Spiritual Digital Twin

Abstract

This document teaches a computer-executed system for simulating the mental state of a human mind. This system includes a memory having a storage area for storing a plurality of instructions, a virtual data storage area, and a human mental data storage area. The mental data storage area stores mental data regarding the mental state of a human mind. A processor can generate at least a digital twin of a human mind by executing one of the plurality of instructions stored in the storage area. The digital twin includes one or more models of a human mind based on human mental data. One or more models of the mind are stored in the virtual data storage area to enable analysis of the human mind by the processor within the virtual data storage area. The digital twin includes a link between the mental data storage area and the virtual data storage area.

Description

Cross - reference to related applications

[0001] This application claims the benefit of the filing date of U.S. Provisional Patent Application No. 63 / 350,115, the entire disclosure of which is incorporated herein by reference. Field of the Invention

[0002] The present invention relates to methods of creating, maintaining, updating, predicting, and otherwise utilizing a periodically updated representation of a human mental state, as well as computer - implemented systems. BACKGROUND OF THE INVENTION

[0003] A digital twin is a virtual representation of an object or process that functions as a real - time digital copy that reflects the state of the object or process by being updated with information from the object's environment and the object itself. The concept of "digital twin" has its origin in engineering science, provides conceptual and ethical implications for treatment and preventive measures, and provides a framework for practicing data - driven health management. When used with respect to humans, the current concept of digital twin constructs a computer - simulated representation of an individual that dynamically reflects the human body and its physiological state. The concept of digital twin is gradually emerging in the health management industry (reviewed in Non - Patent Document 1).

[0004] Patent Document 1 teaches a method of developing an anthropomorphic digital model (or digital twin) of at least a part of a human anatomical structure using a computer system including a processor device and a communication module under the control of the processor device. This method includes receiving, using the processor device, input data related to the actual physical state of at least a part of the human anatomical structure using a communication module, searching a database of digital models that model different physical states of at least a part of the anatomical structure, and selecting from the database a digital model that best matches the actual physical state of at least a part of the human anatomical structure based on at least a part of the received input data. Based on the received input data, the selected digital model is developed using a physiological development model associated with the selected digital model to develop the modeled physical state of the selected digital model. The physically modeled state developed in this way is closer to the actual physical state according to the input data.

[0005] A computerized method for health data management using a digital twin of an individual patient based on health information related to the individual patient is known from Patent Document 2. This patent publication teaches that a digital twin of an individual patient is a digital representation of at least one health state of the individual patient, and forms a digital twin of a patient population based on health information related to the patient population using a computing device of a health data system. The digital twin described in this application can be used to determine whether a human patient's drug usage profile indicates a potential for taking a regulated drug by mistake. Furthermore, when it is determined that there is a possibility of taking a regulated drug by mistake, this method includes sending a notification indicating the possibility of the patient's misuse. However, in this application, the mental state of the patient is not considered.

[0006] A system for managing the stress level and mental health of the human body is known from Patent Document 3. The system of this application has one or more body sensors and a primary processing unit that executes an artificial intelligence system. The body sensors are configured to measure at least one of the physiological parameters of the human body, the body movements of the human body, the heat consumption of the human body, or a combination thereof, and generate body data periodically or in real time. The primary processing unit is configured to receive and process the body data and is configured to determine at least one of the health and stress level of the human body. The primary processing unit is configured to provide treatment and quantitatively provide insights regarding the effectiveness of psychotherapy including CBD, meditation, and mindfulness. In the prior art, digital twins or simulations that describe the physical parameters of the human body are disclosed. To evaluate and predict the mental state of a human, it is useful to simulate the human mind and continuously represent it.

