System for creating a digital twin of an individual, a computer-implemented method for same, a computer program product and a computer-readable storage medium for same

Abstract

The invention relates to a system 15 for creating a digital twin 22 of an individual 16 with the aid of data, comprising a data evaluation module 17 which is connected for data exchange to at least one data storage medium 19, wherein the data evaluation module is designed to perform the following steps during creation of the digital twin 22: providing a template 20 for a digital twin of a defined individual; specifying a data completion criterion 21 for the digital twin of the defined individual; including data of the defined individual in the digital twin; analysing the included data of the defined individual in the digital twin; comparing the analysed data with the specified data completion criterion and specifying missing data inputs for the digital twin on the basis of the data completion criterion; selecting data from statistical medical data of a plurality of individuals 23 from the data storage medium, wherein the statistical medical data comprises at least data of deceased individuals; filling the missing data inputs in the digital twin of the defined individual with the selected data. The invention also relates to a computer-implemented method.

Description

[0001] System for creating a digital twin of an individual, a computer-implemented method therefor and a computer program product and a computer-readable storage medium therefor

[0002] The present invention relates to a system for creating a digital twin of an individual according to patent claim 1, a computer-implemented method according to patent claim 19, a computer program product according to patent claim 22 and a computer-readable storage medium according to patent claim 23.

[0003] Technological background

[0004] Attempts are already being made to transfer the concept of the digital twin, already common in industry, to human applications. The difficulty lies in the fact that, unlike technical systems, every person is unique, both physically and in their response to therapies and medications. Furthermore, the available data on a single person is too sparse to generate any relevant added value, since – primarily for cost and ethical reasons – no systematic medical data collection is carried out on living people (i.e., no experiments or risky examinations on humans without medical necessity).

[0005] In today's clinical trials on living humans, the results are only approximate, as the study group always represents a single person with specific characteristics. All clinical evidence in medicine is based on statistical frequency, as every person is different. Clinical trials and animal experiments are increasingly being replaced by so-called in silico trials (see FDA decision from 2022).

[0006] WO 2022 / 040536 A1 is known from the prior art. A method for processing a medical claim, a system, and a computer program product therefor are disclosed, the method comprising the following steps: receiving a medical claim associated with a patient to create a digital twin of the patient; mathematically analyzing whether the medical claim matches the digital twin; and outputting an indication of possible non-fulfillment if the medical claim does not match the digital twin. The mathematical analysis checks whether the medical claim matches the digital twin and includes determining a degree to which the medical service matches the mathematical model and outputting a score indicating this degree.The mathematical analysis of whether the medical claim conforms to the digital twin may include determining whether a medical service associated with the medical claim was medically indicated based on the digital twin and / or whether a medical service associated with the medical claim conforms to medical norms related to a patient health condition.

[0007] The disadvantage of this known solution is that it only verifies the veracity of a medical claim. This is done using a mathematical comparison between a digital twin and a real patient.

[0008] EP 3 646 331 A1 is known from the prior art. It discloses a device comprising a processor and a memory. The example processor is intended to configure the memory according to a digital twin of a first patient. The digital twin of the example patient is intended to contain a data structure created from a combination of medical patient data, image data, genetic information, and historical information, wherein the combination is extracted from one or more information systems and arranged in the data structure to form a digital representation of the first patient. The digital twin of the example patient is intended to be configured for query and simulation by the processor.The digital twin of the example patient should be capable of being combined with one or more rules in order to generate, with the aid of the processor, a recommendation for a health outcome of the patient based on the modeling of the patient's digital twin according to the instructions of the one or more rules. WO 2019 / 005185 A1 is known from the prior art. A device is disclosed with a digital twin of a patient and with a data structure created from a combination of electronic medical patient data and historical information, wherein the combination is extracted from information systems and arranged in the data structure to form a digital representation of the patient. The example digital twin of the patient is configured for queries and simulations.The digital twin of the sample patient is used to obtain feedback about the patient after a procedure has been performed on the patient. The digital twin of the sample patient is intended to incorporate the feedback into the digital twin of the patient once the elements of the procedure have been completed. The digital twin of the sample patient is intended to process the recorded feedback to generate and output a recommendation for the patient's further treatment based on the digital representation of the patient, including the recorded feedback regarding the performed procedure.

[0009] The disadvantage of the two known solutions mentioned above is that missing data is not added to the digital twin, thus resulting in an incomplete digital twin.

[0010] US 2022 / 013199 A1 is known from the prior art. A computer-implemented method and system for managing digital twins are disclosed, the method comprising: collecting data corresponding to one or more body tissues; generating one or more digital twin versions of the one or more body tissues; identifying a most suitable digital twin version of the one or more digital twin versions based on a health and compatibility of the one or more digital twin versions; and distributing the most suitable digital twin version.

[0011] The disadvantage of this well-known solution is that it only creates models of digital twins. One of the created models can then be used to address a medical question.

