Integrated mental health diagnostic device for attention-deficit / hyperactivity disorder, internet gaming disorder, depression, impulsivity, anxiety, obsessive-compulsiveness, and aggression, continuous mental health monitoring method, and feedback alarm notification device

Abstract

An integrated mental health diagnostic method for attention-deficit / hyperactivity disorder (ADHD), internet gaming disorder (IGD), depression, impulsivity, anxiety, obsessive-compulsiveness, and aggression, according to one embodiment of the present disclosure, comprises: acquiring user data from a user terminal; inputting the user data to a pre-trained artificial intelligence model; obtaining, from the artificial intelligence model, mental disorder prediction index values indicating the possibilities of occurrence of a plurality of mental disorders; and, on the basis of the mental disorder prediction index values, diagnosing the possibility of onset for each of the plurality of mental disorders of a user of the user terminal.

Description

Integrated mental health diagnostic device for attention deficit disorder, internet gaming disorder, depression, impulsivity, anxiety, obsessive-compulsive disorder, and aggression, continuous mental health monitoring method, and feedback alert notification device

[0001] The present invention relates to a mental health integrated diagnosis system based on a digital phenotype, and more specifically, to a technology for simultaneously monitoring and diagnosing multiple mental health conditions experienced by a user by analyzing the user's bio-signal information obtained from the user's terminal device or the user's terminal sensor data such as the user's eye movements, voice patterns, keyboard patterns, touch and stylus pen use, acceleration and angular velocity, and illuminance.

[0002] According to the National Health Insurance Service, the number of patients seeking mental health treatment has increased since the outbreak of the novel coronavirus (COVID-19). Even excluding the exceptional circumstances caused by COVID-19, many adults are experiencing mental health issues due to the many stresses of modern society. Young people are anxious about their future, adults are concerned about employment and retirement, and the elderly are living with anxiety about loneliness, health, and death. As the number of people with mental illness continues to rise, various changes are occurring in the field of mental health treatment.

[0003] Recently, the US government launched a large-scale research project to integrate AI into mental health, and the private sector is actively developing products and services that apply AI to mental health, including digital therapeutics. Furthermore, Europe is pursuing research and development on various mental health issues using AI at the EU level, such as "mental health monitoring after cancer treatment." The UK is also leveraging AI to advance the digitalization of health and medical systems, including mental health issues like early diagnosis of dementia.

[0004] Mental illness has traditionally been assessed based on responses to self-report questionnaires based on the American Psychiatric Association's Diagnostic and Statistical Manual of Mental Illnesses (DSM). While self-report questionnaires operate under the assumption of accurate self-awareness and honest responses, objective responses are often challenging for individuals with mental illness, those with limited language skills, those with low self-awareness of their mental health, and those with intellectual disabilities. Furthermore, the social stigma surrounding mental illness often leads to avoidance or false responses.

[0005] To diagnose symptoms of mental illness, research is focusing on objective tests and indicators rather than subjective assessments. Symptoms such as sadness, euphoria, impulsivity, and difficulty concentrating are inherently subjective. Furthermore, psychiatrists and clinical psychologists often make subjective assessments based on the patient's condition during diagnosis, making it difficult to objectively assess a patient's mental health.

[0006] Additionally, to diagnose symptoms of mental illness, it's necessary to visit a psychiatrist or clinical psychologist. Even when diagnosed by a psychiatrist, patients may become agitated by unfamiliar surroundings or hide their emotions or conditions, making it difficult for the psychiatrist to accurately assess the patient's mental illness.

[0007] Moreover, mental health is a daily issue and possesses a dynamic nature. This is called "dynamic nature." It is characterized by the fact that it cannot be treated with a single prescription.

[0008] Non-integrated diagnostic systems centered on individual, single diagnoses have the following problems:

[0009] 1. Ignore symptoms of congestion and overlapping

[0010] ADHD and Internet Gaming Disorder often coexist with mental health problems such as depression, impulsivity, anxiety, and aggression, and one symptom can exacerbate the other. A non-integrated approach ignores these interconnectedness, leading to individual symptoms being identified and managed separately. Consequently, it is difficult to identify exacerbating factors or interactions between symptoms, limiting the development of comprehensive treatment strategies.

