Medical support system, medical support program, and medical support method

Abstract

To provide a medical support system capable of outputting reasons for extracting given medical information from text information including healthcare-related medical information.SOLUTION: A medical support system 100 for automatically filling healthcare-related medical forms is provided, the system comprising a natural language comprehension module 112 machine-trained to output reasons for extracting given medical information from text information including healthcare-related medical information.SELECTED DRAWING: Figure 25

Description

[Technical Field]

[0001] The present invention relates to a medical support system, a medical support program, and a medical support method. [Background technology]

[0002] When medical personnel such as ambulance personnel, doctors, and nurses check a patient's condition, they may provide verbal explanations. However, they are also required to record the patient's condition on a prescribed form.

[0003] For example, a comprehensive emergency service support system has been disclosed in which information on injured persons received at a firefighting command center is transmitted from an emergency server via a communication line to a terminal used by an emergency responder (Patent Documents 1 and 2). [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Application Laid-Open No. 2014-130516 [Patent Document 2] Patent Publication No. 2021-93228 Summary of the Invention [Problem to be solved by the invention]

[0005] When entering medical information into a form of a specified format, it is necessary to extract the medical information contained in the explanation, embed it in the correct place on the form, and output it accurately and quickly.

[0006] In particular, in the case of illnesses that require urgent treatment, such as stroke and myocardial infarction, it is necessary to fill out a prescribed form so that medical information such as the condition that has been explained orally by medical personnel can be conveyed quickly and accurately.

[0007] Furthermore, when specific medical information is extracted from text information containing medical information related to medical care, it is desirable to also be able to confirm the reasons for the extraction, such as the location from which the medical information was extracted. [Means for solving the problem]

[0008] One aspect of the present invention is a medical support system that automatically fills in medical forms related to medical care, characterized by having a natural language understanding module that has been machine-trained to output reasons for extracting specified medical information from text information containing medical information related to medical care.

[0009] Another aspect of the present invention is a medical support program that automatically fills in medical forms related to medical care, characterized in that the program causes a computer to function as a machine-learned natural language understanding module that outputs reasons for extracting specified medical information from text information containing medical information related to medical care.

[0010] Another aspect of the present invention is a medical support method for automatically filling out a medical form related to medical care, characterized in that a computer functioning as a natural language understanding module that has been machine-learned to output reasons for extracting specified medical information from text information including medical information related to medical care is caused to output reasons for extracting the specified medical information from text information including medical information related to medical care.

[0011] Here, it is preferable that the natural language understanding module is machine-learned to use text information containing medical information related to medical care as input data, and the reason for extracting the specified medical information to be output from the text information as training data, and to use supervised learning data that combines these pieces of information so that when the input data is input, the reason for extracting the specified medical information, which is the training data, is output.

[0012] It is also preferable to include a form embedding module that embeds the medical information output from the natural language understanding module into the medical form and outputs it.

[0013] It is also preferable to have a speech-to-text conversion module that generates the text information from speech information that includes medical information related to medical care. [Effects of the Invention]

[0014] According to the present invention, it is possible to provide a medical support system, a medical support program, and a medical support method that output the reason for extracting predetermined medical information from text information including medical information related to medical care. Other objects of the embodiments of the present invention will become apparent by referring to the entire specification. [Brief explanation of the drawings]

