Information processing systems, information processing methods, and programs

Abstract

To enable support in the differential diagnosis of heart disease. [Solution] An information processing system comprising: an answer input unit that receives input from a patient's answers to questions for differentiating mental illnesses; a video image acquisition unit that acquires video images of the patient; a biological response detection unit that analyzes the video images to detect changes in the patient's biological responses; and an estimation unit that estimates the mental illness the patient is suffering from by providing the received answers and detected changes in biological responses to a learning model that has learned the answers, changes in biological responses, and mental illnesses.

Description

Technical Field

[0006] , , ,

[0005] , ,

[0007] , ,

[0001] The present invention relates to an information processing system, an information processing method, and a program.

Background Art

[0002] In Patent Document 1, the risk of cardiovascular disease is evaluated.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] It is difficult to determine a patient's subjective condition such as mental health.

[0005] The present invention has been made in view of such a background, and an object thereof is to provide a technique capable of assisting in the discrimination of heart disease.

Means for Solving the Problems

[0006] The main invention of the present invention for solving the above problems is an information processing system, comprising: a response input unit that receives an input of a response from a patient to a question for discriminating a mental disorder; a moving image acquisition unit that acquires a moving image of the patient; a biological reaction detection unit that analyzes the moving image to detect a change in the biological reaction of the patient; and an estimation unit that gives the received response and the detected change in the biological reaction to a learning model that has learned the response, the change in the biological reaction, and the mental disorder, and estimates the mental disorder with which the patient is suffering.

[0007] Regarding other problems disclosed in the present application and methods for solving them, they will be clarified by the embodiments of the invention and the drawings. [Effects of the Invention]

[0008] According to the present invention, it is possible to assist in the differential diagnosis of heart disease. [Brief explanation of the drawing]

[0009] [Figure 1] This figure shows an example of the overall configuration of an information processing system. [Figure 2] This figure shows an example of the hardware configuration of management server 2. [Figure 3] This figure shows an example of the software configuration for management server 2. [Figure 4] This diagram illustrates the operation of management server 2. [Modes for carrying out the invention]

[0010] <System Overview> The following describes an information processing system according to one embodiment of the present invention. The information processing system of this embodiment supports the differential diagnosis of mental illnesses (such as depression, bipolar disorder, obsessive-compulsive disorder, sleep disorders, eating disorders, alcohol dependence, adjustment disorder, autism spectrum disorder, attention deficit hyperactivity disorder, schizophrenia, dementia, developmental disorders, panic disorder, and PTSD). The information processing system of this embodiment differentiates between mental illnesses (or healthy individuals) based on the patient's responses to questions for differentiating mental illnesses (tests such as QIDS, DSM-IV, DSM-5, ADOS-2, ASRS, EAT-26, and MMSE) and the results of analysis obtained by analyzing video footage of the patient's conversation (everyday conversation).

[0011] Conversations in a medical setting can be broadly divided into specialized conversations aimed at examination and diagnosis (conversations conducted as part of medical procedures, such as confirming symptoms, explaining test results, and discussing treatment plans) and everyday conversational interactions (natural dialogue unrelated to medical treatment, such as small talk in the waiting room, or casual conversations about the weather or family matters before and after an examination). In this embodiment, everyday conversation refers to linguistic communication as social interaction that occurs naturally among people present, rather than being conducted for a specific job or purpose. Everyday conversation can be described as linguistic communication that has low purpose (not primarily aimed at solving a specific problem or gathering information), spontaneous occurrence (not planned), two-way (not one-way information transmission, but mutual exchange), and a function of maintaining and building social relationships. In other words, even though it is linguistic communication that takes place in the same medical setting, conversations for interviews and diagnoses are excluded from "everyday conversation," while small talk exchanged between examinations is included in "everyday conversation." Furthermore, everyday conversation includes situations where the patient is speaking unilaterally (where medical professionals or agents are primarily listening to the patient).

[0012] Furthermore, the information processing system of this embodiment accepts input of diagnostic results from physicians (including medical professionals who are not physicians; the same applies hereinafter), and determines the validity of the physician's diagnostic results by comparing them with the above-mentioned response and the diagnostic results based on video images.