[0012] EP 3 809 420 A1 is known from the prior art. It discloses a system, a method, and a computer program product.The system comprises a first processing arrangement configured in at least a first mode and performing the following steps: retrieving a digital model of at least a portion of a patient's anatomy, simulating a plurality of physical states of the at least one portion of the anatomy based on adjusting a set of one or more input parameters of the model, and developing the digital model based on each set of adjusted parameters; deriving, for each simulated physical state, one or more physiological or anatomical parameters associated with the simulated at least one portion of the anatomy, and storing the derived parameters in a data storage arrangement together with the set of adjusted input parameters corresponding to the simulated physical state, thereby constructing a reference data set separate from the digital model.There is a further processing arrangement which is identical to the first processing arrangement and which is arranged in at least a second mode to: receive a set of one or more measured parameters relating to the at least one part of the patient's anatomy; associate the measured patient parameters with one of the sets of model input parameters stored in the reference data set; and retrieve the corresponding simulated physiological or anatomical parameters from the reference data set; and generate the output based on the one or more retrieved physiological or anatomical parameters.

[0013] The disadvantage of this known solution is that it only deals with the optimization of the storage space of the data in the processing for a digital twin.

[0014] US 2019 / 198169 A1 is known from the prior art. A device for selecting a medical treatment is provided, comprising a user interface for receiving health information for a patient, wherein the health information includes symptoms of a condition. The example system additionally includes a data analyzer for accessing historical patient information stored in a database for previous patients and for determining a condition based on comparing historical patient information stored in a database for previous patients with the health information for the patient.The example device also includes a machine learning engine to recommend at least one treatment plan to be presented by the user interface, including: a data analysis algorithm server to determine success rates of the at least one treatment plan for the condition, wherein the success rates are based on the health information for the patient and historical health information; and a model generator to generate a patient model, wherein the patient model predicts effects of the at least one treatment plan and symptoms of the condition for a duration of the treatment plan based on the success rates. The device additionally includes a communication interface to facilitate scheduling an appointment with a clinician after the patient selects the treatment plan via the user interface.

[0015] The disadvantage of this known solution is that the data from the database for previous patients is not reused in the patient model, but rather serves only as the basis for a treatment plan for the individual patient. A device and method are provided that are intended solely to improve communication and interaction between patients and healthcare providers in order to optimize patient care and increase efficiency in the healthcare system.

[0016] US 2022 / 375621 A1 is known from the prior art. It discloses a platform for digital patient twins, comprising a clinical subsystem, a patient subsystem, and a performance tracking system. The clinical subsystem generates physiotherapy exercises for a patient. The patient subsystem instructs the patient to perform the exercises generated by the clinical patient subsystem. The performance tracking system captures measurement data, such as videos, of the patient performing at least one physiotherapy exercise prescribed by the clinical subsystem. The output of the performance tracking system is fed into a patient digital twin technology, which updates a digital twin of the patient.

[0017] The disadvantage of this known solution is that it provides a method and system for creating and using a digital twin to improve the efficiency and accuracy of monitoring, controlling, and optimizing physical systems. It only captures continuous changes in the physical system to identify optimization opportunities and improve the physical system's performance.

[0018] WO 2022 / 067189 A1 is known from the prior art. Systems, methods, and computer program products are provided for determining one or more biomarkers and the health status of a target patient. In various embodiments, a method is provided in which a plurality of health data of the target patient and a plurality of first-order features determined from the plurality of health data of the target patient are received as input to a pre-trained artificial neural network. The plurality of health data is derived from a plurality of modalities. A plurality of latent variables based on the plurality of health data and a plurality of first-order features are received by an intermediate layer of the pre-trained artificial neural network. The plurality of latent variables are provided to a pre-trained learning system.The pre-trained learning system is trained to receive the multitude of latent variables as input and output one or more biomarkers and the health status of the target patient.

[0019] The disadvantage of this known solution is that it has only developed a system that assesses individuals' cognitive functions and creates personalized treatment plans. This is supposedly made possible by the use of AI algorithms that analyze large amounts of data to enable precise and individualized diagnoses and therapies.

[0020] Description of the invention

[0021] One object of the invention is to avoid at least one of the disadvantages of the prior art. The aim is to create a system that creates an improved digital twin of an individual so that this can be used for clinical studies instead of living beings, thereby at least significantly reducing the number of risky clinical studies on living beings. Furthermore, a computer-implemented method, a computer program product, and a computer-readable storage medium are to be created that contribute to a significant reduction in clinical studies on living beings.

[0022] This object is achieved by the features of the independent patent claims. Advantageous further developments are set out in the figures and in the dependent patent claims. A system according to the invention for creating a digital twin of an individual using data comprises a data evaluation module with at least one computing unit, which is connected to at least one data storage medium for data exchange, wherein the data evaluation module is designed to carry out at least the following steps when creating the digital twin: a) providing a template for a digital twin of a defined individual; b) specifying at least one data completion criterion for the digital twin of the defined individual; c) recording data of the defined individual in the digital twin; d) analyzing the recorded data of the defined individual in the digital twin;e) Comparing the analyzed data with the defined data completion criterion and determining missing data entries for the digital twin based on the data completion criterion; f) Selecting data from statistical medical data of multiple individuals from the at least one data storage medium, wherein the statistical medical data includes at least data from deceased individuals; g) Supplementing the missing data entries in the digital twin of the defined individual with the selected data.

