Systems and methods for medical device and treatment modeling

Abstract

In one aspect, the present disclosure provides a system including a memory and a processor. The memory can store one or more instructions. The processor can execute one or more of the instructions stored on the memory to perform one or more acts, actions, or steps. For example, the processor can build a digital twin model for an individual based on individual data associated with the individual. The digital twin model can be representative of one or more physiological or anatomical characteristics of the individual. The processor can determine an individual-specific treatment for a diagnosis based on a simulated application of two or more virtual treatment options to the digital twin model.

Description

SYSTEMS AND METHODS FOR MEDICAL DEVICE AND TREATMENT MODELINGFIELD

[0001] The present technology is generally related to systems and methods for medical device and treatment modeling, such as a digital twin for healthcare.BACKGROUND OF THE INVENTION

[0002] A digital twin is a digital representation of a physical object, person, or process, contextualized in a digital version of its environment. Digital twins can help simulate real situations and outcomes. A digital twin serves as an effectively indistinguishable digital counterpart for practical purposes, such as integration, testing, monitoring, and maintenance. The use of a digital twin can allow the intended entity's lifecycle to be modeled and simulated. A digital twin of an existing entity may be used in real-time and regularly synchronized with the corresponding physical system.SUMMARY OF THE INVENTION

[0003] This summary is provided 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 identify key features or essential features of the claimed subject matter, nor is it intended to be used in isolation as an aid in determining the scope of the claimed subject matter.

[0004] Embodiments of the present disclosure are generally directed towards technologies for building and deploying digital twin models including determining individual-specific treatments based on a generated and deployed digital twin model. In particular, the technology disclosed herein facilitates building a digital twin model for an individual based on individual data associated with an individual, such as a human patient. The digital twin model may be representative of one or more physiological or anatomical characteristics of the individual. In some respects, the individual data represents health information for the individual collected from one or a plurality of data sources. For example, aggregate data from a plurality of data sources may be utilized to provide the individual data associated with the individual. In some embodiments, an individual-specific treatment is determined for a diagnosis based on a simulated application of two or more virtual treatment options to the digital twin model. Further, the diagnosis may be determined from the model, such as based on a simulated application of a stress test to the digital twin model in someimplementations. Additionally, some embodiments of the technology disclosed herein optimize a treatment parameter associated with two or more virtual treatment options when determining the individual-specific treatment for a diagnosis.

[0005] Accordingly, in one aspect, the present disclosure provides a computer system including a memory and a processor. The memory can store one or more computerexecutable instructions. The processor can execute the instructions stored on the memory to perform operations. For example, the processor can receive individual data associated with an individual. The data may be received from a custom individual database via a communication interface associated with the computer system. The individual data comprises health information for the individual collected from one or a plurality of data sources. The processor can store the individual data within the memory and generate a digital twin model corresponding to the individual, based on individual data associated with the individual. The digital twin model comprises a set of data and computer logic and is representative of one or more physiological or anatomical characteristics of the individual. In some instances, the processor can remove the individual data from the memory of the computer system after the digital twin model is generated. The processor employs the generated digital twin model to determine a particular individual-specific treatment for a diagnosis, based on a simulated application of two or more virtual treatment options to the digital twin model.

[0006] In another aspect, the disclosure provides a computer-implemented method. The computer-implemented method can include aggregating data from a plurality of data sources to provide individual data associated with the individual, the individual data being representative of health characteristics or attributes for the individual, calibrating a prediction engine based on the individual data associated with the individual, validating the prediction engine based on another part of the individual data to provide a corresponding model representative of one or more physiological characteristics of the individual, and subjecting the corresponding model to stress using to determine an individual-specific treatment for a given diagnosis, in which the stress includes test parameters representative of a simulated application of two or more virtual treatment options to the corresponding model.

[0007] In another aspect, the disclosure provides a system including a memory and a processor. The memory can store one or more instructions. The processor can execute one or more of the instructions stored on the memory to perform one or more acts, actions, or steps. For example, the processor can aggregate data from a plurality of data sources to provideindividual data associated with the individual and build a digital twin model for an individual based on the individual data associated with the individual. The digital twin model is representative of one or more physiological or anatomical characteristics of the individual. The processor can determine a diagnosis for the individual based on a simulated application of a stress test to the digital twin model and determine an individual-specific treatment for the diagnosis based on a simulated application of two or more virtual treatment options to the digital twin model.

[0008] The details of one or more aspects of the disclosure are set forth in the accompanying drawings and the description below. Other features, objects, and advantages of the techniques described in this disclosure will be apparent from the description and drawings, and from the claims.BRIEF DESCRIPTION OF THE DRAWINGS

[0009] FIG. 1A is a block diagram that illustrates a system for determining an individualspecific treatment for a diagnosis based on a digital twin model, in accordance with an embodiment of this disclosure ;

[0010] FIG. IB is a workflow diagram that illustrates an example computer-implemented method for calibrating a prediction engine, in accordance with an embodiment of the disclosure;

[0011] FIG. 2 is a workflow diagram that illustrates an example computer-implemented method for building and utilizing a digital twin model, in accordance with an embodiment of the disclosure;

[0012] FIG. 3 is a flow diagram that illustrates a computer-implemented method for determining an individualspecific treatment for a diagnosis based on a digital twin model, in accordance with an embodiment of the disclosure;

[0013] FIG. 4 is a workflow diagram that illustrates an example computer-implemented method for building a digital twin model, in accordance with an embodiment of the disclosure;

[0014] FIG. 5 is a workflow diagram that illustrates an example computer-implemented method for updating the digital twin model, in accordance with an embodiment of the disclosure; and

[0015] FIG. 6 is a block diagram of an example computing environment suitable for use in implementing an embodiment of the present disclosure.DETAILED DESCRIPTION OF THE INVENTION

[0016] The subject matter of aspects of the present disclosure is described with specificityherein to meet statutory requirements. However, the description itself is not intended to limit the scope of this patent. Rather, it is contemplated that the claimed subject matter might also be embodied in other ways, such as to include different steps or combinations of steps similar to the ones described in this document, in conjunction with other present or future technologies. Moreover, although the terms “step” and / or “block” may be used herein to connote different elements of methods employed, the terms should not be interpreted as implying any particular order among or between various steps herein disclosed unless and except when the order of individual steps is explicitly described. Each method described herein may comprise a computing process that may be performed using any combination of hardware, firmware, and / or software. For instance, various functions may be carried out by a processor executing instructions stored in memory. The methods may also be embodied as computer-useable instructions stored on computer storage media. The methods may be provided by a stand-alone application, a service or hosted service (stand-alone or in combination with another hosted service), or a plug-in to another product, to name a few.

[0017] This disclosure relates generally to systems and methods for medical device and treatment modeling, such as a digital twin model for healthcare. As described herein, a digital twin model comprises a virtual representation of a real-world object, process, or system. In some embodiments, a digital twin model comprises a set of structured data corresponding to the represented real-world object, process, or system, and a set of related computer logic, comprising computer instructions, rules, and / or associations, that specify or control how this data interacts with and represents the real-world system. In some implementations, a digital twin model is built so as to be customizable to an individual. For example, the digital twin model can be built based on data from the general population and also based on data collected directly from a particular individual for which the digital twin model is custom tailored.Some embodiments of the digital twin model described herein comprise a calibrated, physicsbased computer model or a data-based computer model. A computer model, such as the digital twin model described herein, can be a mathematical representation of an object, process, or system. The mathematical representation may correlate inputs with outputs with either the laws of nature (e.g., physics-based model) or statistical correlation (e.g., data-based model). Examples of types of computer models include geometrical representations and representations of natural phenomena. The mathematical structures that support geometrical representations may be space-oriented curves and surfaces. The mathematical structures that support the representation of nature phenomena can be the equations of nature or statisticalcorrelations that provide a statistical image of the phenomena.

[0018] According to one example, the digital twin model may be representative of one or more physiological or anatomical characteristics of the individual and represent one or more assets or features of the individual and be indistinguishable from a physical counterpart for a specified application. For example, the digital twin model can represent a system including a pacemaker and a human heart. The digital twin model can further be subjected to a variety of stimuli, such as during a stress test. The stress test can include subjecting the digital twin model to the stimuli and observing a response to the stimuli for the system represented by the digital twin model. Based on the responses to the stress test, a diagnosis can be determined. Additionally, the digital twin model can be subjected to a variety of treatment options and corresponding responses to the treatment options can be observed. A prediction engine can receive the diagnosis, results from the stress test, and results from the treatment options and generate one or more predictions, such as predicting diseases at an early stage, a suggested optimization for therapies, configure medical devices in an optimal fashion, or generate medical advice, for example.Overview of Technical Problems, Technical Solutions, and Technical Improvements

[0019] The advent of digital technologies and advanced data analytics has revolutionized healthcare, offering unprecedented opportunities for personalized medicine and treatment optimization. However, the current healthcare landscape faces significant challenges in effectively utilizing the vast amounts of individual health data to create accurate, comprehensive models of patient physiology. Traditional approaches often rely on generalized population data or limited snapshots of an individual's health, failing to capture the complex, dynamic nature of human physiology and its response to various treatments. This gap between available data and actionable insights hinders the ability of healthcare providers to deliver truly personalized, optimized care tailored to each individual’s unique physiological characteristics and needs.