[0011] 2. Difficulty in making an accurate diagnosis

[0012] Monitoring only individual symptoms can make it difficult to assess a patient's overall mental health, potentially reducing diagnostic accuracy. For example, treatment methods and difficulties differ between those with ADHD and Internet Gaming Disorder (IGA) and those with both separately. Furthermore, impulsivity and anxiety are common symptoms that can manifest in various conditions, including ADHD, depression, and aggression. A non-integrative approach can lead to interpreting these symptoms as a single cause, hindering accurate diagnosis.

[0013] 3. Decreased sensitivity to symptom fluctuations

[0014] Symptoms such as impulsivity, anxiety, and depression can fluctuate over time and across situations. Non-integrated approaches, unable to monitor these dynamic changes in real time, have limitations in effectively managing symptom fluctuations. Consequently, they fail to respond promptly to rapid symptom changes, potentially delaying the timing of intervention.

[0015] 4. Duplicate evaluations and inefficient use of resources

[0016] Monitoring each symptom independently can lead to redundant assessments, resulting in inefficient use of time, money, and other resources. For example, individual assessments for anxiety and depression can be redundant, which not only consumes significant resources but can also be tiring for users.

[0017] 5. Difficulties in establishing individualized treatment strategies

[0018] A non-integrated approach fails to account for the interplay of multiple symptoms, making it difficult to develop a personalized treatment plan tailored to the user's overall mental health. Without integrated monitoring, only general treatments for each symptom are applied, failing to adequately address the individual needs of the user or patient.

[0019] 6. Limitations of long-term management

[0020] Non-integrated monitoring has limitations in identifying symptom patterns and trends, which are essential for long-term mental health management. Relying on fragmented assessments without ongoing data collection makes it difficult to maintain consistency and continuity in treatment, making it difficult to ensure long-term treatment effectiveness.

[0021] Therefore, there is a need for a method to analyze and diagnose a user's mental health status based on biometric and behavioral signals obtained through digital electronic devices that the user continuously uses in daily life, without having to meet a mental health specialist or clinical psychologist.

[0022] The present invention relates to a digital phenotyping-based integrated mental health diagnosis system, and more specifically, to a technology for simultaneously and continuously monitoring and diagnosing multiple mental health conditions by utilizing a transdiagnostic model that comprehensively monitors ADHD, IGD, depression, anxiety, impulsivity, aggression, and obsessive-compulsive disorder. A transdiagnostic model can refer to characteristics or processes common to various mental illnesses or disorders, transcending specific diagnostic categories or categories among mental illnesses. In other words, a transdiagnostic model is not limited to a specific mental illness or mental disorder, but can simultaneously diagnose multiple mental illnesses. Furthermore, by utilizing digital phenotyping technology, it can overcome the limitations of existing questionnaire-based diagnosis, such as subjective reporting, false reporting due to social stigma, and one-time reporting.

[0023] Embodiments of the present invention utilize digital phenotype or ecological moment assessment (EMA) technology that evaluates mental states through behavioral signals and bio-signals related to mental phenomena in electronic devices such as user terminals to analyze user terminal device usage patterns or user bio-signals to evaluate mental states such as anxiety and depression.

[0024] According to embodiments of the present invention, multiple users' mental health conditions can be simultaneously monitored and diagnosed by analyzing their terminal device usage patterns or their bio-signals. According to the present invention, comorbidity, i.e., when a person experiences multiple mental health issues simultaneously, can be diagnosed and analyzed.

[0025] According to the present invention, the mental health status of multiple users can be monitored and diagnosed simultaneously by analyzing the user's terminal device usage pattern or the user's bio-signals, etc., without a separate questionnaire or consultation with a doctor for the diagnosis of mental status.

[0026] In addition, the mental health diagnosis through an integrated approach based on sensor data of a user terminal according to the present invention has the following advantages.

[0027] 1. Simultaneous monitoring of various symptoms and conditions

[0028] Digital phenotyping using sensor data can simultaneously and continuously track multiple mental health conditions, including ADHD, IGD, depression, anxiety, impulsivity, and aggression. This overcomes the limitations of questionnaires, which require a single, continuous response for each individual mental health issue. Furthermore, because each symptom or condition can interact and influence the other, integrated monitoring allows for a more accurate and comprehensive understanding of mental health.