[0023] This system massively replaces or reduces the number of clinical trials on living beings, as the system creates a digital twin with sufficient data for clinical trials. These digital twins can be used for clinical trials instead of living beings, so that risky clinical trials on living beings can at least be significantly reduced. To create the digital twin of a specific person, for example, existing data on that person is first used (e.g. from medical records but also other information such as activity, place of residence, etc.). To fill data gaps, additional statistical medical data from multiple individuals is then used. Missing medical data is replaced with existing, statistically relevant data from individuals that are as similar as possible.Statistical medical data is therefore medical data from several individuals which, statistically speaking, match the defined individual. This enables the testing of new therapies on the digital twin as well as a parallel analysis on many instances of the same digital twin. The statistical medical data originates from many different individuals from data stores in different hospitals, countries or continents. The great advantage of using statistical medical data from several different individuals is that one does not necessarily aim for an "exact" correspondence of the data for the digital twin, but merely a sufficiently good similarity. This can largely replace clinical studies on living organisms. Compared to the existing gold standard for clinical treatments: There, medical treatments are always tested via clinical studies on completely different (i.e.This is done on (quite dissimilar) living beings that exhibit similar symptoms. A digital twin can be created that can be used to predict different treatment concepts before these treatment concepts are then applied to the real individual.

[0024] Data from deceased individuals includes, on the one hand, autopsy data and, on the other hand, historical patient data or the patient file of a deceased person, or historical data sets that reflect the medical history of a deceased person and all related data of the deceased individual. Furthermore, this data can include the cause of death of the individual, which may, for example, be related to years of medication use. Being able to use this medical data from deceased individuals as statistical medical data is revolutionizing the use of data for clinical studies. By utilizing the vast amount of data from deceased individuals, numerous clinical studies can be conducted on at least one or more digital twins of individual individuals with results that are at least equivalent to those obtained on living beings.

[0025] Thus, no digital "mixed twin" is created, i.e., a digital image consisting of a mixture of different individuals. The digital twin created here is individually based on a defined individual. The fact that missing data for this defined individual is supplemented with data from individuals as similar as possible (according to the data completion criterion or an evaluation algorithm) does not necessarily make it a mixed twin. Thus, individual digital twins can be created if the procedure described here is performed with one individual at a time. The set of individual digital twins then results in a sufficient and significant amount of data for clinical studies.

[0026] Preferably, at least steps b) to g) of the present method are performed multiple times to improve the data quality of the digital twin, or to create many individual digital twins of different individuals. The supplemented data and the selected data are stored together in a structured manner on the data storage medium. The set of individual digital twins then results in an improved and significant data set for clinical studies.

[0027] By using this system, for example, many studies can be conducted on digital twins of humans with results at least equivalent to those of current clinical trials on living humans. One or more digital twins can be used for clinical trials instead of real humans. A digital twin of a human does not necessarily have to be identical to a human, but can also contain statistically relevant data, and thus also statistically relevant medical data from other humans.

[0028] In particular, before selecting data from the statistical medical data of several individuals in step f), this data is verified in the data evaluation module. This ensures that only verified data is used in the digital twin. Some clinical studies require verified data, so these clinical studies can also be conducted with the generated digital twin. For example, one verification criterion could be to exclude the data of an individual with a specific disease.

[0029] A "digital twin" is understood here as a digital representation of an individual, including at least one identifying feature, e.g., date of birth, gender, etc., and at least one information element describing at least one physical characteristic of the individual, such as tumor tissue, blood, imaging data, etc. An individual in this case is a living body or a deceased body of a human or an animal with all its medically and physically detectable parameters.

[0030] The system user defines a data completion criterion. The at least one data completion criterion can be understood as restrictions of the general target definition, e.g., to specific groups of people (age, gender, number, etc.). For example, the data completion criterion requires that comparable statistical medical data from a minimum number of individuals must be consistently available, e.g., that the selected data from the statistical medical data include at least 100 individuals who have type A diabetes. The missing data entries (e.g., certain blood values) in the digital twin of the defined individual are supplemented only with the statistical data from at least 100 individuals.In principle, in the system disclosed here, not all data from the data storage medium should be used to complete the digital twin, but only those that meet the data completion criteria. Thus, when creating each digital twin in the system, only a fraction of the data available in the data storage medium is used—namely, the relevant data for the digital twin of the defined individual.

[0031] In addition to incorporating data from the defined individual into the digital twin, the individual's medical data is also recorded. This allows the existing medical data of the defined individual to be directly used in the digital twin.

[0032] In particular, the order of the aforementioned steps a) to f) is determined so that the selection of the data from the statistical medical data of the plurality of individuals from the at least one data storage medium can be carried out reproducibly.

[0033] The data completion criterion preferably includes at least one medical question. This question is to be answered based on data. The at least one medical question thus has a significant influence on the selection of statistical medical data of the multiple individuals from the at least one data storage medium. For example, the medical question can be formulated such that the individuals must not have any pre-existing lung diseases, such as COVID-19 or whooping cough. Such data from individuals is excluded from the system. Excluding data from individuals with certain pre-existing conditions accelerates the data selection, and the digital twin is supplemented only with relevant data.

[0034] Preferably, an interface is provided through which the data of the digital twin can be made available to a user. The user can be, in addition to a person, an artificial intelligence or a control device, for example, of a medical device. The interface enables individual further processing of the data in the digital twin by the user. In particular, an input device is provided so that the user can enter additional data or, alternatively or additionally, transmit at least one data completion criterion to the data evaluation module.