[0020] Despite the significant advancements in healthcare data collection and analysis, current systems face substantial technical challenges in effectively utilizing this wealth of information to create accurate, comprehensive models of individual patient physiology. Traditional approaches often rely on generalized population data or limited snapshots of an individual's health, failing to capture the complex, dynamic nature of human physiology andits response to various treatments. This gap between available data and actionable insights hinders the ability of healthcare providers to deliver truly personalized, optimized care tailored to each individual's unique physiological characteristics and needs.

[0021] One key technical problem lies in the integration and processing of diverse data sources. Healthcare data is often siloed across multiple systems, including electronic health records, wearable devices, implantable sensors, and genetic testing results. The lack of standardization and interoperability between these systems makes it challenging to aggregate and analyze data holistically. Additionally, the sheer volume and complexity of healthcare data pose significant computational challenges, requiring advanced algorithms and processing capabilities to extract meaningful insights in a timely manner.

[0022] Another critical technical issue is the difficulty in creating accurate, real-time digital representations of individual patients. Conventional modeling approaches lack the sophistication to account for the intricate interplay of various physiological systems and their responses to different stimuli or treatments. This limitation makes it challenging to simulate and predict individual patient outcomes accurately, particularly when considering multiple treatment options or complex medical conditions. As a result, healthcare providers are often forced to rely on generalized treatment protocols rather than truly personalized approaches, potentially leading to suboptimal patient outcomes and inefficient use of healthcare resources.

[0023] The technical challenges described above are addressed by the embodiments of the digital twin modeling system disclosed herein. In particular, some embodiments provided herein provide a comprehensive solution for integrating diverse healthcare data sources, creating accurate digital representations of individual patients, and leveraging these models to simulate and optimize personalized treatments. By employing advanced data aggregation techniques and sophisticated modeling algorithms, some of these embodiments overcomes the limitations of traditional approaches that rely on generalized population data or limited snapshots of an individual's health.

[0024] One key technical improvement offered by certain embodiments of this disclosure include computer functionality to generate a dynamic, multi-faceted digital twin model that accurately represents an individual's physiological and anatomical characteristics. For instance, and as described herein, this model may be built using a combination of physicsbased and data-driven approaches, thereby enabling it to capture complex interactions between various bodily systems and their responses to different stimuli or treatments. These innovative calibration and validation processes facilitate ensuring that the digital twin modelremains an accurate representation of the individual over time, adapting to new data inputs and changes in the patient's condition.

[0025] Furthermore, some embodiments include functionality for performing virtual stress tests and simulations of multiple treatment options on the digital twin model, which represents a significant technical advancement in personalized medicine. These technical features enable healthcare providers to explore and evaluate various treatment scenarios without subjecting the patient to unnecessary risks or invasive procedures. By optimizing treatment parameters based on these simulations, these embodiments enable the existence and utilization of truly individualized patient care plans that are tailored to each patient's unique physiological characteristics and needs. This not only improves patient outcomes but also enhances the efficiency of healthcare delivery by reducing the need for trial-and-error approaches in treatment selection.Additional Description of the Embodiments

[0026] FIG. 1 A is a block diagram that illustrates one example embodiment of a system 100 for determining an individual-specific treatment for a diagnosis based on a digital twin model. With regard to this example embodiment, system 100 includes a processor 112, a memory 114, a storage drive 116, a communication interface 118, a prediction engine 122, and one or more peripherals 124. Respective components of the system 100 can be communicatively coupled by one or more busses (not all shown for simplicity) or one or more wireless connections to facilitate computer communication, for example. The processor 112, memory 114, or storage drive 116 can be utilized to implement the prediction engine 122. The memory 114 can store one or more instructions and the processor 112 can execute one or more of the instructions stored on the memory 114 to perform one or more operations including acts, actions, or steps.

[0027] Continuing with this example embodiment, the prediction engine 122 can utilize any mathematical, statistical, machine-learning (ML), or artificial intelligence model that links inputs and outputs through one or more correlations and can be tuned based on individual data 144 (e.g., individual-specific data). The prediction engine 122 can be physics-based, data-based or a combination of both. Physics-based models, for example, can have equations that describe physics associated with or corresponding to the system, asset, and / or feature being modeled. Examples or physics-based models are finite element models,finite volume models, differential equation models. Examples of data-based models include multilayer perceptron (MLP), deep neural networks, convolutional neural networks (CNN), recurrent neural networks (RNN), long short-term memory (LSTM). Physics-based and data- based models can be combined with a techniques such as multi-fidelity neural networks or statistical methods. For example, physics-informed neural networks (PINNs) or other combination models can be implemented. A digital twin is a prediction engine which is calibrated and validated for a specific individual or population.

[0028] According to one example, the prediction engine 122 is configured to build a digital twin model 120 for an individual based on individual data 144 associated with the individual as an input. The digital twin model 120 is representative of one or more physiological or anatomical characteristics of the individual. Stated another way, the digital twin model 120 can be a model that describes a function of an individual as an in-silico representation of an asset associated with the individual. An example of an asset can be an organ, a medical device, a combination thereof, or a system within the human body. For example, the digital twin model 120 can include a medical device model representing a simulated medical device having configuration parameters and an organ model representing a simulated organ and the interaction therebetween.

[0029] Therefore, the digital twin model 120 can behave similarly to the asset being modelled from the real world. The description of the function can be performed with respect to inputs, such as stress conditions or virtual treatments, and with respect to outputs, such as forecasted, estimated, or predicted reactions by the asset or physiological responses. In some examples, the digital twin model 120 can describe more than one asset associated with the individual or different attributes or features of a given asset. For example, the digital twin model 120 can be representative of a system within the human body including a pacemaker and a human heart. In this example, the digital twin model 120 can include a first model representing the pacemaker and a second model representing the human heart while the digital twin model 120, as a whole, represents the system including the pacemaker and the human heart and how the system responds to different stimuli.

[0030] As an example, the individual data 144 associated with the individual can be data collected from one or more clinical records (e.g., electronic health records), an implanted sensor, a wearable sensor, a stationary device (e.g., a camera or infrared camera, motion sensor), or from genetic testing. Examples of individual data 144 include information from electronic health records (e.g., medical imaging data, laboratory results, histology results,etc.), demographic data, drug prescriptions, other sources of healthcare data, implantable devices, wearable devices (e.g., activity data, sleep data, heartrate data, VO2 max data, weight, etc.), remote individual monitoring systems, etc. The individual data represents health information for the individual collected from a plurality of data sources and the processor 112 can aggregate data from a plurality of data sources to provide the individual data 144 associated with the individual. The individual data 144 associated with the individual can be collected and stored in a database remote from the system 100, such as federated storage, to build a custom individual database 142 using the individual data 144.For example, the system 100 can include an application programming interface (API) to call a function for retrieving data from each database or service that contains corresponding data for the individual. The data can be pre-processed using machine learning (e.g., outlier treatment with random forest and K- means clustering for segregation) as well as other methods.

[0031] Further, or as an alternative, the individual data 144 can include data which relates to the genetics for the individual and data relating to environmental information associated with the individual. For example, family history can provide information related to underlying genetic encoding for the individual. Environmental information, on the other hand, such as diet or physical activity can also impact health status. Communicable diseases, such as CO VID, are an example of an environmental aspect which can impact a state of health for an individual over time or on a permanent basis. The environmental information or data is trackable using wearable technologies. Further, clinical data gathered from regular healthcare visits can be used to validate the wearables information. In this way, the custom individual database 142 can be utilized to build the digital twin model 120 in a manner specific to each individual. In some examples, the individual data and custom individual database 142 is accessed by the system at a time when the digital twin model 120 is being constructed but is not maintained and stored in persistent memory in the custom individual database 142 or other memory after the digital twin model 120 has been generated. For example, the custom individual database 142, comprehensive healthcare database 146, and the wearable database 148 can be remote from the system 100 and / or not included as part of the system 100.