[0029] 2. Accurate diagnosis possible through understanding of interrelationships

[0030] By synthesizing data from various sensors, correlations between symptoms can be identified. This allows for clear separation of overlapping symptoms such as depression, anxiety, and impulsivity, and a better understanding of the causes and patterns of symptom occurrence, enabling more precise diagnosis.

[0031] 3. Timing prediction and optimization for real-time response and intervention.

[0032] Sensor data can detect changes in condition in real time, allowing for the prediction of rapidly worsening symptoms and immediate response. For example, a warning signal can be sent when anxiety or aggression spikes, enabling timely intervention and preventing the worsening of symptoms.

[0033] 4. Personalized management possible

[0034] Based on data collected through sensors, individual symptom patterns can be analyzed to design individualized management plans. For example, a person with both ADHD and anxiety could be provided with management strategies tailored to specific times when anxiety levels are elevated. This personalized management maximizes therapeutic effectiveness.

[0035] 5. Long-term tracking and analysis of change patterns

[0036] By accumulating long-term data, we can identify long-term trends in symptom changes. This can help us understand how each symptom develops or improves, providing users with long-term treatment strategies. For example, we can identify seasonal patterns of worsening symptoms and take proactive measures.

[0037] 6. Efficient resource utilization

[0038] An integrated approach can save time and money by reducing redundant assessment and diagnostic processes. Furthermore, monitoring multiple conditions simultaneously can reduce user fatigue and encourage more sustained engagement.

[0039] 7. Increased interactive therapeutic effects

[0040] An integrative approach allows us to understand how symptoms interact, allowing us to tailor treatment approaches for greater effectiveness. For example, if depression and anxiety are interrelated, managing anxiety can enhance the therapeutic effects of the interaction, potentially improving depressive symptoms.

[0041] An integrated approach based on sensor data has the advantage of comprehensively identifying various symptoms and providing personalized management, making mental health management more effective.

[0042] FIG. 1 is a diagram illustrating an integrated mental health diagnosis system according to one embodiment of the present invention.

[0043] Figure 2 is a block diagram showing the configuration of the user terminal device (110) of Figure 1.

[0044] Figure 3 is a block diagram showing the configuration of a mental health integrated diagnosis server (120) according to one embodiment of the present invention.

[0045] FIG. 4 is a diagram showing a plurality of mental illness prediction index values ​​according to one embodiment of the present invention.

[0046] Figure 5 is a flow chart showing an operation method of a mental health diagnosis system according to one embodiment of the present invention.

[0047] An integrated mental health diagnosis method for attention deficit disorder (ADHD), internet gaming disorder (IGD), depression, impulsivity, anxiety, obsessive-compulsive disorder, and aggression according to one embodiment of the present invention includes the steps of: obtaining user data from a user terminal; inputting the user data into a pre-trained artificial intelligence model, and obtaining mental illness prediction index values ​​indicating the possibility of occurrence of a plurality of mental illnesses from the artificial intelligence model; and diagnosing the possibility of occurrence of each of the plurality of mental illnesses of a user of the user terminal based on the mental illness prediction index values.

[0048] In one embodiment, the user data includes at least one of location information, call records, message content, social media activity, number of steps, heart rate, user's eye movement, eye blinking, user's keyboard typing pattern, touch screen touching pattern, stylus pen input pattern, stylus pen coordinate information, number of keyboard inputs, time interval between screen touches, and time interval between keyboard inputs obtained from the user terminal.

[0049] In one embodiment, the artificial intelligence model may be trained using user data obtained from users suffering from each of the plurality of mental illnesses and user data obtained from users not suffering from the mental illness.

[0050] In one embodiment, the mental illness prediction index value has a value between 0 and 1, and when the mental illness prediction index value is greater than a predetermined threshold value, the onset of the mental illness can be diagnosed.

[0051] A mental health diagnosis device according to one embodiment of the present invention includes a communication unit that obtains user data from a user terminal and transmits a mental illness diagnosis result to the user terminal; and a mental health diagnosis unit that inputs the user data into a pre-trained artificial intelligence model, obtains mental illness prediction index values ​​indicating the possibility of occurrence of a plurality of mental illnesses from the artificial intelligence model, and diagnoses the possibility of occurrence of each of the plurality of mental illnesses of a user of the user terminal based on the mental illness prediction index values.