[0035] Alternatively or additionally, the data of the digital twin can be presented to a user on a display device. The display device is connected to the system and comprises, in particular, a graphics module for graphically displaying the digital twin. The user can directly access the data in the digital twin and process it further as desired. The user also has the option of verifying the data in the digital twin of the defined individual. The user can also perform further actions or make corrections to at least one data completion criterion in order to obtain the desired statistical medical data with the desired relevance. In particular, the display device comprises a display, preferably a touchscreen, so that the selected statistical medical data is available to the user in a user-friendly manner. A touchscreen can be used for data input and data output.The display can be connected to the system. For example, the display device is a portable device, such as a smartphone or tablet. Multiple display devices can be present, so that the digital twin data is available to multiple users simultaneously. The user can also perform further actions or easily make corrections to the data, for example, via the touchscreen and the graphics module in at least one data completion criterion.

[0036] In particular, the data from the digital twin can be presented to a user in an animated manner on a display device. An animated presentation of the data in the digital twin enables the system user to cognitively capture and process the data in the digital twin.

[0037] Preferably, a control device is provided, which causes the interface to output the selected data. This allows the control device to output the selected data automatically, so that no additional steps are required by the user. This simplifies further processing of the data in the digital twin for clinical studies.

[0038] Alternatively or additionally, the control device causes the display device to output at least the selected data. The selected data can be displayed directly to the user, allowing the user to verify the selected data for the digital twin. This ensures that the selected data for the digital twin of the defined individual is verified by a user before further processing.

[0039] Alternatively or additionally, the statistical medical data at least comprise data from living individuals. These living individuals have medical data that statistically match the digital twin of the defined individual. For example, the medical data in the data storage medium matches the digital twin of the defined individual in terms of age, gender, previous illnesses and blood count, so that further verification is not necessary. Data records from living individuals are understood to be historical patient data or the patient file or historical data records that reflect the medical history and all related data of the living individual. For example, years of medication use can be linked to subsequent illnesses. By utilizing the huge amount of data in the data storage medium, at least one or moreMany clinical studies can be conducted on several digital twins of individuals with at least equivalent quality of results as if these studies were conducted on living beings.

[0040] Preferably, the data from deceased individuals was generated using at least one medical data acquisition method. Medical data acquisition methods such as computed tomography (CT), magnetic resonance imaging, radiation therapy, etc., can be used in high doses, but are only permitted to a limited extent on living individuals. However, this data can contain valuable information that is very useful for completing the digital twin of a single individual, making this digital twin usable for a clinical study. Autopsies or post-mortem examinations are performed on deceased individuals to examine the body and determine the cause of death.Typically, such data was not used further and may never have been stored at all. Data is generated from the human body using diagnostic data acquisition techniques, including taking biopsies from the body, taking images, and performing CT scans. Such examinations can be performed on deceased individuals with high radiation doses, in a very short time and with very high resolution, allowing for the generation of very high-quality data from the deceased.

[0041] Alternatively or additionally, the data from living individuals was generated using at least one medical data acquisition method. Recognized medical data acquisition methods are already proven and therefore do not require additional verification, so this data from deceased or living individuals can be safely used in the digital twin of the defined individual. This simplifies the selection of statistical medical data for the digital twin.

[0042] Preferably, the data of the defined individual in the digital twin is stored in a data structure on at least one data storage medium. This makes the selected medical data easy to find and more easily linkable with the data completion criterion. For example, the medical data is grouped into clusters so that they can be easily linked to an expert database. The data structure is known to the system, so that the missing data entries in the digital twin can be added without errors. In particular, the data storage medium contains data from several different digital twins.

[0043] Preferably, the data in the data structure can be further processed in at least one data storage medium. The data is extracted from the data structure in a structured manner and prepared for easy use in clinical studies. This allows clinical studies to be conducted more quickly and reliably. Searching for data from different sources is avoided.

[0044] Preferably, an evaluation algorithm is present. The evaluation algorithm analyzes the recorded data in step d) based on at least one data completion criterion. This ensures that the statistical medical data is evaluated correctly and that the digital twin has been supplemented according to the data completion criteria.

[0045] Alternatively or additionally, the evaluation algorithm is designed to evaluate the data stored in the data structure from different digital twins in order to output them in a clinical study on a predefined question.

[0046] In particular, the evaluation algorithm is trained to identify at least one biomarker. If the aim is to develop biomarkers for specific clinical pictures in order to detect these diseases at an early stage, these can also be determined by analyzing the medical data of deceased persons (including data from when the deceased were still alive). This is done in particular by analyzing digital twins of these deceased persons. Further data (e.g. a biopsy of a tumor) can then be generated, which can be used for validation. The main advantage of this is that clinical pictures can be detected in deceased persons that were not visible in the living person. These clinical pictures could possibly have been detected using various invasive examination methods (e.g. CT scans, biopsies), but for ethical reasons such invasive examinations may not be carried out on living persons without concrete suspicion.

[0047] The evaluation algorithm is preferably designed to optimize the comparison in step e). For this purpose, the evaluation algorithm also has constraints or limits that improve and thus optimize the comparison. The evaluation algorithm can, for example, rank the available data according to additional parameters, such as the location and time of an event, in particular the time of a disease diagnosis in the individual, and thus favor data with earlier diagnosis times in the selection of statistical medical data. The constraints or limits enable further selection of the data for the digital twin.

