressure data can be collected from the synovium, similar to how pressure data is collected for mechanical data targeting a joint.

[0057] The data transmission and cleaning phase 525 includes operation 530 (where raw data is transmitted by the biosensor), operation 535 (where the transmitted raw data is collected by a device (e.g., data collection device 105)), and operation 540 (where data cleaning operations are performed by a data cleaning module 125). Once the raw data has been cleaned, the results can be provided to the preprocessing module 130, and the workflow 500 proceeds to the feature extraction phase 545.

[0058] At feature extraction stage 545, preprocessing module 130 processes the data to extract feature values ​​for various relevant features. As shown in the figure, feature data for two main categories can be obtained by performing musculoskeletal status detection 550 and hematological status detection 555. Musculoskeletal data may include the rate of change in osteoarthritis stage, the rate of change in temperature, the rate of change in cartilage stress distribution, and the rate of change in synovial fluid volume or pressure. Hematological data may include the rate of change in AHF measurement, the rate of change in inhibitor levels, the rate of change in hemoglobin levels, and the rate of change in blood levels in synovial fluid.

[0059] At the feature value processing stage 560, the extracted feature values ​​can be processed by the recommendation system 565 (e.g., the treatment planning module 135 and its trained model). The recommendation system 565 can determine the state of the target joint based on the feature values ​​and can output various different recommendations according to the state of the target joint.

[0060] As shown in treatment phase 570, recommendations may include orthopedic (surgical) recommendations 575 and hematological (clinical) recommendations 580. Orthopedic recommendations may include non-surgical recommendations, such as physical therapy or changes in physical therapy and radiation synovectomy (e.g., using Sharp 90), or surgical recommendations, such as arthroscopy or arthroplasty. Hematological recommendations 580 may include on-demand or prophylactic AHF replacement therapy, or hemoglobin-targeted therapy, such as medications or blood transfusions to increase hemoglobin levels.

[0061] Treatments or combinations of treatments can be applied at operation 585, and the results can be fed back into the recommendation system 565 so that the machine learning components can improve accuracy over time.

[0062] Figure 6 This is a flowchart depicting a method 600 for monitoring and treating a patient using an implanted biosensor according to an embodiment of the present invention.

[0063] At operation 610, the biosensor is implanted into the patient. The biosensor can be implanted into the patient's target joint using microsurgery or other techniques, including minimally invasive techniques. The location of the biosensor can be verified to ensure proper data collection. Depending on the type of biosensor(s) implanted, the biosensor(s) can be implanted intrasynovially or extrasynovially. The biosensor can be inserted into or adjacent to the capillary bed associated with the target joint.

[0064] At operation 620, data is acquired from the biosensor. The data can be acquired by the biosensor, which, when implanted in the patient, can then share data with another device (e.g., data collection device 105) via wireless communication. The acquired data can then be uploaded to a server (such as treatment server 115) for further processing.

[0065] At operation 630, data cleaning is performed on the raw data. Data cleaning can be performed by discarding outliers and / or any data values ​​that exceed a predetermined or machine learning threshold. Different thresholds can be used for each data type; for example, one threshold can be applied to clean hemoglobin data, and another threshold can be applied to clean stress data. Thresholds can include an upper limit, a lower limit, or both.

[0066] At operation 640, feature values ​​are extracted from the cleaned data. The cleaned data can be processed to extract features, including the rates of change of various tracked features and their corresponding standard deviations.

[0067] At operation 650, the condition of the target joint is determined. Feature values ​​are provided to a multi-class classification model, which identifies the specific condition of the target joint and determines recommendations based on that condition. This condition may include the presence or absence of internal bleeding, the severity of the bleeding, and / or other conditions related to the target joint.

[0068] At operation 660, recommendations are provided to mitigate damage to the target joint. These recommendations can be selected using a machine learning model trained on the correlation between the target joint condition and the corresponding treatment; therefore, the machine learning model can output one or more treatment options based on the determined condition of the target joint.

[0069] At operation 670, the recommendation model is updated based on the results of the applied recommendations. When the recommended treatment is applied to a patient, the results of the treatment application are recorded and can be used as additional training data for the recommendation model. Therefore, the recommendation model can be updated over time to improve the accuracy and effectiveness of the recommendations.