[0013] Figure 1 shows an example of the overall configuration of an information processing system. The information processing system in this embodiment includes a management server 2. The management server 2 is connected to a physician terminal 1 and a patient terminal 3 via a communication network. The communication network is, for example, the internet and is constructed using public telephone networks, mobile phone networks, wireless communication channels, Ethernet (registered trademark), etc.

[0014] Doctor's terminal 1 is a computer operated by a physician. Doctor's terminal 1 can be, for example, a smartphone, a tablet computer, or a personal computer.

[0015] Patient terminal 3 is a computer operated by the patient. Patient terminal 3 can be, for example, a smartphone, a tablet computer, or a personal computer.

[0016] The management server 2 may be a general-purpose computer such as a workstation or personal computer, or it may be logically implemented through cloud computing.

[0017] <Management Server> Figure 2 shows an example of the hardware configuration of the management server 2. Note that the illustrated configuration is just one example, and other configurations are also possible. The management server 2 includes a CPU 201, memory 202, storage device 203, communication interface 204, input device 205, and output device 206. The storage device 203 stores various data and programs, such as a hard disk drive, solid-state drive, or flash memory. The communication interface 204 is an interface for connecting to a communication network, such as an adapter for connecting to Ethernet®, a modem for connecting to a public telephone network, a wireless communication device for wireless communication, or a USB (Universal Serial Bus) connector or RS232C connector for serial communication. The input device 205 is for inputting data, such as a keyboard, mouse, touch panel, button, or microphone. The output device 206 is for outputting data, such as a display, printer, or speaker. Furthermore, each functional unit of the management server 2, as described later, is realized by the CPU 201 reading programs stored in the storage device 203 into memory 202 and executing them, and each storage unit of the management server 2 is realized as part of the storage area provided by memory 202 and storage device 203.

[0018] Figure 3 is a diagram showing an example of the software configuration of the management server 2. The management server 2 includes a response input unit 211, a moving image acquisition unit 212, a biological reaction detection unit 213, an estimation unit 214, a diagnosis result input unit 215, a misdiagnosis possibility determination unit 216, a daily conversation processing unit 217, and a learning model storage unit 231.

[0019] <Memory unit> The learning model storage unit 231 stores a trained learning model (hereinafter referred to as a disease model) by machine learning for estimating mental disorders. The disease model stored in the learning model storage unit 231 may be, for example, one learned by machine learning using, as training data, the responses from patients to questions for differentiating mental disorders, changes in biological reactions of the patients, and the mental disorders (or the fact that the patient is healthy) of the patients. The disease model can be a classifier. The disease model may also be a generator (large language model) that has been fine-tuned. Further, the learning model storage unit 231 may be provided in an external server instead of the management server 2, and the management server 2 may be configured to access the external server.

[0020] The learning model storage unit 231 may be configured to store a trained learning model (hereinafter referred to as a determination model) by machine learning for determining the diagnosis results by doctors. The determination model can be created by machine learning using, as training data, the estimation results of mental disorders using the disease model, the diagnosis results by doctors, and the correctness of those diagnosis results. Note that the determination model may be created using, as training data, the responses from patients to questions for differentiating mental disorders, changes in biological reactions of the patients, and the diagnosis results by doctors. Further, as features given to the determination model, the attributes of doctors (which may be information for identifying doctors or may be the age, years of experience, main medical departments, etc. of doctors) may be given. In this case, the management server 2 may be provided with a doctor information storage unit for managing the attributes of doctors, and at the time of inference, the attributes of the doctor who made the diagnosis can be read from the doctor information storage unit and given to the determination model.

[0021] <Functional Unit> The response input unit 211 receives the input of the patient's response to the questions for differentiating mental disorders. The response input unit 211 may be configured to receive the input of the patient's response from a doctor (receive the response from the doctor terminal 1), or may be configured to present questions to the patient and directly receive the input of the response from the patient (receive the response from the patient terminal 3).