[0048] Preferably, the evaluation algorithm is designed to provide at least one medical treatment suggestion for the defined individual. Any treatment options for a patient are validated in real time using the digital twin, since the large amount of available data allows for at least the same level of statistical significance as in a clinical study (an in silico trial with the same level of accuracy as a real clinical trial). This can save significant costs and time associated with conducting clinical trials.

[0049] The analysis algorithm is advantageously designed to capture data from the data structure. This allows the structured data to be analyzed in real time and makes it more suitable for clinical studies.

[0050] Preferably, an artificial intelligence module (KI) is present, which, based on the data completion criterion, suggests supplementing the digital twin of the defined individual with data, particularly automatically. The selected data can be processed directly by the KI module, so that no user needs to verify the selected data for the digital twin. This ensures that the selected data for the digital twin of the defined individual can also be verified by a KI module before further processing.

[0051] The selected statistical medical data can, in turn, serve as training data for the AI ​​module, making the system intelligent and self-learning. This creates a system for identifying statistical medical data that exhibits high data quality and favors the selection of high-quality data. In particular, the AI ​​module assigns a medical relevance to the statistical medical data based on the at least one data completion criterion and thus links the medical data to the data completion criterion. The AI ​​module can, in turn, be designed to assign a medical relevance to the medical data, so that it can be more easily selected when a data completion criterion is subsequently modified.The Kl module can employ a deep learning method (using artificial neural networks), in which multiple layers of artificial neurons link the input variables (feature vectors) with the output variables (classification, regression, etc.). Numerous other machine learning methods, such as random forest algorithms (randomized decision trees), or support vector machines (estimation using support vectors in the vector space of feature vectors), can also be used, particularly to limit computational effort. The Kl module is generally trained using medical data and / or data from an expert database.

[0052] Preferably, the artificial intelligence module is configured to optimize the selection of data in step f). The AI ​​module processes the statistical medical data of the multiple individuals at an optimized speed, thus accelerating the completion of the selected data.

[0053] Alternatively or additionally, the artificial intelligence module is designed to optimize the determination of missing data entries in step e). The AI ​​module processes the statistical medical data of the multiple individuals at an optimized speed, thus accelerating the comparison of the analyzed data with the specified data completion criterion and the determination of missing data entries for the digital twin based on the data completion criterion.

[0054] Preferably, at least one medical examination facility is present that is connected to the data exchange interface. One such examination facility is, for example, the Pathopsy™ Center. This is a facility for collecting medical or physical data from individuals. Its goal is to obtain data from every deceased individual, at least from every individual who died in a hospital (e.g., approximately 75% of all people in Austria). This enables a very rapid buildup of statistical data that can be used to complete a digital twin.

[0055] Alternatively or additionally, at least one medical examination device is available, which is connected to the interface for exchanging control commands. This allows the system to directly access the control system of the medical examination device and, in particular, to obtain missing medical data from living or deceased individuals for the data set of the defined individual.

[0056] Preferably, the data evaluation module is configured to generate control commands for at least one medical examination device. This allows the data evaluation module to directly access the control system of the medical examination device and, in particular, to obtain missing medical data for the defined individual.

[0057] A computer-implemented method according to the invention for creating a digital twin of an individual comprises at least the following steps: a) Providing a template for a digital twin of a defined individual; b) Specifying a data completion criterion for the digital twin of the defined individual; c) Recording data, in particular medical data, of the defined individual in the digital twin; d) Analyzing the recorded data of the defined individual in the digital twin; e) Comparing the analyzed data with the specified

[0058] Data completion criterion and definition of missing

[0059] Data entries for the digital twin based on the

[0060] Data completion criterion; f) selecting data from statistical medical data of several individuals from the at least one data storage medium, wherein the statistical medical data comprises at least data from deceased individuals; g) supplementing the missing data entries in the digital twin of the defined individual with the selected data. This computer-implemented process massively replaces or reduces clinical studies on living beings. To create the digital twin of a specific person, for example, existing data about that person (e.g. from medical records but also other information such as activity, place of residence, etc.) is initially used. To fill data gaps, additional statistically relevant medical data from several individuals that are as similar as possible is then used.This enables the testing of new therapies on the digital twin and enables parallel analysis of multiple instances of the same digital twin. The statistical medical data comes from many different individuals from different hospitals, countries, or continents.

[0061] By using this system, for example, many studies can be conducted on digital twins of humans with results at least equivalent to those of current clinical trials on living humans. Instead of real individuals, one or more digital twins can be used for these studies. A digital twin of a human does not necessarily have to be identical to the human but can also contain statistically relevant data from other humans. The computer-implemented process can be executed on a computer, such as a processing unit or a processor.

[0062] Preferably, the data of the defined individual in the digital twin is stored in a data structure on at least one data storage medium. This makes the selected medical data easy to find and more easily linkable with the data completion criterion. For example, the medical data is grouped into clusters so that they can be easily linked to an expert database. The data structure is known to the method, so that the missing data entries in the digital twin can be added without errors.