[0070] Figure 7This is a block diagram depicting the components of a computer 10 suitable for performing the methods disclosed herein. The computer 10 may implement a data collection device 105 and / or a treatment server 115 according to embodiments of the present invention. It should be understood that... Figure 7 The illustration is provided only as an example and does not imply any limitation regarding the environment in which different embodiments may be implemented. Many modifications may be made to the depicted environment.

[0071] As depicted, computer 10 includes a communication structure 12 that provides communication between computer processor(s) 14, memory 16, persistent storage device 18, communication unit 20, and input / output (I / O) interfaces(s) 22. Communication structure 12 can be implemented using any architecture designed to transfer data and / or control information between processors (such as microprocessors, communication and network processors, etc.), system memory, peripheral devices, and any other hardware components within the system. For example, communication structure 12 can be implemented using one or more buses.

[0072] Memory 16 and persistent storage device 18 are computer-readable storage media. In the depicted embodiment, memory 16 includes random access memory (RAM) 24 and cache memory 26. Typically, memory 16 may include any suitable volatile or non-volatile computer-readable storage medium.

[0073] One or more programs may be stored in persistent storage device 18 for execution by one or more of the respective computer processors 14 via one or more memories in memory 16. Persistent storage device 18 may be a magnetic hard disk drive, a solid-state hard disk drive, a semiconductor storage device, a read-only memory (ROM), an erasable programmable read-only memory (EPROM), flash memory, or any other computer-readable storage medium capable of storing program instructions or digital information.

[0074] The media used by persistent storage device 18 can also be removable. For example, a removable hard disk drive can be used for persistent storage device 18. Other examples include optical discs and disks, thumb drives and smart cards, which are inserted into the drive for transfer to another computer-readable storage medium (which is also part of persistent storage device 18).

[0075] In these examples, communication unit 20 provides communication with other data processing systems or devices. In these examples, communication unit 20 includes one or more network interface cards. Communication unit 20 can provide communication using either or both physical and wireless communication links.

[0076] Multiple I / O interfaces 22 allow data input and output to other devices that can be connected to computer 10. For example, I / O interfaces 22 can provide connectivity to external devices 28, such as keyboards, keypads, touchscreens, and / or other suitable input devices. External devices 28 may also include portable computer-readable storage media, such as thumb drives, portable optical discs or disks, and memory cards.

[0077] Software and data used to practice embodiments of the present invention can be stored on such portable computer-readable storage media and can be loaded onto permanent storage device 18 via I / O interfaces 22(s). I / O interfaces 22 can also be connected to display 30. Display 30 provides a mechanism for displaying data to a user and can be, for example, a computer monitor.

[0078] The procedures described herein are identified based on their application as implemented in specific embodiments of the invention. However, it should be understood that any particular procedural terminology used herein is for convenience only, and therefore the invention should not be limited to use only in any particular application identified and / or implied by such terminology.

[0079] Data related to patient monitoring and treatment using implanted biosensors (e.g., raw biosensor data, cleaned biosensor data, extracted feature values, trained machine learning models, training data, historical mappings of feature values ​​to target joint conditions and / or historical mappings of target joint conditions to recommendations, etc.) can be stored in any conventional or other data structure (e.g., file, array, list, stack, queue, record, etc.) and in any desired storage unit (e.g., database, data repository or other storage, queue, etc.). Data transferred between data collection device 105 and / or treatment server 115 can include any desired format and arrangement and can include any number of fields of any type and size for storing data. The definition and data model of any dataset can indicate the overall structure in any desired manner (e.g., computer-related language, graphical representation, list, etc.).

[0080] Data related to patient monitoring and treatment using implanted biosensors (e.g., raw biosensor data, cleaned biosensor data, extracted feature values, trained machine learning models, training data, historical mappings of feature values ​​to target joint conditions and / or historical mappings of target joint conditions to recommendations, etc.) may include any information provided to or generated by data collection device 105 and / or treatment server 115. Data related to patient monitoring and treatment using implanted biosensors may include any desired format and arrangement, and may include any number of fields of any type and size to store any desired data. Data related to patient monitoring and treatment using implanted biosensors may include any data about the entity collected through any collection mechanism, any combination of collected information, and any information derived from analyzing the collected information.