[0022] The moving image acquisition unit 212 acquires a moving image of the patient. The moving image is assumed to be one that captures the patient engaged in daily conversation. The moving image acquisition unit 212 can acquire a moving image that captures the state of the conversation between the patient and the daily conversation processing unit 217 described later. The moving image acquisition unit 212 can receive from the patient terminal 3 a moving image of the patient's state captured by the patient terminal 3 using a camera. The moving image acquisition unit 212 may be configured to acquire a moving image that captures the state of the conversation with the patient from the doctor terminal 1. In this case, the doctor terminal 1 may perform the shooting, or a moving image file captured by another camera may be transmitted from the doctor terminal 1 to the management server 2.

[0023] The biological reaction detection unit 213 analyzes the moving image and detects changes in the patient's biological reaction.

[0024] The biological reaction detection unit 213 can, for example, separate the moving image into a set of images (a collection of frame images) and audio, and analyze changes in the biological reaction from each. For example, the biological reaction detection unit 213 can analyze changes in the biological reaction related to at least one of the expression, eye line, pulse, and facial movement by analyzing the user's face image using the frame images separated from the moving image. Also, the biological reaction detection unit 2^13 can analyze changes in the biological reaction related to at least one of the user's speech content and voice quality by analyzing the audio separated from the moving image.

[0025] When a person's emotions change, it manifests as changes in biological responses such as facial expression, gaze, pulse, facial movements, speech content, and voice quality. In this embodiment, changes in the user's emotions are analyzed by analyzing changes in the user's biological responses. One example of the emotion analyzed in this embodiment is the degree of pleasure or displeasure. In this embodiment, the biological response detection unit 213 can calculate a biological response index value that reflects the content of the changes in biological responses by quantifying the changes in biological responses according to predetermined criteria.

[0026] The analysis of facial expression changes can be performed, for example, as follows: For each frame image, the facial region is identified within the frame image, and the identified facial expressions are classified into several categories according to a pre-trained image analysis model. Based on the classification results, it is then possible to analyze whether positive or negative facial expression changes have occurred between consecutive frame images, and to determine the magnitude of these changes, and to calculate an facial expression change index value corresponding to the analysis results.

[0027] The analysis of changes in eye movement can be performed, for example, as follows: For each frame image, the eye region is identified within the frame image, and the direction of both eyes is analyzed to determine where the user is looking. For example, it can be analyzed whether the user is looking at the speaker's face, the shared document being displayed, or looking off-screen. It may also be possible to analyze whether the eye movement is large or small, and whether the movement is frequent or infrequent. Changes in eye movement are also related to the user's level of concentration. The bio-response detection unit 213 can calculate an eye movement change index value according to the analysis results of the changes in eye movement.

[0028] The analysis of pulse rate changes is performed, for example, as follows: For each frame image, the facial region is identified within the frame image. Then, the change in the G color of the facial surface is analyzed using a pre-trained image analysis model that captures the numerical value of the facial color information (G in RGB). By arranging the results along the time axis, a waveform representing the change in color information is formed, and the pulse rate is identified from this waveform. A person's pulse rate increases when they are nervous and decreases when they are calm. The bioresponse analysis unit 213 can calculate a pulse rate change index value according to the analysis results of the pulse rate changes.

[0029] The analysis of changes in facial movement is performed, for example, as follows: For each frame image, the facial region is identified within the frame image, and the orientation of the face is analyzed to determine where the user is looking. For example, it is analyzed whether the user is looking at the face of the speaker currently displayed, the shared document currently displayed, or looking off-screen. It may also be analyzed whether the facial movement is large or small, and whether the movement is frequent or infrequent. Facial movement and eye movement may also be analyzed together. For example, it may be analyzed whether the user is looking straight at the face of the speaker currently displayed, looking upwards or downwards, or looking at it from an angle. The bio-response analysis unit 213 can calculate a facial orientation change index value according to the analysis results of changes in facial orientation.

[0030] The analysis of the content of speech is performed, for example, as follows: The bioreaction analysis unit 213 converts the speech into a string by performing known speech recognition processing on the speech for a specified time (for example, a time of about 30 to 150 seconds), and then removes unnecessary words that represent the conversation, such as particles and articles, by performing morphological analysis on the string. Then, it vectorizes the remaining words and analyzes whether a positive or negative emotional change has occurred, and to what extent the emotional change has occurred, and can calculate a speech content index value according to the analysis results.