[0063] Preferably, the data in the data structure is further processed in at least one data storage medium. The data is extracted from the data structure in a structured manner and prepared for easy use in clinical studies. This allows clinical studies to be conducted more quickly and reliably. Searching for data from different sources is no longer necessary.

[0064] Advantageously, the data from the data structure is analyzed in real time, making it more suitable for clinical studies. For example, the aforementioned analysis algorithm is used for this purpose. This is particularly advantageous when using large amounts of data from deceased individuals.

[0065] Furthermore, the computer-implemented method is designed to carry out the method disclosed here.

[0066] A computer program product according to the invention comprises program instructions that, when executed by a computer, cause a system disclosed herein to execute a method described herein. The computer program product can be executed on a computing unit and thus process the program instructions step by step to provide the selected data at an interface. These can then be provided to a user or an AI module, in particular to control an examination device.

[0067] A computer-readable storage medium according to the invention, which comprises at least one computer program product which, when executed by at least one computing unit, causes the latter to carry out at least one method described herein.

[0068] Further advantages, features and details of the invention will become apparent from the following description, in which embodiments of the invention are described with reference to the drawings.

[0069] The list of reference symbols, like the technical content of the patent claims and figures, is part of the disclosure. The figures are described coherently and comprehensively. Identical reference symbols indicate identical components; reference symbols with different indices indicate functionally identical or similar components.

[0070] The invention is explained in more detail using exemplary embodiments in the following figures.

[0071] Positional designations such as "top," "bottom," "right," or "left" refer to the respective illustrations and are not to be understood as limiting. Although the invention is illustrated and described in detail by means of the figures and the associated description, this illustration and this detailed description are to be understood as illustrative and exemplary and not as limiting the invention. It is understood that those skilled in the art may make changes and modifications without departing from the scope of the following claims. In particular, the invention also encompasses embodiments having any combination of features mentioned or shown above for various aspects and / or embodiments.

[0072] The invention also encompasses individual features in the figures, even if they are shown there in conjunction with other features and / or not mentioned above. Furthermore, the term "comprising" and derivatives thereof do not exclude other elements or steps. Likewise, the indefinite article "a" or "an" and derivatives thereof do not exclude a plurality. The functions of several features listed in the claims may be fulfilled by a single unit. The terms "essentially," "about," "approximately," and the like, in connection with a property or value, specifically define the property or value. All reference signs in the claims are not to be understood as limiting the scope of the claims.

[0073] Character description

[0074] The figures are described in a coherent and comprehensive manner. The same reference symbols refer to the same components.

[0075] Fig. 1 : a first system according to the invention for creating a digital twin of an individual in a schematic representation;

[0076] Fig. 2: a further system according to the invention for creating a digital twin of an individual in a schematic representation; and Fig. 3: a computer-implemented method according to the invention with a system according to Fig. 1 in a flowchart.

[0077] Implementation of the invention

[0078] Figure 1 shows a first embodiment of a system 15 for creating a digital twin 22 of an individual 16 using data. The system 15 comprises a data evaluation module 17 with a computing unit 18, which is connected to a data storage medium 19 for data exchange, wherein the data evaluation module 17 is designed to perform at least the following steps when creating the digital twin 22: a) Providing a template 20 for a digital twin 22 of a defined individual 16; b) Specifying a data completion criterion 21 for the digital twin 22 of the defined individual 16; c) Recording data 40 of the defined individual 16 in the digital twin 22; d) Analyzing the recorded data 40 of the defined individual 16 in the digital twin 22;e) Comparing the analyzed data 41 with the defined data completion criterion 21 and determining missing data entries 42 for the digital twin 22 based on the data completion criterion 21; f) Selecting data 43 from statistical medical data 44 of several individuals 23 from the data storage medium 19, wherein the statistical medical data includes at least data from deceased individuals 24; g) Supplementing the missing data entries in the digital twin 22 of the defined individual 16 with the selected data 43;

[0079] To create the digital twin 22, in this case a specific person, existing data from this person's medical record and other information such as activity, place of residence, etc. are first used and fed into the digital twin 22. To fill data gaps, additional data 43 from several individuals 23 are then generated or used from statistical medical data 44.

[0080] The system 15 has an evaluation algorithm 33 that analyzes the recorded data 40 in step d) based on the data completion criterion 21. The evaluation algorithm 33 can be executed in the computing unit 18 and is designed, for example, to identify at least one biomarker. Furthermore, the evaluation algorithm 33 is designed to optimize the comparison of the data in step e). For this purpose, the evaluation algorithm 33 has framework conditions and limits that improve and thus optimize the comparison of the data. The evaluation algorithm 33 can rank the available recorded data 40 according to additional parameters, such as the location and time of an event, in particular the time of a disease diagnosis in the individual 16, and thus prefer data with earlier diagnosis times in the selection of the statistical medical data 44.