[0081] Embodiments of the present invention can employ any number and type of user interface (e.g., graphical user interface (GUI), command line, prompts, etc.) to obtain or provide information (e.g., data related to patient monitoring and treatment using electronic textiles), wherein the interface may include any information arranged in any manner. The interface may include any number and type of input or actuation mechanisms (e.g., buttons, icons, fields, boxes, links, etc.) arranged in any location to input / display information and initiate desired actions via any suitable input device (e.g., mouse, keyboard, etc.). The interface screen may include any suitable actuators (e.g., links, tabs, etc.) to navigate between screens in any manner.

[0082] It will be understood that the embodiments described above and shown in the accompanying drawings represent only a few of the many ways in which implanted biosensors can be used to improve patient monitoring and treatment.

[0083] The environment of this invention embodiment can include any number of computers or other processing systems (e.g., client or end-user systems, server systems, etc.) and databases or other repositories arranged in any desired manner, wherein this invention embodiment can be applied to any desired type of computing environment (e.g., cloud computing, client-server, network computing, mainframe, standalone systems, etc.). The computers or other processing systems employed in this invention can be implemented by any number of any personal or other type of computers or processing systems (e.g., desktop computers, laptop computers, PDAs, mobile devices, etc.), and can include any commercially available operating system and any combination of commercially available software and custom software (e.g., communication software, server software, data cleaning module 125, preprocessing module 130, treatment planning module 135, etc.). These systems can include any type of monitor and input device (e.g., keyboard, mouse, voice recognition, etc.) for inputting and / or viewing information.

[0084] It should be understood that the software of the embodiments of the present invention (e.g., communication software, server software, data cleaning module 125, preprocessing module 130, treatment planning module 135, etc.) can be implemented in any desired computer language and can be developed by those skilled in the art based on the functional descriptions contained in the specification and the flowcharts shown in the accompanying drawings. Furthermore, any references herein to software performing various functions generally refer to a computer system or processor performing these functions under software control. The computer system of the embodiments of the present invention can alternatively be implemented by any type of hardware and / or other processing circuitry.

[0085] Different functionalities of a computer or other processing system can be distributed in any manner across any number of software and / or hardware modules or units, processing or computer systems and / or circuits, wherein the computer or processing system can be arranged locally or remotely and communicate with each other via any suitable communication medium (e.g., LAN, WAN, intranet, Internet, hardwired, modem connection, wireless, etc.). For example, the functionalities of embodiments of the present invention can be distributed in any manner across different end-user / client and server systems and / or any other intermediate processing devices. The software and / or algorithms described above and shown in the flowcharts can be modified in any way to implement the functionalities described herein. Furthermore, the functionalities in the flowcharts or specifications can be executed in any order to achieve the desired operations.

[0086] The software of this invention (e.g., communication software, server software, data cleaning module 125, preprocessing module 130, treatment planning module 135, etc.) can be available on non-transitory computer-usable media (e.g., magnetic or optical media, magneto-optical media, floppy disks, CD-ROMs, DVDs, storage devices, etc.) for use with stand-alone systems or systems connected via networks or other communication media.

[0087] The communication network can be implemented by any number and type of communication network (e.g., LAN, WAN, Internet, intranet, VPN, etc.). The computer or other processing system of embodiments of the present invention may include any conventional or other communication device that communicates over a network via any conventional or other protocol. The computer or other processing system may utilize any type of connection for accessing the network (e.g., wired, wireless, etc.). The local communication medium may be implemented by any suitable communication medium (e.g., local area network (LAN), hardwired, wireless link, intranet, etc.).

[0088] This system can utilize any number of any conventional or other databases, data storage, or storage structures (e.g., files, databases, data structures, data, or other repositories) to store information (e.g., data related to patient monitoring and treatment using implanted biosensors). The database system can be implemented using any number of any conventional or other databases, data storage, or storage structures (e.g., files, databases, data structures, data, or other repositories) to store information (e.g., data related to patient monitoring and treatment using electronic textiles). The database system can be included within or coupled to server and / or client systems. The database system and / or storage structure can be located remotely from or local to a computer or other processing system and can store any desired data (e.g., data related to patient monitoring and treatment using implanted biosensors).