[0015] [Figure 1] 1 is a diagram showing a configuration of a medical support system according to an embodiment of the present invention. [Figure 2] FIG. 2 is a diagram illustrating a configuration of a server according to an embodiment of the present invention. [Figure 3] FIG. 2 is a diagram illustrating a configuration of a client according to an embodiment of the present invention. [Figure 4] 1 is a functional block diagram showing a configuration of a medical support system according to an embodiment of the present invention. [Figure 5] FIG. 2 illustrates a speech-to-text conversion module according to an embodiment of the present invention. [Figure 6] FIG. 2 illustrates a speech-to-text conversion module according to an embodiment of the present invention. [Figure 7] FIG. 2 illustrates a speech-to-text conversion module according to an embodiment of the present invention. [Figure 8] FIG. 2 illustrates a speech-to-text conversion module according to an embodiment of the present invention. [Figure 9] FIG. 2 is a diagram illustrating a medical form generation module according to an embodiment of the present invention. [Figure 10]FIG. 3 is a diagram showing an example of text information according to the embodiment of the present invention. [Figure 11] FIG. 10 is a diagram showing an example of form information according to the embodiment of the present invention. [Figure 12] FIG. 10 is a diagram illustrating an example of output data according to the embodiment of the present invention. [Figure 13] 1 is a flowchart showing a learning method for a natural language understanding module according to an embodiment of the present invention. [Figure 14] FIG. 2 is a diagram illustrating an example of the configuration of learning data for a natural language understanding module according to an embodiment of the present invention. [Figure 15] FIG. 2 is a diagram illustrating an example of the configuration of learning data for a natural language understanding module according to an embodiment of the present invention. [Figure 16] FIG. 2 is a diagram illustrating a learning method for a natural language understanding module according to an embodiment of the present invention. [Figure 17] FIG. 2 is a diagram illustrating a learning method for a natural language understanding module according to an embodiment of the present invention. [Figure 18] 10 is a flowchart illustrating a learning method for a form embedding module according to an embodiment of the present invention. [Figure 19] 10 is a diagram showing an example of the configuration of learning data for a form embedding module according to an embodiment of the present invention; FIG. [Figure 20] FIG. 10 is a diagram illustrating a learning method for a form embedding module according to an embodiment of the present invention. [Figure 21] 10 is a flowchart illustrating a training method for a medical form generation module according to an embodiment of the present invention. [Figure 22] FIG. 1 is a diagram illustrating a method for improving a medical support system according to an embodiment of the present invention. [Figure 23] 1 is a flowchart showing a learning method for a natural language understanding module according to an embodiment of the present invention. [Figure 24] FIG. 2 is a diagram illustrating an example of the configuration of learning data for a natural language understanding module according to an embodiment of the present invention. [Figure 25]FIG. 2 is a diagram illustrating a learning method for a natural language understanding module according to an embodiment of the present invention. [Figure 26] FIG. 3 is a diagram illustrating an example of output information of a natural language understanding module according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0016] [System Configuration] As shown in Fig. 1, a medical support system 100 according to an embodiment of the present invention includes a server 102 and a client 104. There may be one or more clients 104. The server 102 and the client 104 are connected to each other via an information and communication network 106 such as the Internet so that they can exchange information with each other.

[0017] The information and communication network 106 is not limited to the Internet, but may be anything that can connect the server 102 and the client 104 to each other so that they can communicate with each other. For example, it may be a dedicated line, a public line (telephone line, mobile communication line, etc.), a wired LAN (Local Area Network), a wireless LAN, etc., or it may be a combination of the Internet and these.

[0018] As shown in FIG. 2, the server 102 includes a processing unit 10, a memory unit 12, an input unit 14, an output unit 16, and a communication unit 18. The processing unit 10 includes a means for performing arithmetic processing, such as a CPU. The processing unit 10 executes a medical support server program stored in the memory unit 12, thereby realizing a process for accurately and quickly entering medical information on a form of a predetermined format in the medical support system 100 of this embodiment. The memory unit 12 includes a storage means, such as a semiconductor memory or a memory card. The memory unit 12 is accessible and connected to the processing unit 10 and stores the medical support server program and information required for the process. The input unit 14 includes a means for inputting information. The input unit 14 includes, for example, a keyboard, a touch panel, buttons, etc. for receiving input from an administrator. The output unit 16 includes a means for outputting the processing results of the server 102, such as a user interface screen (UI) for receiving input information from the administrator. The output unit 16 includes, for example, a display for presenting images to the administrator. The communication unit 18 includes an interface for communicating information with the client 104 via the information and communication network 106. The communication by the communication unit 18 may be wired or wireless.

[0019] The server 102 accepts access via the information and communication network 106 from the clients 104 of each user who receives medical support services, and provides the services via the information and communication network 106 to each user.

[0020] As shown in FIG. 3 , the client 104 includes a processing unit 20, a storage unit 22, an input unit 24, an output unit 26, and a communication unit 28. The client 104 is also referred to as a communication terminal. The processing unit 20 includes a means for performing arithmetic processing, such as a CPU. The processing unit 20 executes a medical support client program stored in the storage unit 22 to function as a client terminal in the medical support system 100 of this embodiment. The storage unit 22 includes storage means, such as a semiconductor memory or a memory card. The storage unit 22 is accessible to the processing unit 20 and stores the medical support client program and information required for its processing. The input unit 24 includes a means for inputting information. The input unit 24 includes, for example, a keyboard, a touch panel, buttons, a motion sensor, a microphone, etc., for receiving input from a user. The output unit 26 includes a means for outputting information required for processing by the client 104, such as a screen for receiving input information from a user or a display for displaying image information such as a user interface screen (UI). The communication unit 28 includes an interface for communicating information with the server 102 via the information and communication network 106. The communication by the communication unit 28 may be wired or wireless.

[0021] Various information processing devices capable of executing a client program for providing medical support services can be used as the client 104. For example, the client 104 can be a stationary or portable personal computer (PC), a tablet computer, a smartphone, a mobile phone terminal, a PHS (Personal Handy-phone System) terminal, a personal digital assistant (PDA), a multi-function television receiver with information processing capabilities (so-called smart TV), etc.

[0022] In order to ensure secure access from the client 104 to the server 102, processing such as user authentication may be performed when the client 104 accesses the server 102. The user authentication processing may be, for example, a process in which a user ID and password are input from the input unit 24 of the client 104, and if the inputted user ID and password match a combination of the user ID and password pre-stored in the storage unit 12 of the server 102, access from the client 104 to the server 102 is authenticated.