[0032] The individual or individual population data will be collected from the different data sources or databases available to which the individual will grant access. The data aggregator serves as a link between the source database and the application that consumes theinformation (e.g., to avoid storing individual data elsewhere). For example, this can be the calibration of the digital twin. Examples of existing clinical records databases can include medical device databases, wearable databases, implantable individual monitoring databases, genetic data databases.

[0033] Additionally, the prediction engine 122 can build the digital twin model 120 for the individual based on data from a comprehensive healthcare database 146. The digital twin is a calibrated prediction engine. As an example, a prediction engine for a cardiac application counts with a set of input such as individual anatomy geometry, tissue conduction speed, fiber distribution, muscle stiffness, muscle contractility, blood density and viscosity, etc. While the prediction engine is not calibrated, is it merely a set of linked equations. The calibration process reads the individual’s data and calibrates the prediction engine to that specific individual data. As an example, the conduction velocity can be calibrated to match the electrocardiogram (EKG) of the individual. Once the prediction engine is calibrated, the digital twin, a set of inputs read from the individual, correlated with the equations in the model, that behaves like the real individual for the output variables (i.e., not for every single aspect, but for the chosen aspects). As the prediction engine calibration is automated, it can scan a population database to create a virtual population. Once a digital twin is created for an individual or a population, this can be used for the diagnostics or stress test. As an example, a digital twin of an individual’s heart can be subjected to exercise (e.g., increase of the heart rate, reduction of the afterload) to predict if that situation triggers an arrythmia.

[0034] The data from the comprehensive healthcare database 146 can provide the processor 112 with general information (e.g., probability distribution of population) pertaining to the digital twin model 120 to be built. The prediction engine 122 can determine what data should be pulled from what database. For example, the functions of the digital twin model 120 being built can determine the input data utilized to build the digital twin model 120. Inputs can be determined by the prediction engine 122 as variables from the data which are sensitive to the output, for example. In this way, the digital twin model 120 can be utilized to predict diseases at an early stage, optimize therapies, configure medical devices in an optimal fashion, generate medical advice, etc.

[0035] The building of the digital twin model 120 by the prediction engine 122 for the individual can include calibrating and validating the digital twin model 120 based on the individual data 144. The prediction engine 122 can utilize supervised machine learning to train the digital twin model 120, as implemented by the processor 112 and the memory 114.The digital twin model 120 can be stored on the storage drive 116 for runtime execution. Additionally, the digital twin model 120 can be subjected to stimuli which are compared to medical records for the individual and the digital twin model 120 can be tuned accordingly. In this way, calibration can be performed by testing possible inputs until solution values (e.g., validation) are matched. According to one example, the prediction engine 122 can perform data mining to extract and discover patterns or correlations between the digital twin model 120 and the comprehensive healthcare data or the individual data 144.Newly discovered correlations can be incorporated into digital twin model 120 builds. This data mining can facilitate early diagnoses or therapy optimization by enabling early detection of issues.

[0036] According to one example, the individual data 144 is stored in the first database 142 and the comprehensive healthcare data is stored in the second database 146. According to another example, both the individual data 144 and the comprehensive healthcare data are stored in the same database 142 or 146. The communication interface 118 can receive individual data from a medical device 132 (e.g., pacemaker) associated with the individual, a mobile device 134 (e.g., fitness tracker or smart watch) associated with the individual, a peripheral device 136 associated with the individual, a service that stores data associated with one or more user devices, etc. In any event, the communication interface 118 of the system 100 can enable communication between either database 142, 146 and the system 100.

[0037] The processor 112 can be programmed to determine a diagnosis based on a simulated application of one or more stress test to the digital twin model 120. For example, the simulated application of a given stress test enables the system 100 to analyze different scenarios without actually subjecting the individual to each stress condition, while providing simulated stress test results for evaluation. Examples of different types of stress tests include a temperature stress test, an activity stress test, a dehydration stress test, sodium intake, etc. Since subjecting an individual to a stress test is not always feasible due to health risks and costs, the digital twin can be utilized by the prediction engine 122 to execute procedures that could be risky in real life individuals, such as invasive diagnostic procedures, for example. The digital twin model can compare output values of the model with the medical knowledge, such as can be retrieved from one or more diagnostic databases (e.g., a knowledge base, clinical decision support system, or the like).

[0038] The processor 112 can determine an individual-specific treatment for the diagnosis based on a simulated application of two or more virtual treatment options to thedigital twin model 120. For example, the simulated application of a given virtual treatment option enables the system 100 to analyze different treatment options without actually subjecting the individual to the different treatment options, while providing simulated virtual treatment results for evaluation. Treatment options can include a given medical device (e.g., implantable device) and configurations thereof, therapy options (physical and / or pharmacological). Different treatment options thus can further include different device configurations and operating parameters and / or levels or types of therapies.

[0039] The determining of the individual-specific treatment for the diagnosis can include optimizing one or more treatment parameters associated with the two or more virtual treatment options. An optimization loop can be implemented based on different virtual treatment options and device configurations. For example, given a diagnosis, the system can suggest a therapy (e.g., the system 100 can indicate cardiac resynchronization therapy for an individual with low ejection fraction and a wide QRS). The system can also provide the optimized therapy configuration (e.g., as a prediction for optimized therapy or as an optimized treatment) based on the digital twin. After the prediction engine is calibrated for a single individual and the optional stress test was executed, the system can make automatic diagnosis based of medical knowledge. For this diagnosis, the system will provide casespecific measures such as heart contractility, ejection fraction, QRS duration, QT interval, total activation time, action potential duration 90%, longitudinal fiber stress, etc. Ejection fraction is the ratio (e.g., in percent points) of blood leaving the ventricle with every contraction. For example, if the system detects a reduced ejection fraction (less than 50%) the individual will be diagnosed with heart failure with reduced ejection fraction (HFrEF). For example, if the digital twin ECG has a negative deflection in lead V 1 and a QRS width of 120 milliseconds and mid-QRS notching, the individual will be diagnosed with left bundle branch block (LBBB). Once the diagnose is made, a therapy can be automatically recommended given the medical knowledge and the therapy parameters calibrated to return the metrics to healthy values. For example, if the individual is diagnosed with LBBB, the system will automatically assign cardiac resynchronization therapy (CRT). The optimization loop (similar to the calibration loop) will modify the therapy configuration that account for variables such as lead position, atrial-ventricular delay, ventricle- ventricle delay, etc. The optimization loop will test multiple configurations to minimize QRS duration, which means achieving inter-ventricular synchrony.

[0040] One example of a treatment parameter is a device parameter (e.g., configurationparameter) for a medical device 132. The processor 112 can further determine one or more device parameters and configure the medical device 132 using the communication interface 118 based on the one or more device parameters included in the determined individualspecific treatment. In this example, the digital twin model can further include a medical device model representing a simulated medical device having device parameters or configuration parameters. The processor 112 can modify the device parameters or configuration parameters of the medical device model in conjunction with subjecting the corresponding model (e.g., digital twin model), including the medical device model, to a stress and determine the individual-specific treatment based on the configuration parameters for the medical device model.

[0041] By configuring the medical device 132 based on the optimized treatment parameters from virtual treatment options, the digital twin model 120 provides the benefit or advantage of providing predictions or forecasts using group or individual epidemiological data to optimize the configuration of the medical device 132 for the individual. In this way, the medical device 132 can be virtually configured and assessed or simulated on the digital twin model 120 and optimized prior to actual configuration and implementation. According to one example, the communication interface 118 can provide an alert or configuration information to a mobile device 134 (e.g., including speakers, displays, etc.) to be displayed. According to another example, the peripherals 124 (e.g., including speakers, displays, etc.) of the system 100 is configured to provide alerts. In yet other examples, the system is configured to provide device parameters to configure operation of an actual device (e.g., an implantable device or other form of treatment device) that can be used to apply treatment to the individual.

[0042] According to one example, the digital twin model 120 can be a digital twin model of a human heart and simulate depolarization, deformation, blood pumping, etc. Using the individual’s database, the prediction engine 122 can be calibrated for a specific individual. Once the digital twin is obtained, it can be used to execute a stress test and / or execute an automatic diagnose. According to one use case example, this digital twin model can be utilized to predict the individual’s heart behavior under certain cardiomyopathies. For example, the digital twin model of the heart can be subjected to a regional muscular hypoxia (e.g., a heart infarct, often incorrectly called a “heart attack”) as a stress test. On the stress test, the digital twin model will add a non-contractile tissue region that is the effect of the partial muscular hypoxia. This could result in a drop in ejection fraction (e.g., output of theprediction engine 122 as stress test result) similar to an individual’s heart. Therefore, the digital twin model 120 can be used to predict if the resulting ejection fraction after the infarct is compatible with life. The digital twin can also mimic the heart depolarization pattern after the infarct, predicting the risk of ventricular fibrillation and risk of asystole after the infarct.