[0081] All relevant selected data 43 are provided at an interface 25, wherein the interface 25 is connected to a touchscreen 31 as the first display device 30. The system 15 has a control device 29, wherein the control device 29 causes the interface 25 to output the selected data 43 or, additionally, the analyzed data 41. The touchscreen 31 is connected to a graphics module 27 in order to graphically display the digital twin 22, and in particular to display it in an animatable manner. The user can also use the touchscreen 31 to perform further actions or make corrections to at least one data completion criterion 21 in order to obtain the desired statistical medical data 44 with the desired relevance.Additional display devices 32, such as a smartphone 36 or a tablet, are connected to the system 15 in order to make the data 40, 41, 42, 43, 44 available to many users of the system 15. Before selecting the data from the statistical medical data 44 of several individuals 23 in step f), these data 43 are verified in the data evaluation module 17. The arrows shown show, by way of example, an exchange of the aforementioned data with the control device 29, the data storage medium 19 or the display devices 32. The user of the system 15 defines one or more data completion criteria 21. In doing so, restrictions of the general target definition, e.g., to certain groups of people (age, gender, number, etc.), are specified and entered into the system 15 via an input device 26.In the present example, the data completion criterion 21 requires that comparable statistical medical data 44 from a minimum number of 100 individuals 23 who have type A diabetes must be available. The missing data entries 42 (e.g., certain blood values) in the digital twin 22 of the defined individual 16 are only supplemented with the statistical medical data 44 of at least 100 individuals. As an alternative to type A diabetes, other diseases, such as hepatitis, can also be defined as exclusion criteria or framework conditions. The data completion criterion 21 also includes a medical question 34. This involves a text input and is answered based on data. In the present example, the medical question 34 is formulated such that the individuals must not have any pre-existing lung diseases, such as CoV19. Such data from individuals is excluded from the system 15.

[0082] The statistical medical data 44 includes data from deceased individuals 24 and from living individuals 28. Data from deceased individuals 24 includes, on the one hand, autopsy data and, on the other hand, historical patient data or the patient file of a deceased person, or historical data sets that reflect the medical history of a deceased person and all related data of the deceased individual. The living individuals 28 also contain, among other things, medical data that statistically matches the digital twin 22 of the defined individual 16. For example, the data storage medium 19 also contains medical data that matches the digital twin of the defined individual based on age, gender, previous illnesses, and blood count.

[0083] The system 15 has an artificial intelligence module (KI) 35, which, based on the data completion criterion 21, proposes the addition of data to the digital twin 22 of the defined individual 16. The selected statistical medical data 44 are in turn used as training data for the KI module 35, so that the system 15 is intelligent and self-learning. Furthermore, the KI module 35 assigns a medical relevance to the statistical medical data 44 based on the data completion criterion 21 and thus links the statistical medical data 44 to the data completion criterion 21. The KI module 35 can in turn be configured to assign a medical relevance to the statistical medical data 44, so that they can be selected more effectively in a subsequent, modified determination of a data completion criterion.For example, the medical relevance of statistical medical data 44 includes a statement about the degree of narrowing of the heart vessels when determining the probability of a heart attack.

[0084] In a supplementary embodiment of a system according to the system 15, the evaluation algorithm is designed to provide at least one medical treatment suggestion for the defined individual 16.

[0085] In a supplementary embodiment of a system according to system 15, the artificial intelligence module 35 is configured to optimize the selection of data 43 in step f). In doing so, the AI ​​module 35 processes the statistical medical data 44 of multiple individuals at an optimized speed.

[0086] Figure 2 shows a further embodiment of a system 115 for creating a digital twin 22 of an individual 16, wherein the system 115 corresponds to the system 15 according to Figure 1 in terms of its functional and structural features and additionally includes a medical examination device 120 connected to the interface 25 for exchanging medical data 45. One such examination device 120 is, for example, the Pathopsy™ Center. Its goal is to obtain medical data 45 from each deceased individual. The medical examination device 120 is connected to the interface 25 for exchanging control commands 46. Thus, the system 115 can directly access the control of the medical examination device 120 and, in particular, obtain missing data or data entries 42 from the defined individual 16.The data evaluation module 17 is configured to generate control commands 46 for the medical examination device 120. The arrows shown illustrate, by way of example, an exchange of the aforementioned data with the control device 29, the data storage medium 19, the Kl module 35, the medical examination device 120, or the display devices 32.Figure 3 shows an embodiment of a computer-implemented method for creating a digital twin 22 of an individual 16 in a flowchart, which can be executed in a system 15 according to Figure 1 and comprises at least the following steps: a) Providing a template for a digital twin of a defined individual; b) Defining a data completion criterion for the digital twin of the defined individual; c) Recording data, in particular medical data, of the defined individual in the digital twin; d) Analyzing the recorded data of the defined individual in the digital twin; e) Comparing the analyzed data with the defined data.

[0087] Data completion criterion and definition of missing

[0088] Data entries for the digital twin based on the

[0089] Data completion criterion; f) selecting data from statistical medical data of a plurality of individuals from the at least one data storage medium, wherein the statistical medical data comprises at least data from deceased individuals; g) supplementing the missing data entries in the digital twin of the defined individual with the selected data.

[0090] A computer program product can be stored in the processing unit 18 of a computer in the system 15 and includes program instructions which, when the computer program is executed by the computer, cause the computer to execute a method described herein.