Prior Art Documents

Patent Documents

[0007]

Patent Document 1

Patent Document 2

Patent Document 3

Non-Patent Document 1

Non-Patent Document 2

Summary of the Invention

[0008] The present invention is a computer-executed system that simulates the mental state of the human mind. This computer-executed system includes a memory that includes a storage area for storing a plurality of instructions, a virtual data storage area, and a mental data storage area. The mental data storage area stores user data related to the mental state of the human mind represented by the computer-executed system. At least one processor can generate at least a digital twin of the human mind by executing one of the plurality of instructions stored in the storage area. The digital twin includes one or more models of the human mind based on mental data. One or more models of the mind are stored and updated in the virtual data storage area, enabling analysis of the human mind by at least one processor within the virtual data storage area. The digital twin includes a link between the mental data storage area and the virtual data storage area.

[0009] In a further aspect, the computer-executed system includes an interface for creating a representation of the mental state that is available to a user external to the system. This may include a graphical user interface that visualizes human mental data and provides access to human mental data based on the output from the digital twin. This may also include an application interface that enables users of other software or platforms to utilize such data. Furthermore, the computer execution system may include one or more notification devices for executing notifications from the digital twin, and a prediction device for predicting the mental state of a person.

[0010] A computer-executed method for simulating the mental state of a person's mind is also disclosed. This method includes applying one or more simulation parameters to the digital twin, where the digital twin includes one or more models of a person's mind. The digital twin enables the simulation of a person's mind. One or more simulation parameters cause the digital twin to generate an output that simulates the behavior of a person's mind, for example, one or more subsequent states of the mind. Applying one or more simulation parameters to the digital twin may include inputting the one or more simulation parameters into one or more mental models. Simulating the mental state of a person's mind may include calculating the trajectory of the mental state of a person's mind in a state space (or phase space). The trajectory of the mental state of a person's mind in a state space (or phase space) may depend on the simulation parameters input into the digital twin of the person's mind. The computer execution system is receivable and updatable of information that changes the mental state, such as the presence, location, and heart rate of another person, from a sensor, user, or external data source.

[0011] The output of the digital twin includes, for example, a graphical depiction of the mental state, a prediction of mental health, or a connection to an avatar generator.

[0012] In a further aspect, the method enables comparison of the output of the digital twin with a set of responses of another human or a mental digital twin.

[0013] Other aspects, features and advantages of the present invention will become readily apparent from the following detailed description which is merely illustrative of embodiments and implementations. The present invention can take other different embodiments, and without departing from the spirit and scope of the present invention, some of its details can be clearly changed in various respects. Therefore, the drawings and description are illustrative in nature and not restrictive of the present invention. Additional objects and advantages of the present invention are described in part in the following detailed description, and some of them will be apparent from the detailed description or can be understood by practicing the present invention.

Brief Description of the Drawings

[0014] To more fully understand the present invention and its advantages, reference should be made to the following detailed description and the accompanying drawings.

[0015]

Figure 1

[0016]

Figure 2

[0017]

Figure 3

[0018]

Figure 4

Modes for Carrying Out the Invention

[0019] The present invention will be described with reference to the drawings. The embodiments and aspects of the present invention described in this specification are merely illustrative and do not limit the scope of protection of the claims in any way. The present invention is defined by the claims and their equivalents. The features of one aspect or one embodiment of the present invention can be combined with the features of another aspect and / or embodiment of the present invention.

[0020] FIG. 1 shows an overview of a computer-executed system 10 that simulates the mental state of a human mind. The computer-executed system 10 includes a memory 20 having a storage area 23 for storing a plurality of instructions, a virtual data storage area 25, and a mental data storage area 27. As will be described later, the computer-executed system 10 has at least one processor 30 capable of generating a digital twin 40 of a human mind by executing one of the plurality of instructions 23 stored in the storage area. The virtual data storage area 25 is used by the processor 30 to generate the digital twin 40. The mental data storage area 27 stores mental data related to the mental state of a human.