[0043] According to another use case example, the digital twin model 120 can be subjected to heart valve calcification parameters to determine an increase on afterload (e.g., outlet pressure for the ventricle) similarly to the individual’s heart, reduce aortic resistance, simulate exercise, etc. The digital twin mode of the heart can be programmed to predict the reduction on the afterload after the calcified valve is replaced by a transcatheter aortic valve implantation (TAVI). The digital twin model the heart further can simulate multiple TAVI models and select the optimal model for the individual based on different simulation conditions or inputs.

[0044] According to yet another use case example, the digital twin model the heart of an atrial fibrillation individual can be programmed to predict an outcome of a cardiac ablation and be used for surgical planning. In this way, the digital twin model 120 can be a guide for optimal ablation regions. Other use case examples include subjecting the digital twin model the heart to exercise while measuring (e.g., which can be invasive) a left atrial pressure which indicates an existence of heart failure with preserved ejection fraction (HFpEF) before symptoms appear, measuring mechanical and electrical dyssynchrony to predict severity of left bundle branch block (LBBB) in an individual, estimate a pressure drop, flow direction, and an amount of a sick or damaged heart valve and suggest an optimal surgery timing based on the digital twin model 120.

[0045] The digital twin model 120 can be subjected to a virtual stress test by controlling parameters for augmenting the heart rate and contractility and reducing the aortic resistance or mimicking a light exercise. The left atrial pressure (LAp) can be measured and compared with basal LAp information from one of the databases. Given the increase or lack of increase of LAp during the stress test, a first risk index can be obtained from the digital twin model 120. The digital twin model 120 can include a deep neural network (e.g., trained with retrospective data) to estimate the risk of heart failure with preserved ejection fraction (HFpEF), which can be represented in the digital twin model 120 based on the individual EHR history or other individual data 144 that has been acquired. The digital twin model 120 can be a combination of the weighted results from the physics-based model and the data- driven model. Given the severity of the case, such as ranked by the deviation of thequantities of interest from the baseline, the system 100 can further be programmed suggest one or a combination of the following treatment options or therapies: change in diet, exercise, or pharmacotherapy. The optimization of these or other therapies can be executed on the digital twin model 120.

[0046] According to one example, the digital twin model 120 can be a digital twin model of a human response to glucose consumption and exercise, etc. Using the individual’s database, the prediction engine 122 can be calibrated for a specific individual. Once the digital twin is obtained, it can be used to execute a stress test and / or execute an automatic diagnose. According to one use case example, this digital twin model 120 can be utilized to develop treatment plans for individuals with diabetes. For example, the digital twin will read nutrition data and activity data (e.g., collected by wearable devices) from the individual database. Using these inputs, the digital twin model 120 can predict an insulin requirement for an individual, a dose for a bolus, or set a glucose monitor threshold value.

[0047] According to one example, the digital twin model 120 can be a digital twin model a human brain and model vasculature of the brain, etc. The digital twin model 120 modeling cerebral vessels can be calibrated and used to predict a risk of rupture related to having an aneurism. The digital twin can be used to predict a fractional flow reserve of the cerebral vasculature and estimate a risk of ischemic brain stroke.

[0048] FIG. IB is a workflow diagram that illustrates an example embodiment of a computer-implemented method for calibrating a prediction engine, such as the prediction engine 122 of the system 100 for determining the individual-specific treatment for the diagnosis based on the digital twin model FIG. 1A. In FIG. IB, the prediction engine receives information (e.g., individual data 144 or general data pertaining to the population) from databases (e.g., custom individual database 142, comprehensive healthcare database 146, wearable database 148), which can be external to the system 100. For example, the prediction engine receives the information from the database through one or more of the peripherals 124. The prediction engine 122 also receives one or more stress inputs 222.The stress input can include an input for a simulated application of a stimuli to the digital twin model 120, such as representative of a stress test as described herein. The prediction engine 122 can receive information from previous iterations of calibration and / or optimization 154. At 152, a check is performed to determine if calibration is complete. For example, individual data 144 from the custom individual database 142 can be compared with digital twin measurements from the digital twin model 120 in response to the stressinput 222. If the difference is less than a predetermined threshold, the digital twin model 120 calibration is complete, and the digital twin model 120 is stored in the storage drive 116. If the difference is greater than the predetermined threshold, the digital twin model 120 calibration and / or optimization 154 continues (e.g., new set of inputs are provided to train the digital twin model 120), and the calibration and / or optimization 154 information is fed back to the prediction engine 122.

[0049] FIG. 2 is a workflow diagram that illustrates an example computer-implemented method for building and utilizing the digital twin model 120. For example, data from a database, such as the custom individual database 142 or the comprehensive healthcare database 146 can be fed to the prediction engine 122. The prediction engine 122 can receive data and generate the digital twin model 120 based on the received data. Based on the individual data from the custom individual database 142 or the comprehensive data from the comprehensive healthcare database 146, the prediction engine 122 can generate the digital twin model 120. The digital twin model 120 can be tuned, such as via calibration and validation at 214. Calibration of the digital twin model 120 can include adjusting prediction for the digital twin model 120 to improve both accuracy and confidence in those predictions. Calibration is the process via which the prediction engine takes the form of a digital twin, by adjusting the prediction engine parameters in order to mimic or simulate the individual physiology. The calibration loop suggests a set of input parameters for the prediction engine. After the prediction engine is executed, the outputs are compared with basal-state (e.g., basal can mean not exercising, on distress, under the influence of any temporary drugs, etc. It can be assumed that the basal-state is the average of the individual metrics, this is the average individual heart rate, the average arterial pressure, the average blood viscosity, etc.) individual data. If the difference between he calculated outputs and the individual data is below certain tolerance threshold, the input parameters are assumed as correct, and the prediction engine with the calibrated parameters now becomes the “digital twin”.

[0050] Validation can include a set of processes and activities designed to ensure that the digital twin model is performing as expected. To proceed with the validation, the calibrated prediction engine is tested under non-basal conditions read from the individual’s database and minimizes the difference between the digital twin model output and the corresponding individual data (e.g., taken as ground truth). As an example, for a heart-application, the comparison data is extracted while the individual is doing light exercise (e.g., walking), andthe calibrated prediction engine (or digital twin) is subjected to the same light exercise (by e.g., increasing the HR and reducing the arterial resistance). If the differences between the digital twin predictions and the individual data are under a certain threshold the model can be considered validated. If they are not, the model should be re-calibrated utilizing other set of basal conditions. In this way, the prediction engine 122 can generate and update the digital twin model 120.

[0051] Additionally, a stress test 222 can be performed on the digital twin model 120. As discussed herein, a stress test can include a simulated application of a stimuli to the digital twin model 120, such as a change in temperature, an exercise, providing food or water, etc. and observing the effects or results of the stimuli on the system simulated by the digital twin model 120. Results from the stress test 222 or simulations from the digital twin model 120 can be utilized to generate a diagnosis 224. For example, the prediction engine 122 can generate the diagnoses 224 based on the results from the stress test 222 and medical knowledge 228. An individual specific diagnosis 226 can be generated based on the digital twin model 120. Medical knowledge 228 can facilitate determination of the diagnosis 224.

[0052] The prediction engine 122 can determine two or more virtual treatment options 252 based on the diagnosis 224 and the information from the digital twin model 120 and apply these virtual treatment options 252 to the digital twin model 120. For example, the prediction engine 122 can apply a variety of virtual treatment options 252 to the digital twin model 120 and observe responses to the virtual treatments by the digital twin model 120. Repeated simulations can facilitate optimization 262 of the virtual treatment options 252. Based on simulations of the application of the virtual treatment options 252 to the digital twin model 120 and the optimization 262, an individual-specific treatment 292 can be generated. As discussed herein, the individual-specific treatment 292 can include physical configuration of a medical device.

[0053] FIG. 3 is a flow diagram that illustrates a computer-implemented method 300 for determining an individual-specific treatment for a diagnosis based on a digital twin model. The computer-implemented method 300 can include aggregating 302 data from a plurality of data sources to provide individual data associated with the individual, the individual data being representative of health characteristics or attributes for the individual. The digital twin model 120 can be representative of one or more physiological or anatomical characteristics of the individual. The computer-implemented method 300 can include calibrating 304 aprediction engine based on the individual data associated with the individual, validating 306 the prediction engine based on another part of the individual data to provide a corresponding model representative of one or more physiological characteristics of the individual, and subjecting 308 the corresponding model to stress using to determine an individual-specific treatment for a given diagnosis, in which the stress includes test parameters representative of a simulated application of two or more virtual treatment options to the corresponding model.