[0031] Voice quality analysis is performed, for example, as follows: The bioreaction analysis unit 12 identifies the acoustic characteristics of the voice by performing known voice analysis processing on the voice for a specified time (for example, a time of about 30 to 150 seconds). Based on these acoustic characteristics, it analyzes whether a positive or negative voice quality change has occurred, and to what degree the voice quality change has occurred, and can calculate a voice quality change index value according to the analysis results.

[0032] The bioresponse analysis unit 213 calculates a bioresponse index value using at least one of the facial expression change index value, eye gaze change index value, pulse rate change index value, face orientation change index value, speech content index value, and voice quality change index value calculated as described above. For example, the bioresponse index value can be calculated by weighting the facial expression change index value, eye gaze change index value, pulse rate change index value, face orientation change index value, speech content index value, and voice quality change index value.

[0033] The estimation unit 214 estimates the patient's mental illness. The estimation unit 214 can estimate the mental illness in accordance with the above-mentioned responses from the patient and changes in biological responses (biological response index values). The estimation unit 214 can estimate the mental illness the patient is suffering from by providing the received responses and detected changes in biological responses to the disease model stored in the learning model memory unit 231.

[0034] Furthermore, the estimation unit 214 can determine the probability that a patient has multiple mental illnesses based on the confidence level of the inferences made by the disease model. Specifically, it utilizes the probability of belonging to each class of mental illness calculated in the output layer of the disease model. Typically, a disease model receives input data (patient responses and biometric values) and probabilistically outputs which class of mental illness the data belongs to. For example, the probability of having multiple mental illnesses is calculated as follows: 70% for depression, 20% for bipolar disorder, and 10% for schizophrenia. Note that confidence level may also be used as "probability." In other words, the probability can be defined as the non-linear likelihood of having an illness.

[0035] The estimation unit 214 can evaluate the certainty of the diagnosis based on the estimated probability of the mental disorder. The reliability of the estimation result can be evaluated by the absolute value of the probability of the most likely mental disorder. For example, if the probability of depression is 90% or higher, it can be judged that there is a very high possibility of depression, while if the probability of the most likely mental disorder is around 50%, it can be judged that the certainty of the diagnosis is not very high.

[0036] Furthermore, the estimation unit 214 can suggest the possibility of comorbidities based on the estimated probability of mental illness. If the probability of suffering from two or more mental illnesses is high, it may suggest the possibility of co-occurrence of those illnesses.

[0037] The diagnostic result input unit 215 receives the results of the differential diagnosis of the patient's mental illness made by the physician. The differential diagnosis by the physician is made based on the above response. The diagnostic result input unit 215 can receive the diagnostic results from the physician terminal 1.

[0038] The misdiagnosis possibility determination unit 216 determines the possibility of misdiagnosis in the differential diagnosis results received by the diagnosis result input unit 215. The misdiagnosis possibility determination unit 216 can determine the possibility of misdiagnosis based on whether the estimated results of the mental illness by the estimation unit 214 match the diagnosis results received by the diagnosis result input unit 215. The misdiagnosis possibility determination unit 216 can determine the probability of misdiagnosis according to the probability of each mental illness estimated by the estimation unit 214.

[0039] The misdiagnosis possibility determination unit 216 may infer the possibility of misdiagnosis (or the possibility of a correct diagnosis) by providing the judgment model stored in the learning model memory unit 231 with the estimated result of the mental illness by the estimation unit 214 and the diagnosis result by the physician. The misdiagnosis possibility determination unit 216 may also infer the possibility of misdiagnosis (or the possibility of a correct diagnosis) by providing the judgment model stored in the learning model memory unit 231 with the patient's response, the detected change in biological response, and the diagnosis result by the physician.

[0040] The misdiagnosis possibility determination unit 216 can transmit the possibility of misdiagnosis to the physician's terminal 1 to warn the physician of the possibility of misdiagnosis. The misdiagnosis possibility determination unit 216 may also transmit the possibility of misdiagnosis to the terminal of, for example, the administrator of the medical institution. In this case, an administrator storage unit is provided for each physician to store information indicating the administrator, etc. If the possibility of misdiagnosis exceeds a predetermined value, or regardless of the degree of possibility of misdiagnosis, the administrator, etc. corresponding to the physician can be read from the administrator storage unit and the possibility of misdiagnosis can be transmitted to the read administrator. The misdiagnosis possibility determination unit 216 may also transmit the possibility of misdiagnosis to the patient's terminal 3 to notify the patient.