[0021] In one aspect of the present disclosure, the mental data storage area 27 can store collective mental data related to the mental states of other humans, including mental data of healthy humans and / or humans suffering from mental and / or physical diseases. In another aspect, mental data related to the mental state of a human (e.g., representing the mental state of a human) can be used, and for example, depending on the monitoring situation of the onset or progression of a mental disease or mental disorder, the mental data can be defined and / or modified. The definition and / or modification of mental data includes the update of mental data. The update of mental data may include, for example, updating the mental data daily or hourly depending on the mental state of the simulated human and the availability of newly collected mental data. The update may be performed at a predetermined time or may be prompted by a person or an event. The definition and / or modification of mental data may be performed at the discretion of a person and / or depending on the mental state of a person. In a further aspect, the mental data may be related to and / or not limited to one or more of a person's mood, emotion, arousal, attention, energy level, hunger, thirst, aggression, or level of fear.

[0022] In yet another aspect, the mental data storage area 27 may further include one or more of physiological data, environmental (or situation) data, and / or personal data. The physiological data may include data related to the quality of sleep, physical activity, body movement, skin electrical activity, electrocardiogram, heart rate variability, and / or neuroimaging. The environmental (or situation) data may include location data (e.g., GPS data) and data related to social communication such as emails, text messages, social media posts, voice mails, calls, weather conditions, voice patterns, movement patterns, etc. The personal data may include medical data (e.g., from electronic health records) such as sleep time, number of steps, time spent outdoors, time spent in green spaces, time spent on exercise, data related to behavioral therapy or medication. A person's mental state may be directly or indirectly linked to physiological data, environmental (or situation) data, and / or personal data.

[0023] In one aspect, some of the mental data, physiological data, environmental (or situation) data, and / or personal data may be measured or inferred from objective data collected by sensors 80, 90 (e.g., energy level from movement patterns) and / or the user's response. The data may be collected using the method of ecological momentary assessment (EMA) (see Non-Patent Document 2). The EMA may include using the motion sensor 90, location data (e.g., GPS), and data regarding the use of digital apps (e.g., based on the number of logins or parts of digital apps used).

[0024] The digital twin 40 includes one or more models 50 of the human mind and is created by the processor 30 within the virtual data storage area 25. The digital twin 40 also includes a link between the mental data storage area 27 and the virtual data storage area 25, whereby the digital twin 40 can create a simulation of the human mind within the virtual data storage area 25.

[0025] The computer-execution system 10 also has a graphical user interface 60, for example, displayed on a computer screen, tablet, or smartphone. The graphical user interface 60 visualizes the simulated mental state based on the data stored in the mental data storage area 27 and output from the digital twin 40. The graphical user interface 60 also enables a user of the computer-execution system 10, such as a medical practitioner, to interact with the digital twin 40. For example, a user may want to input data into the computer-execution system 10 to understand what impact it has on the human mental state. The digital twin 40 can simulate the patient's reaction.

[0026] The computer-execution system 10 can also connect to an external application or platform. One example is the graphics engine 62 that creates an avatar within the metaverse. The avatar includes one or more graphical representations of a human and can take the form of a human or a creature. The graphics engine 62 includes a generator for creating a human avatar, and by being supplied with parameters from the digital twin 40, the reaction of the avatar in the metaverse can be improved. For example, these parameters may affect the expression and body language of the avatar. As another example, information regarding the mental state can be made available from outside, so that services or products suitable for the mood can be proposed, or a call can be made to a therapist.

[0027] Physiological data may be collected from at least one physiological sensor 80 attached to the living human. This physiological data may be supplied to the computer execution system 10 and stored in the mental data storage area 27. By applying the physiological data to the digital twin 40, the reaction of a human may be simulated. At least one physiological sensor 80 measures, for example, heart rate, blood pressure, heart rate variability, skin conductance, pupil reaction, vocal activity, body movement, eye movement, etc.

[0028] At least one environmental sensor 90 may be connected to the computer execution system 10. This at least one environmental sensor 90 measures, for example, the intensity of light, temperature, air pressure, humidity, proximity to other humans, or proximity to devices worn by other humans.