[0052] The following merely exemplifies the principles of the present invention. Therefore, those skilled in the art will be able to implement the principles of the present invention and invent various devices within the scope and spirit of the present invention, even if not explicitly described or illustrated herein. Furthermore, all conditional terms and embodiments listed herein are expressly intended, in principle, to facilitate understanding of the present invention, and should be understood as being in no way limiting to the specifically enumerated embodiments and conditions.

[0053] Furthermore, all detailed descriptions of the principles, aspects, and embodiments of the present invention, as well as specific embodiments, should be understood to encompass structural and functional equivalents thereof. Furthermore, such equivalents should be understood to encompass not only currently known equivalents but also equivalents developed in the future, i.e., all devices invented to perform the same function, regardless of structure.

[0054] Thus, for example, the block diagrams herein should be understood as representing conceptual views of exemplary circuits embodying the principles of the present invention. Similarly, all flowcharts, state transition diagrams, pseudocode, and the like, which may be substantially represented on a computer-readable medium, should be understood as representing various processes performed by a computer or processor, regardless of whether a computer or processor is explicitly depicted.

[0055] Furthermore, any explicit use of terms such as processor, controller, or similar concepts should not be construed as exclusively referring to hardware capable of executing software, but should be understood to implicitly include, without limitation, digital signal processor (DSP) hardware, read-only memory (ROM), random access memory (RAM), and non-volatile memory for storing software. Other commonly used hardware may also be included.

[0056] The above-described purposes, features, and advantages will become more apparent through the following detailed description, taken in conjunction with the accompanying drawings. Accordingly, those skilled in the art will be able to readily implement the technical concepts of the present invention. Furthermore, in describing the present invention, detailed descriptions of known technologies related to the present invention will be omitted if they are deemed to unnecessarily obscure the gist of the invention.

[0057] Hereinafter, a preferred embodiment of the present invention will be described in detail with reference to the attached drawings.

[0058] FIG. 1 is a diagram illustrating an integrated mental health diagnosis system according to one embodiment of the present invention. Referring to FIG. 1, the integrated mental health diagnosis system (100) includes a user terminal (110) and an integrated mental health diagnosis server (120). The integrated mental health diagnosis system (100) of FIG. 1 is merely one embodiment of the present invention and is not limited to the configuration of FIG. 1.

[0059] The user terminal (110) is connected to the mental health integrated diagnosis server (120) through a network. The network refers to a connection structure that enables information exchange between each node, such as a plurality of terminals and servers, and examples of such networks include a local area network (LAN), a wide area network (WAN), the Internet (WWW), a wired and wireless data communication network, a telephone network, a wired and wireless television communication network, etc. Examples of wireless data communication networks may include 3G, 4G, 5G, 6G, WIMAX (World Interoperability for Microwave Access), Wi-Fi, the Internet, a local area network (LAN), a wireless local area network (Wireless LAN), a wide area network (WAN), a personal area network (PAN), a radio frequency (RF), a Bluetooth network, a near-field communication (NFC) network, a satellite broadcasting network, an analog broadcasting network, etc.

[0060] The user terminal (110) transmits user data acquired through equipped sensors or cameras, etc., to a mental health integrated diagnosis server (120) connected via a network. In addition, the user terminal (110) can output a report or data on the user's mental health diagnosis results from the mental health integrated diagnosis server (120).

[0061] The user terminal (110) may be a wearable electronic device mounted on a part of the user's body, or an electronic device used by the user. For example, the user terminal (110) may be an electronic device such as a computer, smartphone, tablet PC, smart pad, smart watch, smart ring, smart glasses, or HMD, which has a CPU, memory, and a communication module capable of communicating with other external devices, and can connect to a remote server or terminal via a network.

[0062] User data can be either passive or active. Passive data is not data that the user actively creates, but rather data generated through unconscious actions. This data includes data generated by actions such as clicking or swiping to use the terminal. Active data is data consciously provided by the user, such as (emotional) diaries, learning problem solving, and social media comments. While active data is not the core of the present invention, it can be used in conjunction with passive data and interactive data.

[0063] Embodiments of the present invention can solve the problem of users intentionally hiding their mental health status or misdiagnosing their mental health status through false behavior by monitoring, evaluating, or diagnosing the mental health status of users using such passive data, i.e., behavioral patterns that users perform unconsciously without being conscious of it.

[0064] User data may be acquired using the Ecological Momentary Assessment (EMA) technique using digital phenotypes. Digital phenotypes are models that utilize extracted passive data to monitor a user's attention span and mental health.