[0041] The everyday conversation processing unit 217 engages in conversation (everyday conversation) with the patient. The everyday conversation processing unit 217 can achieve natural conversation with the patient by generating appropriate responses according to the patient's utterances, for example, using a large-scale language model. The large-scale language model used in this embodiment is a model that learns from a large amount of text data to understand the context and meaning of language and can generate natural, human-like sentences according to the given context. Specifically, large-scale language models such as the GPT (Generative Pre-trained Transformer) series, BERT (Bidirectional Encoder Representations from Transformers) series, XLNet, and ELMO (Embeddings from Language Models) can be used.

[0042] The everyday conversation processing unit 217 generates the content to be conveyed to the patient using a large-scale language model, based on the content of the patient's speech (which can be obtained by analyzing the speech content extracted from the audio received from the video image on the patient terminal 3). The specific process can be thought of as follows: (1) Convert the patient's speech into text data. (2) The converted text data is tokenized to conform to the input format for the large-scale language model. (3) The tokenized data is input into a large-scale language model to generate appropriate response sentences according to the context. (4) The generated response sentence is converted into speech data using a speech synthesis engine and output to the patient.

[0043] The everyday conversation processing unit 217 can generate conversational content to be conveyed to the patient by providing a large-scale language model with prompts that include, for example, a history of the patient's utterances (which may be the entire history or a constant for immediate neighbors) and instructions to generate utterances to be spoken to the patient in accordance with those utterances. Specific examples of prompts include the following:

[0044] Example prompt: The following is a record of conversations with the patient. Patient: "Hello." System: "Hello. It's warm today, isn't it?" Patient: "Oh, yes, it might be warm." Next, please think of what the system should say to the patient.

[0045] The everyday conversation processing unit 217 can obtain useful information about the patient's mental state by conducting the above-described conversation with the patient and analyzing its content. Furthermore, the everyday conversation processing unit 217 can obtain more detailed information by analyzing not only the patient's responses but also the reaction time to the response, the choice of words in the response, the speed and intonation of speech, and other factors.

[0046] <Operation> Figure 4 is a diagram illustrating the operation of the management server 2.

[0047] The management server 2 presents the patient with questions to differentiate mental illnesses (S301), receives the patient's answers to the questions (S302), acquires video footage of the patient engaging in everyday conversation (S303), analyzes the acquired video footage to detect changes in biological responses (S304), and feeds the above answers and changes in biological responses to a learning model to estimate the mental illness (S305). The management server 2 also receives the doctor's diagnosis (S306), determines the possibility of misdiagnosis in the diagnosis (S307), and can notify the doctor of the determination (S308).

[0048] As described above, the information processing system of this embodiment can differentiate a patient's mental illness from video footage of the patient, in addition to providing answers to questions. It is known that experts can infer a mental illness from a person's behavior without asking questions such as tests, and adding analysis from video footage is expected to improve the accuracy of the inference. Furthermore, even for patients who arbitrarily give answers that do not reflect their actual situation, it is expected that the possibility of a mental illness can be accurately estimated by also performing differentiation from video footage.

[0049] Furthermore, the information processing system of this embodiment can determine the possibility of misdiagnosis based on the doctor's diagnosis and the computer's reasoning results.

[0050] Although these embodiments have been described above, they are intended to facilitate understanding of the present invention and are not intended to limit its interpretation. The present invention can be modified and improved without departing from its spirit, and equivalents thereof are also included.

[0051] For example, the processing performed by each functional unit of the management server 2 described above may be executed by any of the functional units. Furthermore, different functional units may be added to perform some of the processing performed by each of the functional units described above. Also, the functional units of the management server 2 may be distributed across multiple computers.

[0052] Furthermore, the information stored in each memory unit of the management server 2 may be stored in any of the memory units. That is, the information stored in the multiple memory units mentioned above may be stored in a single memory unit, or a portion of the information stored in one memory unit may be stored in another memory unit.