[0089] Embodiments of the present invention can employ any number and type of user interface (e.g., graphical user interface (GUI), command line, prompts, etc.) to obtain or provide information (e.g., data related to patient monitoring and treatment using implanted biosensors), wherein the interface may include any information arranged in any manner. The interface may include any number and type of input or actuation mechanisms (e.g., buttons, icons, fields, boxes, links, etc.) arranged in any location to input / display information and initiate desired actions via any suitable input device (e.g., mouse, keyboard, etc.). The interface screen may include any suitable actuators (e.g., links, tabs, etc.) to navigate between screens in any manner.

[0090] The embodiments of the present invention are not limited to the specific tasks or algorithms described above, but can be used in any number of applications in the relevant fields, including but not limited to providing improved health care to patients by identifying and treating health conditions of joints and other body parts, including providing health care to patients who do not have hemophilia, have parahemorrhodonia, or have acquired hemophilia.

[0091] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the invention. As used herein, unless the context clearly indicates otherwise, the singular forms “a,” “an,” and “the” are intended to also include the plural forms. It should also be understood that when the terms “comprises,” “comprising,” “includes,” “including,” “has,” “have,” “having,” “with,” etc., are used in this specification, they specify the presence of the stated features, integers, steps, operations, elements, and / or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or combinations thereof.

[0092] All components or steps in the following claims, along with their corresponding structures, materials, actions, and equivalents of functional elements, are intended to encompass any structure, material, or action used to perform a function in conjunction with other claimed elements, as specifically claimed. The description of the invention has been presented for purposes of illustration and description, but is not intended to be exhaustive or limited to the forms disclosed. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the invention. Embodiments were chosen and described in order to best explain the principles and practical application of the invention, and to enable others skilled in the art to understand various embodiments of the invention with various modifications suitable for the intended particular purpose.

[0093] Various embodiments of the invention have been described for illustrative purposes, but are not intended to be exhaustive or limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments. The terminology used herein has been chosen to best explain the principles of the embodiments, their practical application, or technical improvements to technologies found in the market, or to enable those skilled in the art to understand the embodiments disclosed herein.

[0094] This invention can be a system, method, and / or computer program product with any possible level of technical detail. The computer program product may include a computer-readable storage medium (or media) having computer-readable program instructions thereon for causing a processor to execute aspects of the invention.

[0095] Computer-readable storage media can be tangible devices that can hold and store instructions for use by an instruction execution device. Computer-readable storage media can be, for example, but not limited to, electronic storage devices, magnetic storage devices, optical storage devices, electromagnetic storage devices, semiconductor storage devices, or any suitable combination of the foregoing. A non-exhaustive list of more specific examples of computer-readable storage media includes: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), static random access memory (SRAM), portable compact disk read-only memory (CD-ROM), digital universal disk (DVD), memory sticks, floppy disks, mechanical encoding devices such as punch cards, or protrusions in slots having instructions recorded thereon, and any suitable combination of the foregoing. As used herein, computer-readable storage media should not be construed as transient signals themselves, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through waveguides or other transmission media (e.g., light pulses passing through fiber optic cables), or electrical signals transmitted through wires.

[0096] The computer-readable program instructions described herein can be downloaded from a computer-readable storage medium to a corresponding computing / processing device or to an external computer or external storage device via a network (e.g., the Internet, a local area network, a wide area network, and / or a wireless network). The network may include copper transmission cables, optical transmission fibers, wireless transmissions, routers, firewalls, switches, gateway computers, and / or edge servers. A network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards them to a computer-readable storage medium within the corresponding computing / processing device.