[0054] FIG. 4 is a workflow diagram 400 that illustrates an example computer- implemented method for building a digital twin model 120. For example, individual data 144 from the custom individual database 142 can be compared 402 with digital twin measurements from the digital twin model 120. If the comparison shows that the difference is greater than a predetermined threshold, a new set of inputs 404 can be provided to train the digital twin model 120. Examples of inputs for heart-specific calibration can include depolarization velocity, muscle stiffness, arterial resistance, blood viscosity, EKG shape, heart rate, etc. Based on these inputs, the prediction engine 122 can generate digital twin output measurements, such as QRS duration, QT duration, ejection fraction, aortic pressure, etc. Once the differences from the comparison 402 are within the predetermined threshold, the digital twin model 120 is generated.

[0055] As discussed herein, the digital twin model of the heart can be subjected to stress, including simulated exercise while measuring the left atrial pressure which indicates whether heart failure with preserved ejection fraction (HFpEF) is in an early stage. The advantage or benefit of using the digital twin model 120 is that current state of the art for early detection of asymptomatic HFpEF requires invasive measurement of the left atrial pressure during a stress test, which is impractical and possibly unsafe as a screening test. By utilizing the digital twin model 120, a digital off-line test provides an easy-access, repeatable risk index for HFpEF of individuals in question.

[0056] Examples of individual data 144 for building a digital twin model 120 associated with HFpEF can include healthcare data such as medical imaging, and lab work history data (e.g., lipidic profile, kidney function), wearable sensor data, such as weight, daily heart rate histogram, nightly average heart rate, heart rate variability, daily activity minutes, and V02max, demographic data, such as age, gender, height, and zip code, etc. The individual data 144 can be pre-processed. For example, a mesh (e.g., mathematical description of the geometry through tessellation useful for computational modeling) can be created, curves can be digitized, data can be parsed or organized in a structured database, etc.

[0057] The digital twin model 120 associated with HFpEF can be calibrated using the individual’s electronic health records (EHR) to execute for a first assessment of the individual’s left atrial pressure (LAp). The unknown inputs can be estimated using the probability distribution of the individual population with the data. The calibration can include iteratively modifying the prediction engine parameters until the output matches the individual measurements of the individual data. At that point, the prediction engine 122 becomes the digital twin model 120, as calibrated. An initial guess for prediction engine parameters can be made for electrophysiology (EP) calibration, and measurements obtained from the prediction engine are compared with the individual measurements through one or more metrics such as ID convolution, for example. If the metrics are under certain tolerance, the electrophysiology model (EM) is calibrated and mechanical calibration can occur. During mechanical calibration, EP parameters are kept unmodified. According to one example, the initial guess for the myocardium mechanical properties are taken from a histogram of oriented gradients (HOG) model and the initial guess for the arterial system parameters can be made from a known model. The prediction engine 122 is executed with the initial parameters and measurements obtained from the prediction engine are compared with the individual measurements through one or more metrics, such as ID convolution. If the metrics are under certain tolerance, the mechanical model is considered calibrated (e.g., ‘yes’ at 402). If the metrics are under certain tolerance, the process can be repeated (e.g., ‘no’ at 402), such as with a different estimation technique (e.g., gradient decent).

[0058] After the digital twin model 120 is calibrated, validation can be performed (e.g., calculation of uncertainty or precision of digital twin). Validation can include detecting another non-basal condition on the individual database (e.g., light exercise as walking) and modify the prediction engine parameters to reproduce the non-basal conditions, e.g., increasing heart rate and reducing systemic artery resistance. Again, the prediction engine 122 can be executed with the mentioned parameters and the measurements obtained from the prediction engine are compared with the non-basal individual measurements through one or more metrics, such as ID convolution. This metric provides the digital twin uncertainty or precision, which is a measurement of the accuracy. This accuracy is individual dependent, and changes with the quality of the individual data, the amount of the individual data and the drift out of average values of the individual physiology e.g., an individual with a low- resolution CT or low-resolution MRI, no lab work, no wearable data, and a genetic disorder will likely have a low precision or high uncertainty. Thus, the prediction engine 122 canhave the input parameters calibrated to the individual data. Therefore, the calibrated prediction engine 122 becomes the digital twin model 120 for the individual. From time to time, the digital twin model 120 will be updated, such as responsive to one or more new events or new data points are discovered. Also, or as an alternative, the calibration can be performed periodically to update the digital twin model over time to reflect current conditions and demographics of the individual.

[0059] The stress test is designed to assess if the individual has an underlying pathology, such as HFpEF, that may express during a physically stressful condition. The digital twin stress test enables variables to be measured that would otherwise be unsafe and / or unethical to measure on the physical individual. One such example of this type of measurement is the left atrial pressure which involves catheterism and light to moderate exercise. This test includes the increase of the heart rate and a reduction of the systemic resistance. Therefore, protocol for the stress test to the digital twin model to assess the existence of HFpEF can include providing the calibrated parameters from the prediction engine for the digital twin model, increasing the heart rate, reducing the arterial resistance to mimic light to moderate exercise, and measuring the left atrial pressure (LAp) during the virtual stress test. If there is a considerable increase in the LAp during the stress test (e.g., greater than a threshold amount), the individual is tagged as high-risk for HFpEF because the increase of LAp during exercise is an indication of an HFpEF individual.

[0060] In this way, the digital twin model 120 can be subjected to a stress test by augmenting the HR and contractility and reducing the aortic resistance, mimicking a light exercise. The LAp will be measured and compared with the basal LAp. Given the increase or not of LAp during the stress test, a first risk index can be obtained. A deep neural network (e.g., trained with retrospective data) can estimate the HFpEF risk given the individual HER history. The final index can be a combination of the weighted results from the physics-based method and the statistical method.

[0061] According to one example, given the severity of the case, ranked by the deviation of the quantities of interest from the baseline, the system may suggest one or a combination of the following: change in diet and / or exercise, atrial shunt therapy, pharmacotherapy. The optimization of these therapies can be executed on a simplified physics-based model of the human body and the results ratified by a high resolution three physics model of the heart.

[0062] In this regard, the prediction engine 122 can determine an individual-specific treatment for the HFpEF based on a simulated application of two or more virtual treatmentatrial shunt therapy variations to the digital twin model. For example, shunt diameter and shunt position can be varied as different virtual treatment options. The initial guess parameters for the shunt therapy can be extracted from the medical recommendations. The prediction engine 122 can calculate LAp during the virtual treatment by executing the prediction engine with the digital twin parameter plus the device therapy parameters. Select a new set of parameters for the shunt therapy given the selected optimization method, such as gradient decent, for example. If the change in the reduction of the LAp during the stress test is below certain predetermined threshold, optimization is considered complete and the shunt therapy is considered optimized in this way.

[0063] FIG. 5 is a workflow diagram that illustrates an example computer-implemented method 500 for updating the digital twin model. For example, the computer-implemented method 500 for updating the digital twin model is a data management decision tree indicative of how healthcare database data (e.g., custom individual data or individual data) is stored when building or updating the digital twin model. At 502, the prediction engine is initially calibrated. At 504, the custom individual database 142 is accessed and the corresponding individual data is read and temporarily stored for processing while the digital twin model 120 is built by the prediction engine 122. At 512, a check is performed to determine whether the calibration of the digital twin model 120 should be updated. Examples of when the digital twin model 120 will be updated include a new event, such as a new data point (e.g., a new medical imaging, such as an MRI, a new X-Ray, a new laboratory result associated with the individual, etc.), a model update (e.g., additional physics equations or updated equations), an on-demand request for a rebuild, a periodic rebuild command, etc. If there is not an update, the computer-implemented method 500 will loop and continuously check for new events. When a new event is found, the custom individual database 142 is accessed and the corresponding individual data and / or updated model data is read and temporarily stored for processing, if necessary, while the digital twin model 120 is rebuilt or updated by the prediction engine 122.Example Computing Environments

[0064] Having described various implementations, an example computing environment suitable for implementing embodiments of the disclosure is now described, including an example computing device. With reference to FIG. 6, an example computing device is providedand referred to generally as computing device 600. The computing device 600 is but one example of a suitable computing environment and is not intended to suggest any limitation as to the scope of use or functionality of embodiments of the disclosure, and nor should the computing device 600 be interpreted as having any dependency or requirement relating to any one or combination of components illustrated. Accordingly, computing device 600 comprises any type of computing device capable of performing aspects of the operations described in connection with FIGS. 1A, IB and 2, or methods 300 and 500, and workflow 400, in FIGS. 3- 5. By way of example and not limitation, computing device 600 may be embodied as a personal computer (PC), a laptop computer, a mobile device, a smartphone, a tablet computer, a smart watch, a wearable computer, a virtual-reality (VR) or augmented-reality (AR) device or headset, a smart display such as a smart television, a handheld scanner device, a workstation, any other suitable computer device, or any combination of these delineated devices.