[0053] <Example 1>

[0054] By establishing specific physiological response indicators that should be given particular importance for each type of mental illness and focusing on analyzing changes in these indicators, more accurate diagnostic support can be provided. Below are examples of representative mental illnesses and the physiological response indicators that should be given importance in their diagnosis.

[0055] (1) Depression In diagnosing depression, it is effective to emphasize the following physiological response indicators. • Facial expressions: The movement of facial muscles is limited, and many expressions are seen that convey sadness or pain. • Gaze: The gaze is directed downwards, and the frequency of eye contact decreases. • Voice quality: The voice tends to lack inflection, be monotonous, and quiet.

[0056] (2) Bipolar disorder In diagnosing bipolar disorder, it is effective to emphasize the following physiological response indicators. • Facial expressions: (Hypomanic state) Facial muscles are highly active, and many expressions of joy and excitement are observed. (Depressed state) Facial muscle movement is limited, and expressions of sadness and pain are frequently observed. • Gaze: (Hypomanic state) Gaze is directed upward, and the frequency of eye contact increases. (Depressed state) The gaze is directed downwards, and the frequency of eye contact decreases. • Voice quality: (Hypomanic state) The voice tends to have rich intonation and become louder. (Depressed state) The voice tends to have little intonation, becoming monotonous and quiet.

[0057] (3) Schizophrenia In diagnosing schizophrenia, it is effective to emphasize the following physiological response indicators. • Facial expressions: Due to the flattening of emotions, facial expressions become less varied. • Gaze: The gaze may be fixed on one point, or conversely, the gaze may be unfocused. • Content of statements: Often incoherent and reflecting a decline in the ability to assess reality.

[0058] (4) Panic disorder In diagnosing panic disorder, it is effective to emphasize the following physiological response indicators. • Pulse: Marked tachycardia is observed during panic attacks. • Facial expressions: During panic attacks, they may show facial expressions that indicate intense fear. • Voice quality: During panic attacks, the voice tends to tremble and speech becomes stammery.

[0059] (5) Dementia In diagnosing dementia, it is effective to emphasize the following biological response indicators. • Content of statements: Statements regarding time and place may be inaccurate, reflecting memory impairment or disorientation. • Voice quality: There is a tendency for speech speed to slow down and voice volume to decrease. • Facial expressions: Due to the flattening of emotions, facial expressions become less varied.

[0060] It should be noted that the biological response indicators exemplified above are merely representative examples, and each biological response indicator is only one example.

[0061] <Modification 2>

[0062] When estimating a mental illness or determining the likelihood of misdiagnosis, attribute information such as the patient's age and gender can be considered. Examples of attribute information that should be considered during estimation and diagnosis are provided below.

[0063] (1) Age The prevalence and manifestation of symptoms of mental illnesses can vary with age. For example, dementia is more common in the elderly and rare in younger people. Also, childhood autism spectrum disorder can exhibit behavioral characteristics different from those of adults. Therefore, considering the patient's age allows for evaluation using diagnostic criteria more appropriate to that age group.

[0064] The estimation unit 214 may also provide the patient's age as one of the input pieces of information to the disease model stored in the learning model memory unit 231. This makes it possible to estimate mental illness according to age.

[0065] The misdiagnosis probability determination unit 216 may also provide the patient's age as one of the input pieces of information to the determination model stored in the learning model memory unit 231. This makes it possible to determine the probability of misdiagnosis while taking age into consideration.

[0066] (2)Gender It is known that some mental illnesses show gender differences in prevalence and symptoms. For example, the lifetime prevalence of depression is higher in women, while the lifetime prevalence of alcoholism is higher in men. Eating disorders are also more common in women. Therefore, considering the patient's gender allows for a diagnosis that takes into account gender-specific symptoms and prevalence.

[0067] The estimation unit 214 may also provide the patient's gender as one of the input pieces of information to the disease model stored in the learning model memory unit 231. This makes it possible to estimate mental illness according to gender.

[0068] The misdiagnosis probability determination unit 216 may also provide the patient's gender as one of the input pieces of information to the determination model stored in the learning model memory unit 231. This makes it possible to determine the probability of misdiagnosis while taking gender into consideration.