[0065] Ecological Momentary Assessment (EMA) is a method for measuring the emotions, thoughts, and behaviors people experience at specific moments in their daily lives in real time. Unlike traditional self-report surveys, the EMA technique utilizes digital phenotypes to assess mental health, reducing stigma and reminiscence-related distortions because the assessment is conducted immediately.

[0066] User data can be collected at regular intervals through an app installed on the user terminal (110) or a camera or sensor equipped on the user terminal (110). Digital phenotype is a method of measuring user data by continuously monitoring the user's behavior, activity, physiological state, etc. through the user terminal (110). The user terminal (110) can acquire various user data, such as location information, call records, message content, social media activity, number of steps, heart rate, the user's electrocardiogram (ECG), pulse (Photoplethysmography, PPG), blood pressure, etc., through sensors.

[0067] In addition, as user data obtained by applying such EMA technique or digital phenotype, all bio-signals or behavioral pattern information that can be obtained from the user when the user uses the user terminal (110) can be used, including location information, call records, message content, social media activity, number of steps, heart rate, user's eye movements, eye blinking, user's keyboard typing pattern, touch screen touching pattern, stylus pen input pattern, stylus pen coordinates, number of screen touches during a given period of time, number of keyboard inputs, time intervals between screen touches, time intervals between keyboard inputs, etc.

[0068] User data acquired from the user terminal (110) can be used to diagnose the user's mental health. For example, depression can lead to slowed thinking and behavior. This can result in a slow initiation of action when given a task, a slow keyboard input speed, or the repetition of certain words. In contrast, impulsive disorder can lead to a user attempting to answer a problem without reading it, failing to space between words, or a rapid typing speed. Furthermore, anxiety can lead to behavioral patterns such as frequent eye movements, increased blinking, or increased touches on the user terminal. In obsessive-compulsive disorder, a user tends to press message buttons for extended periods or with greater force, and the number of times they delete information exceeds the amount of information typed. Therefore, the present invention can utilize such user data to analyze and diagnose a user's mental health.

[0069] Furthermore, the present invention is not limited to a specific mental illness, but can simultaneously diagnose multiple mental illnesses. By modeling and processing raw data extracted from various sensors, impulsivity, depression, anxiety, and other conditions can be simultaneously monitored, assessed, and diagnosed. For example, in the case of attention deficit disorder (ADHD), even users with the same diagnosis may exhibit varying degrees of impulsivity and depression, which are complex symptoms. Furthermore, cases of strong impulsivity and strong depression may manifest differently.

[0070] Users with both Internet gaming disorder and ADHD, or those with only one of the two, may experience differences in behavioral patterns, treatment methods, and outcomes.

[0071] Patients suffering from both depression and anxiety may exhibit similar tendencies in user data, and the present invention can diagnose multiple mental illnesses from user data using a pre-trained artificial intelligence model using user data obtained from users suffering from multiple mental illnesses.

[0072] Figure 2 is a block diagram showing the configuration of the user terminal device (110) of Figure 1.

[0073] The user terminal device (110) includes an input / output unit (210), a display unit (220), a sensor unit (230), and a communication unit (240).

[0074] The input / output unit (210) may include a microphone and / or a speaker, and may output an audio signal through the speaker or receive a user's voice data or audio data generated unconsciously by the user through the microphone.

[0075] The sensor unit (230) may include at least one of a camera sensor, a microphone sensor, a keyboard sensor, a touch and stylus sensor, a GPS sensor for measuring the user's location, an acceleration sensor for measuring the movement of the user and the device, a gyro sensor, or other sensors for measuring the user's bio-signals such as an electrocardiogram (ECG), a photoplethysmography (PPG), or blood pressure. However, the present invention is not limited thereto, and various sensors for measuring the user's bio-signals or the user's activity status may be used.

[0076] The sensor unit (230) can obtain various user data, such as the number of times the user touches the screen during a certain period of time, the number of times the user inputs a keyboard during a certain period of time, the strength of touching the screen, information on the time interval between touching or inputting the screen, and the length of the screen touch, through various sensors.