[0065] Embodiments of the disclosure are described in the general context of computer code or machine-useable instructions, including computer-useable or computer-executable instructions, such as program modules, being executed by a computer or other machine such as a smartphone, a tablet PC, or other mobile device, server, or client device. Generally, program modules, including routines, programs, objects, components, data structures, and the like, refer to code that performs particular tasks or implements particular abstract data types. Embodiments of the disclosure are practiced in a variety of system configurations, including mobile devices, consumer electronics, general -purpose computers, more specialty computing devices, or the like. Embodiments of the disclosure can also be practiced in distributed computing environments where tasks are performed by remote -processing devices that are linked through a communications network. In a distributed computing environment, program modules may be located in both local and remote computer storage media including memory storage devices.

[0066] Some embodiments comprise an end-to-end software-based system that operates within system components described herein to operate computer hardware, which may include computerized equipment, to provide system functionality. At a low level, hardware processors generally execute instructions selected from a machine language (also referred to as machine code or native) instruction set for a given processor. The processor recognizes the native instructions and performs corresponding low-level functions related to, for example, logic, control, and memory operations. Low-level software written in machine code can provide more complex functionality to higher-level software. Accordingly, in some embodiments, computer-executable instructions include any software, including low-level software written in machine code, higher-level software such as application software, and any combination thereof. In this regard, the system components can manage resources and provide services for system functionality. Any other variations and combinations thereof are contemplated with the embodiments of the present disclosure.

[0067] With reference to FIG. 6, computing device 600 includes a bus 610 that directly or indirectly couples the following devices: memory 612, one or more processors 614, one or more presentation components 616, one or more input / output (VO) ports 618, one or more I / O components 620, and an illustrative power supply 622. In one example, bus 610 represents one or more buses (such as an address bus, data bus, or combination thereof). Although the various blocks of FIG. 6 are shown with lines for the sake of clarity, in reality, these blocks represent logical, not necessarily actual, components. For example, a presentation component includes a display device, such as an I / O component. Also, processors have memory. It is recognized that such is the nature of the art and reiterate that the diagram of FIG. 6 is merely illustrative of an exemplary computing device that can be used in connection with one or more embodiments of the present disclosure. Distinction is not made between such categories as “workstation,” “server,” “laptop,” or “handheld device,” as all are contemplated within the scope of FIG. 6 and with reference to “computing device.”

[0068] Computing device 600 typically includes a variety of computer-readable media. Computer-readable media can be any available media that can be accessed by computing device 600 and includes both volatile and nonvolatile, removable and non-removable media. By way of example, and not limitation, computer-readable media comprises computer storage media and communication media. Computer storage media includes both volatile and nonvolatile, removable and non-removable media implemented in any method or technology for storage of information such as computer-readable instructions, data structures, program modules, or other data. Computer storage media includes, but is not limited to, random-access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, CD-ROM, digital versatile disks (DVDs) or other optical disk storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to store the desired information and which can be accessed by computing device 600. Computer storage media does not comprise signals per se. Communication media typically embodies computer- readable instructions, data structures, program modules, or other data in a modulated data signalsuch as a carrier wave or other transport mechanism and includes any information delivery media. The term “modulated data signal” means a signal that has one or more of its characteristics set or changed in such a manner so as to encode information in the signal. By way of example, and not limitation, communication media includes wired media, such as a wired network or direct-wired connection, and wireless media, such as acoustic, radio frequency (RF), infrared, and other wireless media. Combinations of any of the above should also be included within the scope of computer-readable media.

[0069] Memory 612 includes computer storage media in the form of volatile and / or nonvolatile memory. In one example, the memory is removable, non-removable, or a combination thereof. Hardware devices include, for example, solid-state memory, hard drives, and optical- disc drives. Computing device 600 includes one or more processors 614 that read data from various entities such as memory 612 or I / O components 620. As used herein and in one example, the term processor or “a processer” refers to more than one computer processor. For example, the term processor (or “a processor”) refers to at least one processor, which may be a physical or virtual processor, such as a computer processor on a virtual machine. The term processor (or “a processor”) also may refer to a plurality of processors, each of which may be physical or virtual, such as a multiprocessor system, distributed processing or distributed computing architecture, cloud computing system, or parallel processing by more than a single processor. Further, various operations described herein as being executed or performed by a processor are performed by more than one processor.

[0070] Presentation component(s) 616 presents data indications to a user or other device. Presentation components include, for example, a display device, speaker, printing component, vibrating component, and the like.

[0071] The I / O ports 618 allow computing device 600 to be logically coupled to other devices, including I / O components 620, some of which are built-in. Illustrative components include a microphone, joystick, game pad, satellite dish, scanner, printer, or a wireless device. The VO components 620 can provide a natural user interface (NUI) that processes air gestures, voice, or other physiological inputs generated by a user. In some instances, inputs are transmitted to an appropriate network element for further processing. An NUI may implement any combination of speech recognition, touch and stylus recognition, facial recognition, biometric recognition, gesture recognition both on screen and adjacent to the screen, air gestures, head and eye tracking, and touch recognition associated with displays on the computing device 600. In one example, the computing device 600 is equipped withdepth cameras, such as stereoscopic camera systems, infrared camera systems, red-green-blue (RGB) camera systems, and combinations of these, for gesture detection and recognition. Additionally, the computing device 600 may be equipped with accelerometers or gyroscopes that enable detection of motion. The output of the accelerometers or gyroscopes may be provided to the display of the computing device 600 to render immersive augmented reality or virtual reality.

[0072] Some embodiments of computing device 600 include one or more radio(s) 624 (or similar wireless communication components). The radio transmits and receives radio or wireless communications. Example computing device 600 is a wireless terminal adapted to receive communications and media over various wireless networks. Computing device 600 may communicate via wireless protocols, such as code-division multiple access (“CDMA”), Global System for Mobile (“GSM”) communication, or time-division multiple access (“TDMA”), as well as others, to communicate with other devices. In one embodiment, the radio communication is a short-range connection, a long-range connection, or a combination of both a short-range and a long-range wireless telecommunications connection. In various embodiments, references to “short” and “long” types of connections do not refer to the spatial relation between two devices. Instead, in general references short range and long range as different categories, or types, of connections (for example, a primary connection and a secondary connection). A short-range connection includes, by way of example and not limitation, a Wi-Fi® connection to a device (for example, mobile hotspot) that provides access to a wireless communications network, such as a wireless local area network (WLAN) connection using the 802.11 protocol; a Bluetooth connection to another computing device is a second example of a short-range connection, or a near-field communication connection. A long-range connection may include a connection using, by way of example and not limitation, one or more of Code-Division Multiple Access (CDMA), General Packet Radio Service (GPRS), Global System for Mobile Communication (GSM), Time-Division Multiple Access (TDMA), and 802.16 protocols.Other Embodiments

[0073] In some embodiments, a system, such as the computerized system described in any of the embodiments above, comprise at least one computer processor, one or more computer storage media, such as a memory, storing computer-useable instructions that, when used by theat least one computer processor, cause the at least one computer processor to perform operations. The operations comprise receiving individual data associated with an individual at the communication interface from a custom individual database, the individual data representing health information for the individual collected from a plurality of data sources; storing the individual data within the memory; building a digital twin model for the individual based on individual data associated with the individual, wherein the digital twin model is representative of one or more physiological or anatomical characteristics of the individual; removing the individual data from the memory after the digital twin model is built; and determining an individual-specific treatment for a diagnosis based on a simulated application of two or more virtual treatment options to the digital twin model.

[0074] Advantageously, these and other embodiments, as described herein provide improved systems and methods for personalized healthcare modeling and treatment optimization. By building a digital twin model based on individual health data and using it to simulate treatment options, these embodiments enable more accurate and tailored treatment recommendations while maintaining data privacy through removal of raw individual data after model creation. This approach allows for comprehensive analysis of potential treatments without subjecting patients to unnecessary risks or invasive procedures, potentially improving patient outcomes and healthcare efficiency.