[0069] (3) Medical history A patient's medical history, including mental and physical illnesses, significantly influences the interpretation and diagnosis of their current symptoms. For example, if a patient with a history of depression re-exhibits depressive symptoms, it is highly likely to be a relapse of depression rather than simply a stress response. Furthermore, a history of thyroid disease or cerebrovascular disorders can contribute to mental symptoms. Therefore, considering a patient's medical history can lead to a more accurate diagnosis.

[0070] The estimation unit 214 may also provide the patient's medical history as one of the input pieces of information to the disease model stored in the learning model memory unit 231. This makes it possible to estimate mental illnesses based on the patient's medical history.

[0071] The misdiagnosis probability determination unit 216 may also provide the patient's medical history as one of the input pieces of information to the determination model stored in the learning model memory unit 231. This makes it possible to determine the probability of misdiagnosis while taking the patient's medical history into consideration.

[0072] In this embodiment, the information processing system may include an attribute information storage unit in which the management server 2 manages attribute information such as the patient's age, gender, and medical history. The estimation unit 214 and the misdiagnosis probability determination unit 216 can read the necessary attribute information from the attribute information storage unit and use it for estimation and determination processing.

[0073] Furthermore, attribute information may be entered by doctors or patients, or it may be automatically extracted from medical questionnaires, medical records, etc.

[0074] <Variation 3>

[0075] If the misdiagnosis probability determination unit 216 determines that there is a high probability of misdiagnosis, it can issue alerts to doctors and patients in various ways.

[0076] (1) Display of alert messages on the physician's terminal The misdiagnosis probability determination unit 216 may display an alert message on the physician terminal 1 if it determines that there is a high probability of misdiagnosis. The alert message may include a statement that there is a high probability of misdiagnosis and the reason for this (for example, that the disease estimated from the patient's responses and physiological responses differs from the physician's diagnosis). The alert message may also include wording that encourages reconsideration or suggests additional tests or consultations.

[0077] (2) Inclusion of the possibility of misdiagnosis in the diagnostic report The misdiagnosis probability determination unit 216 may automatically include a note indicating a high probability of misdiagnosis in the diagnostic report prepared by the physician if it determines that there is a high probability of misdiagnosis. This allows the physician to reaffirm the possibility of misdiagnosis when reviewing the diagnostic report. Furthermore, the information on the possibility of misdiagnosis included in the diagnostic report can be used to determine subsequent treatment plans and to share information with other medical staff.

[0078] (3) Sending alert messages to administrators, etc. The misdiagnosis probability determination unit 216 may send an alert message to the administrator of the medical institution or the head of the relevant clinical department if it determines that there is a high probability of misdiagnosis. This allows for quality control of diagnoses and, if necessary, guidance and support for physicians. The alert message may include information such as the high probability of misdiagnosis, the reason for the high probability, and the names of the physicians and patients involved.

[0079] (4) Display of alert messages on the patient's terminal The misdiagnosis possibility determination unit 216 may display an alert message on the patient terminal 3 if it determines that there is a high possibility of misdiagnosis. The alert message may include information indicating that the current diagnosis needs to be re-examined and encouraging additional tests or a consultation with a doctor. However, sufficient consideration must be given to the content and wording of the message so as not to cause anxiety to the patient.

[0080] (5) Recording and analysis of alert history The misdiagnosis possibility determination unit 216 may record and analyze the history of alerts that occur. The alert history may include information such as the date and time the alert occurred, the name of the doctor or patient involved, and the reason for the possibility of misdiagnosis. By analyzing this information, it can be used to identify issues for improving the quality of diagnosis and to provide feedback to doctors.

[0081] The alert methods exemplified above may be used individually or in combination. The selection of alert methods can be made flexibly according to the policies of the medical institution, the characteristics of the medical department, the proficiency level of individual physicians, and other factors.