[0091] 15 Systems

[0092] 16 defined individual

[0093] 17 Data evaluation module

[0094] 18 computing unit

[0095] 19 Data storage medium

[0096] 20 templates

[0097] 21 Data completion criterion

[0098] 22 digital twin

[0099] 23 several individuals

[0100] 24 deceased individuals

[0101] 25 Interface

[0102] 26 Input device

[0103] 27 Graphics module

[0104] 28 living individuals

[0105] 29 Control device

[0106] 30 first display device

[0107] 31 touchscreen

[0108] 32 additional display devices

[0109] 33 Evaluation algorithm

[0110] 34 medical questions

[0111] 35 Kl-Modul

[0112] 36 smartphones

[0113] 40 recorded data

[0114] 41 analyzed data

[0115] 42 missing data / data entries

[0116] 43 selected dates

[0117] 44 statistical medical data

[0118] 45 medical data

[0119] 46 control commands

[0120] 115 System

[0121] 120 medical examination facilities

Claims

Patent claims 1. A system (15; 115) for creating a digital twin (22) of an individual (16), in particular a human or an animal, using data, comprising a data evaluation module (17) with at least one computing unit (18), which is connected to at least one data storage medium (19) for data exchange, wherein the data evaluation module (17) is designed to carry out at least the following steps when creating the digital twin (22), wherein in particular the order of the following steps is specified: a) providing a template (20) for a digital twin (22) of a defined individual (16); b) specifying at least one data completion criterion (21) for the digital twin (22) of the defined individual (16); c) recording data, in particular medical data, of the defined individual (16) in the digital twin (22);d) analyzing the recorded data of the defined individual (16) in the digital twin (22); e) comparing the analyzed data with the defined data completion criterion (21) and determining missing data entries for the digital twin (22) based on the data completion criterion (21); f) selecting data from statistical medical data of a plurality of individuals (23) from the at least one data storage medium (19), wherein the statistical medical data comprises at least data from deceased individuals (24); g) supplementing the missing data entries in the digital twin (22) of the defined individual (16) with the selected data.

2. System according to claim 1, characterized in that at least steps b) to g) are carried out several times.

3. System according to claim 1 or 2, characterized in that the data completion criterion (21) comprises at least one medical question (34).

4. System according to one of the preceding claims, characterized in that an interface (25) is provided at which the data of the digital twin (25) can be made available to a user and / or can be displayed, in particular animated, on a display device (30), wherein a control device (29) is provided and the control device (29) causes the interface (25) and / or the display device (30) to output the selected data.

5. System according to one of the preceding claims, characterized in that the display device (30) comprises a graphics module (27) for graphically displaying the digital twin.

6. System according to one of the preceding claims, characterized in that the data of deceased individuals (24) were generated by means of at least one medical data acquisition method.

7. System according to one of the preceding claims, characterized in that the statistical medical data comprise at least data from living individuals (28).

8. System according to claim 7, characterized in that the data were generated from living individuals (28) by means of at least one medical data acquisition method.

9. System according to one of the preceding claims, characterized in that the data of the defined individual (16) in the digital twin (22) are stored in a data structure in at least one data storage medium (19).

10. System according to claim 9, characterized in that the data in the data structure can be further processed in the at least one data storage medium (19). 11 . System according to one of the preceding claims, characterized in that an evaluation algorithm (33) is present.

12. System according to claim 11, characterized in that the evaluation algorithm (33) is designed to identify at least one biomarker.

13. System according to claim 11 or 12, characterized in that the evaluation algorithm (33) is designed to provide at least one medical treatment suggestion for the defined individual (16).

14. System according to one of claims 11 to 13, characterized in that the evaluation algorithm (33) is designed to evaluate the data stored in the data structure from different digital twins.

15. System according to one of the preceding claims, characterized in that an artificial intelligence module (35) is present which, on the basis of the data completion criterion (21), proposes the completion of the digital twin (22) of the defined individual (16) with data, in particular automatically.

16. System according to claim 15, characterized in that the artificial intelligence module (35) is designed to optimize the selection of the data in step f).

17. System according to claim 15 or 16, characterized in that the artificial intelligence module (35) is designed to optimize the determination of the missing data entries in step e).

18. System according to one of the preceding claims, characterized in that at least one medical examination device (120) is present, which is connected to the interface (25) for the exchange of data and / or control commands.

19. A computer-implemented method for creating a digital twin (22) of an individual (16), comprising at least the following steps: a) providing a template for a digital twin of a defined individual; b) specifying a data completion criterion for the digital twin of the defined individual; c) Recording data, in particular medical data, of the defined individual in the digital twin; d) Analyzing the recorded data of the defined individual in the digital twin; e) Comparing the analyzed data with the defined Data completion criterion and definition of missing Data entries for the digital twin based on the Data completion criterion; f) selecting data from statistical medical data of a plurality of individuals from the at least one data storage medium, wherein the statistical medical data comprises at least data from deceased individuals; g) supplementing the missing data entries in the digital twin of the defined individual with the selected data.

20. Computer-implemented method according to claim 19, characterized in that the data of the defined individual (16) in the digital twin (22) are stored in a data structure in the at least one data storage medium (19).

21. Computer-implemented method according to claim 19 or 20, characterized in that the data in the data structure are further processed in the at least one data storage medium (19).

22. A computer program product comprising program instructions which, when the computer program is executed by a computer, cause a system according to one of claims 1 to 18 to carry out a method according to one of claims 19 to 21.

23. A computer-readable storage medium comprising at least one computer program product which, when executed by at least one computing unit, causes the latter to carry out at least one method according to one of claims 19 to 21.

Images