[0075] In any combination of the above embodiments, the system comprises a computer system.

[0076] In any combination of the above embodiments, the system comprises communications interface.

[0077] In any combination of the above embodiments, the digital twin model is built based on a data-based model or a physics-based model.

[0078] In any combination of the above embodiments, the processor builds a custom individual database based on the individual data associated with the individual, and wherein the processor collects the individual data from at least one of a clinical record, an implanted sensor, a wearable sensor, a stationary sensor, or genetic testing.

[0079] In any combination of the above embodiments, building the digital twin model for the individual includes calibrating and validating the digital twin model based on the individual data.

[0080] In any combination of the above embodiments, the processor is further programmed to determine the diagnosis based on a simulated application of a stress test to the digital twin model.

[0081] In any combination of the above embodiments, building the digital twin model is based on supervised or unsupervised machine learning.

[0082] In any combination of the above embodiments, the determining the individualspecific treatment for the diagnosis includes optimizing one or more treatment parameters associated with the two or more virtual treatment options.

[0083] In any combination of the above embodiments, the one or more treatment parameters includes a device parameter for a medical device.

[0084] In any combination of the above embodiments, the processor is programmed to configure the medical device based on the determined individual-specific treatment.

[0085] In any combination of the above embodiments, the processor is further programmed to build the digital twin model for the individual based on data from a comprehensive healthcare database.

[0086] In some embodiments, a computer-implemented method comprises aggregating data from a plurality of data sources to provide individual data associated with the individual, the individual data being representative of health characteristics or attributes for the individual; calibrating a prediction engine based on the individual data associated with the individual; validating the prediction engine based on another part of the individual data to provide a corresponding model representative of one or more physiological characteristics of the individual; and subjecting the corresponding model to stress using to determine an individualspecific treatment for a given diagnosis, in which the stress includes test parameters representative of a simulated application of two or more virtual treatment options to the corresponding model.

[0087] Advantageously, these and other embodiments, as described herein offer a robust approach to creating and utilizing personalized health models. By incorporating a calibration and validation process, these embodiments ensure the accuracy and reliability of the digital twin model. The ability to subject the model to simulated stress tests and treatment options allows for comprehensive evaluation of potential treatments without risk to the patient, potentially leading to more effective and personalized healthcare strategies.

[0088] In some embodiments, a system comprises a memory storing one or more instructions, and a processor executing one or more of the instructions stored on the memoryto perform: aggregating data from a plurality of data sources to provide individual data associated with the individual; building a digital twin model for an individual based on the individual data associated with the individual, wherein the digital twin model is representative of one or more physiological or anatomical characteristics of the individual; determining a diagnosis for the individual based on a simulated application of a stress test to the digital twin model; and determining an individual-specific treatment for the diagnosis based on a simulated application of two or more virtual treatment options to the digital twin model.

[0089] Advantageously, these and other embodiments, as described herein provide a comprehensive system for personalized healthcare modeling, diagnosis, and treatment planning. By integrating data aggregation, model building, simulated diagnostics, and treatment optimization into a single system, these embodiments offer a streamlined approach to personalized medicine. The ability to perform simulated stress tests and evaluate multiple treatment options on the digital twin model allows for more informed decision-making in healthcare, potentially improving diagnostic accuracy and treatment efficacy while minimizing patient risk.

[0090] In any combination of the above embodiments, the corresponding model is built based on a data-based model or a physics-based model.

[0091] In any combination of the above embodiments, the individual data associated with the individual is data collected from at least one of a clinical record, an implanted sensor, a wearable sensor, a stationary sensor, or genetic testing.

[0092] In any combination of the above embodiments, the corresponding model defines a digital twin model for the individual.

[0093] In any combination of the above embodiments, the digital twin model further includes a medical device model representing a simulated medical device having configuration parameters, the method further comprising: modifying the configuration parameters of the medical device model in conjunction with subjecting the corresponding model, including the medical device model, to the stress; and determining the individual-specific treatment based on the configuration parameters for the medical device model.

[0094] In some embodiments, a system comprises a memory storing one or more instructions, and a processor executing one or more of the instructions stored on the memory to perform: aggregating data from a plurality of data sources to provide individual data associated with the individual; building a digital twin model for an individual based on the individual data associated with the individual, wherein the digital twin model is representativeof one or more physiological or anatomical characteristics of the individual; determining a diagnosis for the individual based on a simulated application of a stress test to the digital twin model; and determining an individual-specific treatment for the diagnosis based on a simulated application of two or more virtual treatment options to the digital twin model.

[0095] Advantageously, these and other embodiments, as described herein provide a comprehensive system for personalized healthcare modeling, diagnosis, and treatment planning. By integrating data aggregation, model building, simulated diagnostics, and treatment optimization into a single system, these embodiments offer a streamlined approach to personalized medicine. The ability to perform simulated stress tests and evaluate multiple treatment options on the digital twin model allows for more informed decision-making in healthcare, potentially improving diagnostic accuracy and treatment efficacy while minimizing patient risk.

[0096] In any combination of the above embodiments, the digital twin model is built based on a data-based model or a physics-based model.

[0097] In any combination of the above embodiments, the individual data associated with the individual is data collected from at least one of a clinical record, an implanted sensor, a wearable sensor, a stationary sensor, or genetic testing.

[0098] In any combination of the above embodiments, building the digital twin model for the individual includes calibrating and validating the digital twin model based on the individual data.

[0099] In any combination of the above embodiments, building the digital twin model is based on supervised or unsupervised machine learning.Additional Structural and Functional Features of Embodiments of Technical Solutions

[0100] Having identified various components utilized herein, it should be understood that any number of components and arrangements may be employed to achieve the desired functionality within the scope of the present disclosure. For example, the components in the embodiments depicted in the figures are shown with lines for the sake of conceptual clarity. Other arrangements of these and other components may also be implemented. For example, although some components are depicted as single components, many of the elements described herein may be implemented as discrete or distributed components or in conjunction with other components, and in any suitable combination and location. Some elements may be omittedaltogether. Moreover, various functions described herein as being performed by one or more entities may be carried out by hardware, firmware, and / or software, as described below. For instance, various functions may be carried out by a processor executing instructions stored in memory. As such, other arrangements and elements (for example, machines, interfaces, functions, orders, and groupings of functions) can be used in addition to or instead of those shown.

[0101] Embodiments described in the paragraphs herein may be combined with one or more of the specifically described alternatives. In particular, an embodiment that is claimed may contain a reference, in the alternative, to more than one other embodiment. The embodiment that is claimed may specify a further limitation of the subject matter claimed.

[0102] For purposes of this disclosure, the word “including” has the same broad meaning as the word “comprising,” and the word “accessing” comprises “receiving,” “referencing,” or “retrieving.” Furthermore, the word “communicating” has the same broad meaning as the word “receiving” or “transmitting” facilitated by software or hardware-based buses, receivers, or transmitters using communication media described herein. In addition, words such as “a” and “an,” unless otherwise indicated to the contrary, include the plural as well as the singular. Thus, for example, the constraint of “a feature” is satisfied where one or more features are present. Also, the term “or” includes the conjunctive, the disjunctive, and both (for example, “a or b” includes either a or b, as well as a and b).

[0103] As used herein, the terms “application” or “app” may be employed interchangeably to refer to any software-based program, package, or product that is executable via one or more (physical or virtual) computing machines or devices. An application may be any set of software products that, when executed, provide an end user one or more computational and / or data services. In some embodiments, an application may refer to a set of applications that may be executed together to provide the one or more computational and / or data services. The applications included in a set of applications may be executed serially, in parallel, or any combination thereof. The execution of multiple applications (comprising a single application) may be interleaved. For example, an application may include a first application and a second application. An execution of the application may include the serial execution of the first and second application or a parallel execution of the first and second applications. In other embodiments, the execution of the first and second application may be interleaved.

[0104] For purposes of a detailed discussion above, embodiments of the present disclosure are described with reference to a computing device or a distributed computing environment;however, the computing device and distributed computing environment described herein are non-limiting examples. Moreover, the terms computer system and computing system may be used interchangeably herein, such that a computer system is not limited to a single computing device, and nor does a computing system require a plurality of computing devices. Rather, various aspects of the embodiments of this disclosure may be carried out on a single computing device or a plurality of computing devices, as described herein. Additionally, components can be configured for performing novel aspects of embodiments, where the term “configured for” can refer to “programmed to” perform particular tasks or implement particular abstract data types using code. Further, while embodiments of the present disclosure may generally refer to the technical solution environment and the schematics described herein, it is understood that the techniques described may be extended to other implementation contexts.