[0097] Computer-readable program instructions used to perform the operations of this invention may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state setting data, integrated circuit configuration data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages ​​(such as Smalltalk, C++, etc.) and procedural programming languages ​​(such as the "C" programming language or similar programming languages). The computer-readable program instructions may be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the latter scenario, the remote computer may be connected to the user's computer via any type of network (including a local area network (LAN) or a wide area network (WAN)) or may be connected to an external computer (e.g., via the Internet using an Internet service provider). In some embodiments, electronic circuitry including, for example, programmable logic circuitry, field-programmable gate arrays (FPGAs), or programmable logic arrays (PLAs) may execute computer-readable program instructions by utilizing state information from the computer-readable program instructions to personalize the electronic circuitry in order to perform aspects of this invention.

[0098] This document describes aspects of the invention with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer-readable program instructions.

[0099] These computer-readable program instructions may be provided to a computer processor or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the computer processor or other programmable data processing apparatus, create methods for implementing the functions / actions specified in one or more boxes of a flowchart and / or block diagram. These computer-readable program instructions may also be stored in a computer-readable storage medium that directs a computer, programmable data processing apparatus, and / or other device to operate in a particular manner, such that the computer-readable storage medium storing the instructions comprises an article of manufacture containing instructions that implement aspects of the functions / actions specified in one or more boxes of a flowchart and / or block diagram.

[0100] These computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus or other device to cause a series of operational steps to be performed on the computer, other programmable apparatus or other device to produce a computer-implemented process, such that the instructions, which execute on the computer, other programmable apparatus or other device, implement the functions / actions specified in one or more boxes of a flowchart and / or block diagram.

[0101] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. Each block in a flowchart or block diagram may represent a module, segment, or portion of instructions, including one or more executable instructions for implementing a specified logical function. In some alternative implementations, the functions marked in the blocks may occur in a different order than those marked in the figures. For example, two blocks shown consecutively may actually be completed as a single step, executed simultaneously, substantially simultaneously, or with partial or complete temporal overlap, or the blocks may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented using a dedicated hardware-based system that performs the specified function or action or executes a combination of dedicated hardware and computer instructions.

Claims

1. A computer-implemented method for monitoring a patient using an implanted biosensor, the computer-implemented method comprising: Data associated with a target joint in a hemophilia patient is collected via one or more implanted biosensors, wherein the target joint corresponds to the natural joint of the hemophilia patient; Before extracting multiple feature values ​​from the data, data cleaning is performed on the data to generate cleaned data, wherein the data cleaning includes: Apply machine learning models to identify different thresholds for each of the multiple data types in the data; and Discard data values ​​of a first data type from the plurality of data types that violate a first threshold, wherein the first threshold is derived from the different thresholds and is identified for the first data type. Extract the multiple feature values ​​from the cleaned data; A trained classification model is used to process the multiple feature values ​​from the cleaned data to select a recommendation; and The recommendations are provided to mitigate hemophilia-related damage to the target joint.

2. The computer-implemented method according to claim 1, wherein, The data includes one or more of the following: cyclic data, mechanical data, and synovial fluid data.

3. The computer-implemented method according to claim 1, wherein, The multiple feature values ​​extracted from the cleaned data include one or more of the following: the rate of change of osteoarthritis stage, the rate of change of temperature, the rate of change of cartilage stress distribution, the rate of change of synovial fluid pressure, the rate of change of antihemophilic factor, the rate of change of Bethesda units, the rate of change of hemoglobin, and the rate of change of blood density in synovial fluid.

4. The computer-implemented method according to claim 1, wherein, The trained classification model includes an extreme gradient boosting model.

5. The computer-implemented method according to claim 1, wherein, The recommendations include one or more of the following: physical therapy, synovectomy, recommendations to perform surgery on the target joint after a defined time period, antihemophilic factor replacement therapy, drug therapy to increase hemoglobin levels, and blood transfusion therapy.

6. The computer-implemented method according to claim 1, further comprising: The trained classification model is updated based on the results of applying the recommendations to the target joint.

7. A computer system for monitoring a patient using an implanted biosensor, the computer system comprising: One or more computer processors; One or more computer-readable storage media; Program instructions stored on the one or more computer-readable storage media for execution by at least one of the one or more computer processors, the program instructions including instructions for implementing the steps of the method of any one of claims 1-6.

8. A computer program product for monitoring a patient using an implanted biosensor, the computer program product comprising program instructions executable by a computer to cause the computer to perform the steps of the method of any one of claims 1-6.

Images