[0029] In one aspect, a mobile device such as a smartphone may be equipped with at least one physiological sensor 80 and / or at least one environmental sensor 90. This mobile device may be used for generating a part of the physiological data from at least one physiological sensor 80, a part of the environmental (or situation) data from at least one environmental sensor 90, and / or personal data. The mobile device can record the distance walked by the human carrying the mobile device using GPS. A mobile device can further detect communication with other humans by interacting with another mobile device carried by another human. These communications are recorded along with a timestamp indicating the length of the communication time and the communication time. The mobile device can also ask a human questions to determine the human's health status and record the answers to the questions. Another human can also ask the questions. The physiological data, environmental data, and / or personal data collected by the mobile device are stored in the mental data storage area 27. In another disclosed aspect, the mobile device includes wearable devices such as smartwatches, smart bands, smart rings, smart chest straps, and / or XR goggles.

[0030] Other personal data and / or environmental data that can be recorded by the mobile device may include, but are not limited to, walking data such as walking speed and acceleration, and geographical location.

[0031] The digital twin 40 is stored within the virtual data storage area 25 and within a device having a local processor (such as the user's smartphone). The representation is executed by a recurrent neural network (RNN) that is pre-trained using a series of datasets that characterize the mental state according to human behavior. This recurrent neural network (RNN) is, for example, a deep recurrent neural network (deep RNN) and / or a piecewise linear recurrent neural network (PLRNN). The above training is performed using code described in MATLAB or other coding languages. For example, by combining the health status of a user that is periodically evaluated with movement measurements (e.g., continuous or periodic), a dataset can be constructed to demonstrate that the activity has a beneficial effect on the mood, or by combining green space information based on satellite images with geographical location information based on GPS, a dataset can be constructed to show that the user's mood improves when discovering green space in an urban environment.

[0032] The input to this deep neural network is user input, sensor data, and external datasets linked to that information (e.g., information about the type of local environment extracted from geographical information based on the user's current location provided by a sensor).

[0033] The trained neural network outputs a set of mental parameters. The training of the neural network is executed within the virtual data storage area 25. The neural network thus trained updates the representation of the mental state and forms the digital twin 40. This representation is updated whenever user input, sensor, and environmental data become available. As published in previous research by the inventors, network training uses new training, which is a dynamic system that optimally reconstructs and predicts the human mind based on the network. This means that the network fundamentally regenerates the mental state after training, and as will be explained below, it means that the network has become a functional reflector available for prediction.

[0034] The trained RNN model uses the discrete-time dynamic system xt = Fθ(xt-1,ut). Here, $x_t$ represents the state of a dynamic system (such as the human brain) that evolves over time according to the recurrence map $F_{\theta}$ parameterized by the vector of parameters $\theta$, and $u_t$ represents the time series of external inputs (input variables) to the computer execution system 10. Also, in conventional statistics, $F_{\theta}$ is usually linear and is an important class of autoregressive moving average (ARMA) models, but in the case of RNNs, $F_{\theta}$ is (highly) non - linear. When the maximum length $T$ of the time sequence for training the RNN is fixed as a known value, $F_{\theta}$ is "unraveled" in time, and a technique called "backpropagation through time (BPTT)" may be adopted for training [20, 182]. An important property of RNNs is not only that they can be used as time - series analysis and prediction tools, but also that because RNNs can autonomously generate sequences and temporal behavior, they can function as powerful AI devices even in situations where goal - directed actions and planning are required. By using dynamic systems, it is possible to associate the development of activities in RNNs with brain processes based on stable states, oscillations (limit cycles), or chaotic states. The result of the branching of the dynamic system may be associated with the transition of the human mental state to a desirable or undesirable state. When the mental state is transitioning to an undesirable state, the computer execution system 10 and / or the digital twin 40 may issue a warning. The warning may be displayed, for example, on the graphical user interface 60. The warning may be provided to the application interface, for example, in the form of a notification 64 for transmission to the therapist (see below).