[0105] Further, unless specified otherwise, “first”, “second”, or the like are not intended to imply a temporal aspect, a spatial aspect, an ordering, etc. Rather, such terms are merely used as identifiers, names, etc. for features, elements, items, etc. For example, a first channel and a second channel generally correspond to channel A and channel B or two different or two identical channels or the same channel. Additionally, “comprising”, “comprises”, “including”, “includes”, or the like generally means comprising or including, but not limited to.

[0106] It should be understood that various aspects disclosed herein can be combined in different combinations than the combinations specifically presented in the description and accompanying drawings. It should also be understood that, depending on the example, certain acts or events of any of the processes or methods described herein can be performed in a different sequence, can be added, merged, or left out altogether (e.g., all described acts or events are not necessary to conduct the techniques). In addition, while certain aspects of this disclosure are described as being performed by a single module or unit for purposes of clarity, it should be understood that the techniques of this disclosure can be performed by a combination of units or modules associated with, for example, a medical device.

[0107] In one or more examples, the described techniques can be implemented in hardware, software, firmware, or any combination thereof., such as described herein. If implemented in software, the functions can be stored as one or more instructions or code on a computer-readable medium and executed by a hardware-based processing unit. Computer- readable media can include non-transitory computer- readable media, which corresponds to a tangible medium such as data storage media (e.g., RAM, ROM, EEPROM, flash memory, orany other medium that can be used to store desired program code in the form of instructions or data structures and that can be accessed by a computer).

[0108] Instructions can be executed by one or more processors, such as one or more digital signal processors (DSPs), general purpose microprocessors, application specific integrated circuits (ASICs), field programmable logic arrays (FPGAs), or other equivalent integrated or discrete logic circuitry. Accordingly, the term “processor” as used herein can refer to any of the foregoing structure or any other physical structure suitable for implementation of the described techniques. Also, the techniques could be fully implemented in one or more circuits or logic elements.

[0109] A “processor”, as used herein, processes signals and performs general computing and arithmetic functions. Signals processed by the processor can include digital signals, data signals, computer instructions, processor instructions, messages, a bit, a bit stream, or other means that can be received, transmitted, or detected. Generally, the processor can be a variety of various processors including multiple single and multicore processors and coprocessors and other multiple single and multicore processor and co-processor architectures. The processor can include various modules to execute various functions.

[0110] A “memory”, as used herein, can include volatile memory or non-volatile memory. Non-volatile memory can include, for example, ROM (read only memory), PROM (programmable read only memory), EPROM (erasable PROM), and EEPROM (electrically erasable PROM). Volatile memory can include, for example, RAM (random access memory), synchronous RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), and direct RAM bus RAM (DRRAM). The memory can store an operating system that controls or allocates resources of a computing device.

[0111] A “disk” or “drive”, as used herein, can be a magnetic disk drive, a solid-state disk drive, a floppy disk drive, a tape drive, a Zip drive, a flash memory card, or a memory stick. Furthermore, the disk can be a CD-ROM (compact disk ROM), a CD recordable drive (CD-R drive), a CD rewritable drive (CD-RW drive), or a digital video ROM drive (DVD-ROM). The disk can store an operating system that controls or allocates resources of a computing device.

[0112] A “bus”, as used herein, refers to an interconnected architecture that is operably connected to other computer components inside a computer or between computers. The bus can transfer data between the computer components. The bus can be a memory bus, amemory controller, a peripheral bus, an external bus, a crossbar switch, or a local bus, among others.

[0113] A "database", as used herein, can refer to a table, a set of tables, and a set of data stores (e.g., disks) or methods for accessing or manipulating those data stores.

[0114] An "operable connection", or a connection by which entities are "operably connected", is one in which signals, physical communications, or logical communications can be sent or received. An operable connection can include a wireless interface, a physical interface, a data interface, or an electrical interface.

[0115] A "computer communication", as used herein, refers to a communication between two or more computing devices (e.g., computer, personal digital assistant, cellular telephone, network device) and can be, for example, a network transfer, a file transfer, an applet transfer, an email, a hypertext transfer protocol (HTTP) transfer, etc. A computer communication can occur across, for example, a wireless system (e.g., IEEE 802.11), an Ethernet system (e.g., IEEE 802.3), a token ring system (e.g., IEEE 802.5), a local area network (LAN), a wide area network (WAN), a point-to-point system, a circuit switching system, a packet switching system, among others.

[0116] A “mobile device”, as used herein, can be a computing device typically having a display screen with a user input (e.g., touch, keyboard) and a processor for computing.Mobile devices include handheld devices, portable electronic devices, smart phones, laptops, tablets, smart watches, wearable devices, and e-readers, etc.

[0117] It will be appreciated that various of the above-disclosed and other features and functions, or alternatives or varieties thereof, can be desirably combined into many other different systems or applications. Also, that various presently unforeseen or unanticipated alternatives, modifications, variations, or improvements therein can be subsequently made by those skilled in the art which are also intended to be encompassed by the following claims.

[0118] Many different arrangements of the various components depicted, as well as components not shown, are possible without departing from the scope of the claims below. Embodiments of the present disclosure have been described with the intent to be illustrative rather than restrictive. Alternative embodiments will become apparent to readers of this disclosure after and because of reading it. Alternative means of implementing the aforementioned can be completed without departing from the scope of the claims below. Certain features and subcombinations are of utility and may be employed without reference to other features and subcombinations and are contemplated within the scope of the claims.

Claims

WHAT IS CLAIMED IS:

1. A system (100), comprising: a communication interface (118); a memory (114) storing one or more instructions; and at least one processor (112) executing one or more of the instructions stored on the memory to perform: receiving (118) individual data associated with an individual at the communication interface from a custom individual database, the individual data representing health information for the individual collected from a plurality of data sources; storing the individual data within the memory; building a digital twin model for the individual based on individual data associated with the individual, wherein the digital twin model is representative of one or more physiological or anatomical characteristics of the individual; removing the individual data from the memory after the digital twin model is built; and determining an individual-specific treatment for a diagnosis based on a simulated application of two or more virtual treatment options to the digital twin model.

2. The system of claim 1, wherein the digital twin model is built based on a data-based model or a physics-based model.

3. The system according to claim 1 or 2, wherein the processor builds a custom individual database based on the individual data associated with the individual, and wherein the processor collects the individual data from at least one of a clinical record, an implanted sensor, a wearable sensor, a stationary sensor, or genetic testing.

4. The system according to claim 1 or 2, wherein building the digital twin model for the individual includes calibrating and validating the digital twin model based on the individual data.

5. The system according to any preceding claim, wherein the processor is further programmed to determine the diagnosis based on a simulated application of a stress test to the digital twin model.

6. The system according to claim 1, wherein building the digital twin model is based on supervised or unsupervised machine learning.

7. The system according to any preceding claim, wherein the determining the individual-specific treatment for the diagnosis includes optimizing one or more treatment parameters associated with the two or more virtual treatment options.

8. The system of claim 7, wherein the one or more treatment parameters includes a device parameter for a medical device.

9. The system of claim 8, wherein the processor is programmed to configure the medical device based on the determined individual-specific treatment.

10. The system of claim 1, wherein the processor is further programmed to build the digital twin model for the individual based on data from a comprehensive healthcare database.

11. A computer-implemented method, comprising: aggregating (302) data from a plurality of data sources to provide individual data associated with the individual, the individual data being representative of health characteristics or attributes for the individual; calibrating (304) a prediction engine based on the individual data associated with the individual; validating (306_ the prediction engine based on another part of the individual data to provide a corresponding model representative of one or more physiological characteristics of the individual; and subjecting (308) the corresponding model to stress using to determine an individual- specific treatment for a given diagnosis, in which the stress includes test parameters representative of a simulated application of two or more virtual treatment options to the corresponding model.

12. The computer-implemented method of claim 11 , wherein the corresponding model is built based on a data-based model or a physics-based model.

13. The computer-implemented method of claim 11, wherein the individual data associated with the individual is data collected from at least one of a clinical record, an implanted sensor, a wearable sensor, a stationary sensor, or genetic testing.

14. The computer- implemented method of claim 11, wherein the corresponding model defines a digital twin model for the individual.

15. The computer-implemented method of claim 14, wherein the digital twin model further includes a medical device model representing a simulated medical device having configuration parameters, the method further comprising: modifying the configuration parameters of the medical device model in conjunction with subjecting the corresponding model, including the medical device model, to the stress; and determining the individualspecific treatment based on the configuration parameters for the medical device model.

Images