[0077] In addition, the sensor unit (230) can obtain gaze data, facial expression data, body temperature, and heart rate estimation data by measuring the user's gaze or eye movements using gaze tracking technology. Gaze tracking technology is a technology that can determine where and how much the user looks or extract the user's eye movements and movement path, etc. The gaze tracking technology used in the sensor unit (230) can generate gaze data that includes information on the user's status and behavioral patterns regarding gaze movements. The gaze data can be used to determine the user's interest and fascination, and the user's psychological or mental state. Specifically, the gaze data can include information on the movement of the eyes and gaze, path, and the number of times the eyes blink during a given period of time.

[0078] The various user data acquired from the sensor unit (230) can be used to diagnose the user's mental health. Preprocessing of various user data allows the user data to be identified or weighted to predict mental illness from multiple user data sets, thereby determining a digital phenotype. The digital phenotype can then be input into a pre-trained artificial intelligence model to predict multiple mental illnesses.

[0079] The display unit (220) outputs various contents or information to the user and may include a touch screen. The display unit (220) may receive touch, gesture, proximity, drag, swipe, or hovering input using an electronic pen or a part of the user's body through the touch screen.

[0080] The communication unit (240) is a component that performs communication with an external device, and the communication unit (240) transmits user data acquired through the display unit (220) and sensor unit (230) to the mental health integrated diagnosis server (120).

[0081] Figure 3 is a block diagram showing the configuration of a mental health integrated diagnosis server (120) according to one embodiment of the present invention.

[0082] Referring to FIG. 3, the mental health integrated diagnosis server (120) includes a storage unit (310), a mental health diagnosis unit (320), and a communication unit (330).

[0083] The communication unit (330) receives user data transmitted from the user terminal (110), the storage unit (310) stores the user data transmitted from the user terminal (110), and the mental health diagnosis unit (320) stores the results of each user's mental analysis in a database. The storage unit (310) can store login information such as ID and password for each user, the user's personal information, and the results of each user's mental analysis. The storage unit (310) can store the state of change in the user's mental health by accumulating and storing the user data acquired for each user.

[0084] The mental health diagnosis unit (320) can analyze and diagnose the mental health status of a user using user data received from a user terminal (110). The mental health diagnosis unit (320) determines a digital phenotype by processing the user data by specifying or weighting user data used to predict mental illness from multiple user data through pre-processing of various user data, and inputs the digital phenotype into a pre-trained artificial intelligence model to predict multiple mental illnesses.

[0085] The mental health diagnosis unit (320) can analyze and diagnose the mental health status of a user by using an artificial intelligence model trained using user data obtained from users suffering from various mental illnesses and user data obtained from users who do not suffer from the mental illness. For example, an artificial intelligence model can be trained using user data obtained from users suffering from mental illnesses such as Internet Game Disorder (IGD), Attention Deficit Disorder (ADHD), impulsivity, depression, aggression, anxiety, obsessive-compulsive disorder, and pre-psychotic symptoms, and user data obtained from healthy users who do not suffer from the mental illness.

[0086] An artificial intelligence model can be constructed by applying any one neural network or a composite neural network among support vector machines (SVM), convolutional neural networks (CNN), recurrent neural networks (RNN), and decision tree models.

[0087] SVM is a model that defines a decision boundary, i.e., a baseline for classification, and takes user data as input. Support vectors refer to data points close to the decision boundary, and margin refers to the distance between the decision boundary and the support vector. Since the optimal decision boundary maximizes the margin, learning can be performed using this. The mental health diagnosis unit (320) can train the model by determining a decision boundary for distinguishing multiple mental illnesses using user data. In addition, the mental health diagnosis unit (320) can utilize an SVM model trained using features extracted from user data through a CNN or RNN.

[0088] The mental health diagnosis unit (320) may perform preprocessing on user data to diagnose mental illness. For example, the mental health diagnosis unit (320) may extract data with a high contribution to a specific mental illness from the user data, determine a digital phenotype, and use it as input values ​​for an artificial intelligence model, or perform preprocessing by assigning weights to each user data. The various user data acquired by the sensor unit (230) may be used to diagnose the user's mental health.

[0089] The mental health diagnosis unit (320) inputs user data received from the user terminal (110) into a pre-trained artificial intelligence model as input values, and the artificial intelligence model analyzes and diagnoses the user's mental health status based on the user data, and can output multiple mental illness prediction index values.