[0035] RNNs can be estimated from mental data, physiological data, environmental (or situational data), and / or personal data. In particular, this method involves collecting time series of mental data, physiological data, environment (or situational data), and / or personal data, which may be referred to as a multimodal experiential observation time series (a time series that experientially observes multiple modalities). The RNN can be estimated from the multimodal experiential observation time series using maximum likelihood estimation or Bayesian estimation. The estimation may be based on the model p(yt|θ), where yt is the measured time series of mental data, physiological data, environmental (or situational) data, and / or personal data, and θ is the vector of parameters that defines the recursive map Fθ representing the evolution of the dynamic system represented by xt. In inference, the evidence lower bound (ELBO) may be adopted to approximate the calculation of the integral.

[0036] An example of the multimodal experiential observation time series used for training the RNN is described below. 356 healthy participants aged 18 to 28 were recruited. All participants wore an accelerometer and a research smartphone and answered an additional questionnaire (see Table 1 in Figure 3). The participants answered a questionnaire including sociodemographic information, height and weight, and several psychological evaluations as detailed in Figure 3. According to the established procedure, participants who met the following criteria were excluded. (i) There were significant technical problems with the accelerometer such as the measurement ending midway (N = 28), (ii) the compliance rate of the electronic diary was less than 30% (N = 2), or (iii) the questionnaire data was missing (N = 9). The final sample consisted of 317 healthy participants (57.09% female) with a mean age of 23.08 years (SD = 2.83, see Table 2 in Figure 4).

[0037] Table 2 shows the results of three samples: "Full sample", "fMRI sample", and "COVID-19 sample". In Table 2 shown in FIG. 4, the column of sample size n indicates the number of individuals for whom information on the corresponding variable is available. SD means standard deviation. Household income was evaluated by classifying it into 13 grades as the monthly household income after tax: 1) less than 500 euros, 2) 500 - 749 euros, 3) 750 - 999 euros, 4) 1000 - 1249 euros, 5) 1250 - 1499 euros, 6) 1500 - 1749 euros, 7) 1750 - 1999 euros, 8) 2000 - 2249 euros, 9) 2250 - 2499 euros, 10) 2500 - 2999 euros, 11) 3000 - 3999 euros, 12) 4000 - 4999 euros, 13) more than 5000 euros. To descriptively compare the samples in this table, category means were assigned to individuals. For example, a value of 624.5 euros was assigned to the participants belonging to the second category. The values of exercise acceleration intensity were averaged for the entire group of participants and for the entire research week. For the emotional valence, the intraclass correlation coefficient (ICC) was used to calculate the estimated variance of the outcome variable. In this study, 35.0% of the variance in emotional valence is considered to be due to within-subject variation.

[0038] The participants were explained about the study, submitted written consent, and received a monetary reward for their participation at the end of the study. The participants received an extensive technical explanation including the test in a face-to-face format and then carried a research smartphone and an accelerometer continuously for 7 days in their daily lives. After one week, the participants returned the devices and reported on the most important places they visited. To enhance the participants' memory, an established procedure similar to the Day Reconstruction Method 1 was applied. Briefly, a timestamped digital map (movisens Geocoder) showing all the geographical locations visited and the routes covered (by tracking with the smartphone) was used. Participants were asked to retrospectively label all situations (at home, at work, out with friends, etc.). These location labels were later assigned to three categories representing situations: "home", "work", and "other". Prior to these procedures, participants filled out a series of questionnaires including sociodemographic information, height and weight, and several psychological assessments as detailed in Figure 3.