[0082] <Disclosure Items> Furthermore, this disclosure also includes the following configurations. [Item 1] A response input unit that receives patient responses to questions used to differentiate mental illnesses, A video acquisition unit that acquires video images of the aforementioned patient, A biological response detection unit that analyzes the aforementioned moving image to detect changes in the patient's biological response, An estimation unit estimates the mental illness the patient is suffering from by providing the received response and the detected change in the biological response to a learning model that has learned the aforementioned response and the changes in the biological response and the mental illness; An information processing system characterized by comprising the following features. [Item 2] The information processing system described in item 1, A diagnostic result input unit that receives the diagnostic result of the patient's mental illness from a medical professional based on the above response, A misdiagnosis possibility determination unit that determines whether the estimated result of the mental disorder by the estimation unit matches the diagnosis result, An information processing system characterized by comprising the following features. [Item 3] The information processing system described in item 1, The estimation unit calculates the probability that the patient suffers from multiple mental illnesses. An information processing system characterized by the following. [Item 4] The information processing system described in item 1, The aforementioned video acquisition unit acquires the video footage of the patient engaging in everyday conversation. An information processing system characterized by the following. [Item 5] The information processing system described in item 4, A daily conversation processing unit for conducting daily conversations with the patient, comprising a daily conversation processing unit that generates a second utterance to be conveyed to the patient in response to a first utterance from the patient, The video acquisition unit acquires the video footage of the conversation between the patient and the daily conversation processing unit. An information processing system characterized by the following. [Item 6] A step to receive input from patients regarding questions for differentiating mental illnesses, The steps include: acquiring video footage of the aforementioned patient; The steps include: analyzing the aforementioned video image to detect changes in the patient's biological response, A step of estimating the mental illness the patient is suffering from by providing the received response and the detected change in the biological response to a learning model that has learned the aforementioned response and the changes in the biological response and the mental illness; An information processing method characterized by a computer executing the following. [Item 7] A step to receive input from patients regarding questions for differentiating mental illnesses, The steps include: acquiring video footage of the aforementioned patient; The steps include: analyzing the aforementioned video image to detect changes in the patient's biological response, A step of estimating the mental illness the patient is suffering from by providing the received response and the detected change in the biological response to a learning model that has learned the aforementioned response and the changes in the biological response and the mental illness; A program that causes a computer to execute something. [Explanation of symbols]

[0083] 1. Doctor's terminal 2 Management Server 3. Patient terminal

Claims

1. A response input unit that receives patient responses to questions used to differentiate mental illnesses, A video acquisition unit that acquires video images of the aforementioned patient, A biological response detection unit that analyzes the aforementioned moving image to detect changes in the patient's biological response, An estimation unit estimates the mental illness the patient is suffering from by providing the received response and the detected change in the biological response to a learning model that has learned the aforementioned response and the changes in the biological response and the mental illness; An information processing system characterized by comprising the following features.

2. The information processing system according to claim 1, A diagnostic result input unit that receives input of the diagnosis result of the patient's mental illness by a medical professional based on the above response, A misdiagnosis possibility determination unit that determines whether the estimated result of the mental disorder by the estimation unit matches the diagnosis result, An information processing system characterized by comprising the following features.

3. The information processing system according to claim 1, The estimation unit calculates the probability that the patient suffers from multiple mental illnesses. An information processing system characterized by the following.

4. The information processing system according to claim 1, The aforementioned video acquisition unit acquires the video footage of the patient engaging in everyday conversation. An information processing system characterized by the following.

5. The information processing system according to claim 4, A daily conversation processing unit for conducting daily conversations with the patient, comprising a daily conversation processing unit that generates a second utterance to be conveyed to the patient in response to a first utterance from the patient, The video acquisition unit acquires the video footage of the conversation between the patient and the daily conversation processing unit. An information processing system characterized by the following.

6. A step to receive input from patients regarding questions for differentiating mental illnesses, The steps include: acquiring video footage of the aforementioned patient; The steps include: analyzing the aforementioned video image to detect changes in the patient's biological response, A step of estimating the mental illness the patient is suffering from by providing the received response and the detected change in the biological response to a learning model that has learned the aforementioned response and the changes in the biological response and the mental illness; An information processing method characterized by a computer executing the following.

7. A step to receive input from patients regarding questions for differentiating mental illnesses, The steps include: acquiring video footage of the aforementioned patient; The steps include: analyzing the aforementioned video image to detect changes in the patient's biological response, A step of estimating the mental illness the patient is suffering from by providing the received response and the detected change in the biological response to a learning model that has learned the aforementioned response and the changes in the biological response and the mental illness; A program that causes a computer to execute something.

Images