[0090] The mental health diagnosis unit (320) can transmit multiple mental illness prediction index values ​​output from the artificial intelligence model to the user terminal (110) via the communication unit (330), or output a report or data on the user's mental health diagnosis results.

[0091] FIG. 4 is a diagram showing a plurality of mental illness prediction index values ​​according to one embodiment of the present invention.

[0092] The mental health diagnosis unit (320) can determine multiple mental illness prediction index values ​​from user data using an artificial intelligence model. The mental health diagnosis unit (320) can diagnose the possibility of onset of each of the multiple mental illnesses based on the mental illness prediction index value for each mental illness. For example, the artificial intelligence model of the mental health diagnosis unit (320) can determine and output a value between 0 and 1 as a mental illness prediction index value for each of the multiple mental illnesses. The closer the mental illness prediction index value is to 0 as the possibility of onset of the corresponding mental illness is lower, and the closer the mental illness prediction index value is to 1 as the possibility of onset of the corresponding mental illness is higher. The mental health diagnosis unit (320) can determine that the user is suffering from the corresponding mental illness when each mental illness prediction index value exceeds a predetermined threshold. For example, if the threshold is 0.5 and each mental illness prediction index value exceeds 0.5 and is diagnosed as suffering from the corresponding mental illness, then in Figure 4, user Park OO can be determined to be suffering from mental illness related to depression and impulsivity, patient Oh OO can be determined to be suffering from mental illness related to ADHD, impulsivity, and anxiety, and patient Jo OO can be determined to be suffering from mental illness related to IGD, depression, anxiety, and obsessive-compulsive disorder.

[0093] In this way, according to the present invention, it is possible to diagnose multiple mental illnesses simultaneously using user data received from a user terminal (110).

[0094] Figure 5 is a flow chart showing an operation method of a mental health diagnosis system according to one embodiment of the present invention.

[0095] Referring to FIG. 5, the user terminal (110) acquires user data using the EMA technique or digital phenotype (S510) and transmits the acquired user data to the mental health integrated diagnosis server (120) (S520). The user data may include location information, call records, message content, social media activity, number of steps, heart rate, user's eye movements, eye blinking, user's keyboard typing pattern, touch screen touching pattern, stylus pen input pattern, stylus pen coordinate information, number of keyboard inputs, time interval between screen touches, time interval between keyboard inputs, etc. As described above, the user data is not limited thereto, and all passive data or active data acquired through the user terminal (110) may be used.

[0096] The mental health integrated diagnosis server (120) inputs the received user data as input values ​​into a pre-trained artificial intelligence model, and the artificial intelligence model analyzes and diagnoses the user's mental health status based on the user data, and can output multiple mental illness prediction index values ​​(S530). The mental health integrated diagnosis server (120) determines a digital phenotype by processing the user data by specifying or weighting user data used to predict mental illness from multiple user data through pre-processing of various user data, and inputs the digital phenotype into a pre-trained artificial intelligence model to predict multiple mental illnesses.

[0097] The mental health integrated diagnosis server (120) can diagnose the mental health of a user of a user terminal (110) based on multiple mental illness prediction index values ​​output from an artificial intelligence model and transmit the mental health diagnosis results to the user terminal (110) (S540), or output a report or data on the user's mental health diagnosis results.

[0098] The user terminal (110) analyzes the received mental health diagnosis results, and if it is predicted that the user of the user terminal (110) is likely to be suffering from a mental illness, a feedback warning that informs the user of the user terminal (110) of the possibility of the occurrence of the mental illness can be output to draw the user's attention (S550).

[0099] This integrated mental health diagnosis process is performed periodically and can be used to monitor the mental health status of the user of the user terminal (110).

[0100] Meanwhile, the methods according to the various embodiments of the present invention described above may be implemented in the form of installation data for execution on a terminal device and provided to each server or device in a state stored on various non-transitory computer-readable media. Accordingly, the user terminal (110) may access the server or device and download the installation data.

[0101] A non-transitory readable medium refers to a medium that permanently stores data and can be read by a device, rather than a medium that stores data for a short period of time, such as a register, cache, or memory. Specifically, the various applications or programs described above may be stored and provided on non-transitory readable media, such as a CD, DVD, hard disk, Blu-ray disc, USB, memory card, or ROM.