[0039] To assess physical activity, participants wore a triaxial accelerometer (Move II or Move III, movisens GmbH, Germany) on the right hip continuously for 7 days during waking hours. The accelerometer captures movements up to ±8g with a 12-bit resolution and a sampling frequency of 64Hz, and appropriately assesses human physical activity. To calculate the exercise acceleration intensity, i.e., the vector amplitude of the acceleration evaluated on the three sensor axes (in milligrams [(g) / 1000]), the software DataAnalyzer (version 1.6.12129) from movisens GmbH was used. That is, the gravity component was removed by a high-pass filter (0.25Hz), and artificial factors (e.g., vibrations during cycling on a rough road surface and sensor impacts) were removed by a low-pass filter (11Hz). To distinguish light physical activity, the metabolic equivalent of task (MET) was calculated. This is a measure of energy consumption and is defined as the ratio of the metabolic rate during the task to the metabolic rate at standard rest of 1.0 (4.184kJ)*kg-1*h-1, where 1MET represents the resting metabolic rate while sitting quietly. Based on MET, activities can be classified, for example, as light physical activity (1.6 - 2.9MET). MET was calculated using the software DataAnalyzer (movisens GmbH, Germany). Before MET calculation, the gravity component was removed by a high-pass filter (0.25 Hz), and artificial factors (e.g., vibrations during cycling on rough roads and impacts on sensors) were removed by a low-pass filter (11 Hz). MET was calculated in two steps. First, the activity class was estimated based on the acceleration and barometric pressure signals. Based on the detected class, a model for MET calculation was selected, and the MET value was calculated based on the moving acceleration, altitude change extracted from barometric pressure data, age, gender, weight, and height, and the procedures described elsewhere in this specification were established.

[0040] The method using the electronic diary and sampling was executed using the ecological momentary assessment software movisensXS, version 0.6.3658 (movisensXS12). After thorough guidance, the participants carried a smartphone (Motorola Moto G, Motorola Mobility) continuously for 7 days and were prompted by voice, visual, and vibration signals to fill in the electronic diary multiple times a day. This prompt could be postponed by 5 minutes, 10 minutes, or 15 minutes. The prompt was executed based on a sampling method that combined time and location. This method is superior to the conventional (long-interval and easily forgotten) time-based sampling method, and the individual's attention is dispersed. For each day of the week when the study was conducted, the electronic diary prompt was executed between 7:30 am and 10:30 pm, and the time interval between the prompts was at least 40 minutes and at most 100 minutes. As a result, the participants were prompted to fill in the electronic diary a total of 9 - 23 times a day. In the location-based prompt execution algorithm, the distance between the participant's current location and previous location was continuously monitored. When the distance exceeded 500 meters, the prompt was executed. Furthermore, the participants were prompted at two fixed times every day (8:00 am and 10:20 pm).

[0041] To evaluate the emotional value, the study used a shortened scale of two established items with appropriate reliability and sensitivity for measuring mood fluctuations within the subjects. These two items were presented as bipolar scales with a score range of 0 to 100 on two visual analog scales on the computer, with the ascending and descending orders reversed (from "satisfied" to "dissatisfied", from "zufrieden" to "unzufrieden" in German, from "poor" to "good", from "unwohl" to "wohl" in German). The scores of the two items were later corrected, averaged, and used as the dependent variable in the multi - step analysis. The actual social contact at the prompting of the electronic diary was evaluated by an established binary scale that asked the participants whether they were with other people.

[0042] The trained neural network can be maintained both on a local processor (such as an individual's smartphone) and / or within the virtual data storage area 25. If the model is (also) maintained locally, it is periodically transferred to the virtual storage area, and an updated digital mind twin 40 is provided in both the local and virtual spaces.

[0043] A flowchart of a computer - implemented method for simulating the mental state of the human mind is shown in FIG. 2. This computer - implemented method starts at step 200 and includes applying one or more simulation parameters to the digital twin 40 at step 210. The digital twin 40 is pre - programmed as described above. The simulation parameters 43 can be one or more readings from a plurality of physiological sensors 80 and / or one or more readings from environmental sensors 90.

[0044] In step 215, the digital twin 40 simulates the human mind, and in step 220, based on one or more simulation parameters 43, the digital twin 40 generates a digital twin output 45 that simulates the behavior of the human mind in response to the one or more simulation parameters 43 and the notification sent in step 230.

[0045] Examples of the digital twin output 45 include indicating or predicting early signs of mental health problems such as predicting a recurrence of alcohol abuse, graphically displaying to the user the relationship between mood and levels of physical activity or social interaction to promote mental health behavior, further promoting such behavior, or outputting mental state parameters that inform the facial and body expressions of an avatar representing a human user within the metaverse.

[0046] In one aspect, the mental data storage area 27 may include data on healthy humans, and this method includes, in step 225, comparing the digital twin output 45 with the responses of healthy humans. This comparison can be used to improve the signs / predictions of mental health problems. By the notification 64 in step 230, a text message or a phone call is made to the human and / or the doctor or caregiver, enabling them to intervene. In another aspect, the digital twin output 45 can be transferred to an external application or platform. This may be a graphics engine 62 that generates expressions linked to emotions on a human avatar, or an application that targets therapeutic interventions or proposes services and products using the output mental state parameters.

[0047] Furthermore, the digital twin 40 can be used in the simulation of psychotherapeutic mediation and other forms of social interaction based on the prediction characteristics of a trained neural network representing mental state parameters, thereby assisting in the human treatment plan.

Description of Symbols

[0048] 10 ··· Computer Execution System 20 ··· Memory 23 ··· Multiple Instructions 25 ··· Virtual Data Storage Area 27 ··· Mental Data Storage Area 30 ··· Processor 40 ··· Digital Twin 43 ··· Stimulation Parameters 45 ··· Digital Twin Output 50 ··· Model 60 ··· Graphical User Interface 62 ··· Graphics Engine 64 ··· Notification 70 ··· Notification Device 80 ··· Physiological Sensor 90 ··· Environmental Sensor

Claims

1. A computer execution system (10) that simulates the mental state of the human mind, A memory (20) includes a memory area (23) that stores multiple commands, a virtual data storage area (25), and a mental data storage area (27), wherein the mental data storage area (27) stores human mental data relating to the mental state of the human mind. A processor (30) generates a digital twin (40) of the human mind by executing one of a plurality of commands stored in the memory area (23), The digital twin (40) includes one or more models (50) of the human mind based on mental data, One or more models (50) are stored in a virtual data storage area (25) to enable the processor (30) to analyze the human mind within the virtual data storage area (25). A computer execution system (10) in which the digital twin (40) includes a link between the mental data storage area (27) and the virtual data storage area (25).

2. The computer execution system (10) according to claim 1, further comprising a graphical user interface (60) that visualizes the human mental data and provides access to the human mental data based on the output from the digital twin (40).

3. The computer execution system (10) according to claim 1 or claim 2, further comprising a notification device (70) for executing notifications from the digital twin (40).

4. The computer execution system (10) according to claim 1 further comprises a predictive device (80) for predicting the mental state of the human mind.

5. It also includes a graphics engine (62) for generating avatars. The computer execution system (10) according to claim 1, wherein the graphics engine (62) is connected to the digital twin (40), thereby enabling the generation of an avatar representing the mental state of the human mind.

6. A computer execution method for simulating the mental state of the human mind, This includes applying one or more simulation parameters (43) to the digital twin (40) (210), The digital twin (40) simulates the human mind, A computer execution method in which one or more simulation parameters (43) cause the digital twin (40) to generate a digital twin output (45) that simulates the behavior of the human mind (220) according to the one or more simulation parameters (43).

7. The computer execution method according to claim 6, wherein the digital twin output (45) includes mental health prediction.

8. A computer execution method according to claim 6 or 7, further comprising comparing the digital twin output (45) with the response of a healthy human (225).

9. The computer execution method according to claim 6, further comprising generating a notification relating to the digital twin output (45).

10. The computer execution method according to claim 6, further comprising generating an avatar based on the data in the digital twin (40).

11. A computer program product stored in memory (20), A computer program product comprising a storage area (23) storing a plurality of instructions that enable a processor to perform the computer execution method described in claim 6.

12. A method for training a neural network to create a digital twin (40), Inputting one or more datasets that represent characteristics of human mental states into a neural network, Training the neural network using the one or more datasets mentioned above, A method for training a neural network to create a digital twin (40), including [specific example].

Images