[0102] In addition, although the preferred embodiments of the present invention have been illustrated and described above, the present invention is not limited to the specific embodiments described above, and various modifications can be made by a person having ordinary skill in the art to which the present invention pertains without departing from the gist of the present invention as claimed in the claims, and such modifications should not be understood individually from the technical idea or prospect of the present invention.

Claims

1. In the integrated mental health diagnosis method for attention deficit disorder (ADHD), Internet gaming disorder (IGD), depression, impulsivity, anxiety, obsessive-compulsive disorder, and aggression, A step of obtaining user data from a user terminal; A step of inputting the user data into a pre-trained artificial intelligence model and obtaining mental illness prediction index values ​​indicating the possibility of occurrence of multiple mental illnesses from the artificial intelligence model; and An integrated mental health diagnosis method, comprising a step of diagnosing the possibility of onset of each of the plurality of mental illnesses of a user of the user terminal based on the mental illness prediction index values.

2. In paragraph 1, An integrated mental health diagnosis method, wherein the user data includes at least one of location information, call records, message content, social media activity, number of steps, heart rate, user's eye movements, eye blinks, user's keyboard typing pattern, touch screen touching pattern, stylus pen input pattern, stylus pen coordinate information, number of keyboard inputs, time interval between screen touches, and time interval between keyboard inputs obtained from the user terminal.

3. In paragraph 1, An integrated mental health diagnosis method further comprising a step of determining a digital phenotype by processing the user data by specifying or weighting user data used to predict mental illness from multiple user data through pre-processing of the user data, and inputting the digital phenotype into a pre-trained artificial intelligence model.

4. In paragraph 1, An integrated mental health diagnosis method, characterized in that the artificial intelligence model is learned using user data obtained from users suffering from each mental illness and user data obtained from users not suffering from the mental illness, for each of the plurality of mental illnesses.

5. In paragraph 1, An integrated mental health diagnosis method in which the above mental illness prediction index value has a value between 0 and 1, and a diagnosis is made as the onset of the mental illness when the above mental illness prediction index value is greater than a predetermined threshold value.

6. In paragraph 1, An integrated mental health diagnosis method, wherein, if the user of the user terminal is predicted to be suffering from a mental illness based on the diagnosis results of the possibility of onset of each of the above multiple mental illnesses, a feedback warning is output to the user of the user terminal informing the user of the possibility of onset of the mental illness.

7. In the integrated mental health diagnostic device for attention deficit disorder (ADHD), Internet gaming disorder (IGD), depression, impulsivity, anxiety, obsessive-compulsive disorder, and aggression, A communication unit that obtains user data from a user terminal and transmits mental illness diagnosis results to the user terminal; A mental health diagnosis device comprising a mental health diagnosis unit that inputs the user data into a pre-learned artificial intelligence model, obtains mental illness prediction index values ​​indicating the possibility of occurrence of a plurality of mental illnesses from the artificial intelligence model, and diagnoses the possibility of occurrence of each of the plurality of mental illnesses of a user of the user terminal based on the mental illness prediction index values.

8. In paragraph 7, A mental health diagnosis device, wherein the user data includes at least one of location information, call records, message content, social media activity, number of steps, heart rate, user's eye movements, eye blinks, user's keyboard typing pattern, touch screen touching pattern, stylus pen input pattern, stylus pen coordinate information, number of keyboard inputs, time interval between screen touches, and time interval between keyboard inputs obtained from the user terminal.

9. In paragraph 7, A mental health diagnosis device that determines a digital phenotype by processing the user data by specifying or weighting user data used to predict mental illness from multiple user data through pre-processing of the above user data, and inputs the digital phenotype into a pre-trained artificial intelligence model.

10. In paragraph 7, A mental health diagnosis device, characterized in that the artificial intelligence model is learned using user data obtained from users suffering from each mental illness and user data obtained from users not suffering from the mental illness, for each of the plurality of mental illnesses.

11. In paragraph 7, A mental health diagnosis device in which the above mental illness prediction index value has a value between 0 and 1, and the mental illness diagnosis unit diagnoses the onset of the mental illness when the above mental illness prediction index value is greater than a predetermined threshold value.

12. In paragraph 7, A mental health diagnosis device that outputs a feedback warning to the user of the user terminal informing the user of the possibility of occurrence of the mental illness when the possibility of occurrence of the mental illness is predicted based on the diagnosis result of the possibility of occurrence of the multiple mental illnesses above.

Images