[0023] [Medical Support Processing] 4 shows a system configuration diagram of the medical support system 100 according to this embodiment. As shown in FIG. 4, the medical support system 100 includes a voice-to-text conversion module 110 and a medical form generation module 112.

[0024] The speech-to-text conversion module 110 is a module that performs processing to extract medical text information from voice information obtained from medical professionals, patients, etc. The medical form generation module 112 is a module that extracts and outputs medical information corresponding to a predetermined form for medical information based on the medical text information.

[0025] As shown in Fig. 4, the medical form generation module 112 can be configured to include a natural language understanding module 112a and a form embedding module 112b. The natural language understanding module 112a is a module that performs processing to extract medical information necessary for generating a medical form from medical text information. The form embedding module 112b is a module that performs processing to embed the medical information output from the natural language understanding module 112a into a predetermined medical form and output the medical form. The natural language understanding module 112a and the form embedding module 112b may be independent modules, or may be integrated into a single module.

[0026] A medical form is a document such as a medical record used in a medical institution. However, a medical form is not limited to a medical record, but may be any form that includes medical information corresponding to predetermined items related to the patient's condition and medical care. A medical form may be electronic data, or may be printed out onto a paper medium from the output unit 16 of the server 102 or the output unit 26 of the client 104.

[0027] The speech-to-text module 110 and the medical form generation module 112 will now be described.

[0028] [Speech-to-text module] As shown in Figure 5, the speech-to-text conversion module 110 receives voice information from medical professionals involved in medical care, such as doctors, nurses, clinic staff, and emergency personnel, about the medical condition and treatment instructions of a patient, and converts the voice information into text information.

[0029] The speech-to-text conversion module 110 can apply existing speech-to-text conversion processing algorithms and AI modules.

[0030] For example, the speech-to-text conversion module 110 can be configured by applying statistical speech recognition technology. Statistical speech recognition technology consists of a feature extractor that extracts features from speech, a decoder, and a speech recognition model (a model that integrates an acoustic model, a pronunciation dictionary, and a language model). The feature extractor analyzes a speech signal sequence (e.g., time-series speech information) and extracts a feature vector with several tens of dimensions called a logarithmic Mel-frequency spectrum or a Mel-frequency cepstrum. The decoder searches for and outputs the most likely word sequence based on the feature vector. The speech recognition model integrates an acoustic model, a pronunciation dictionary, and a language model. The acoustic model is a model that obtains the most likely phoneme sequence from a feature vector sequence. The acoustic model typically uses an HMM, a non-deterministic state transition model with finite states. The pronunciation dictionary is a probabilistic model that generates words from a phoneme sequence. The language model is a model that obtains the most likely word sequence from words obtained from the pronunciation dictionary. In other words, the language model is a model that determines the language-likelihood of a given symbol sequence (word sequence).

[0031] Furthermore, for example, the speech-to-text conversion module 110 can be configured by applying an AI module based on deep learning. The AI ​​module can be configured in various ways, such as CNN or RNN, but it is often used to learn a deep neural network (DNN) that outputs a vector whose elements are the probability of belonging to an HMM state when a feature vector is input.

[0032] Specifically, for example, speech information from a single medical professional and text information transcribed from the speech information are combined as training data to create supervised learning data, and the speech data is input into an AI module composed of a neural network, which is trained to produce output that matches the training data, thereby forming the speech-to-text conversion module 110.

[0033] Here, the speech-to-text conversion module 110 can be configured to convert speech information from a single medical professional into text information, as shown in Fig. 5. Fig. 5 shows an example in which speech information related to medical care spoken by Doctor 1 is input and converted into text information. Alternatively, as shown in Fig. 6, the speech-to-text conversion module 110 may be configured to convert speech information in the form of a conversation between multiple medical professionals or between a medical professional and a patient into text information. Fig. 6 shows an example in which speech information related to medical care spoken by Doctor 1, Doctor 2, and Nurse 1 is input and converted into text information in the form of a chat conversation.

[0034] 7, in order to make it easier to recognize medical terminology, a medical language model and a speech recognition model may be integrated for processing. For example, the speech-to-text conversion module 110 may be constructed using existing techniques such as shallow fusion, cold fusion, deep fusion, and density ratio approach.

[0035] Additionally, free text entries containing only medical terms and abbreviations can be converted into structured data that is standardized into a predefined format related to a particular medical condition by relying on key-value pairs containing medical terms and abbreviations used to train the algorithms built into the medical form generation module 112, described below.

[0036] In noisy environments such as hospitals and clinics, it is preferable to apply a noise reduction algorithm as shown in Figure 8. The noise reduction algorithm plays an important role in improving the accuracy of the processing in the speech-to-text conversion module 110. In particular, in medical speech-to-text conversion processing, clarity of the speech information is essential due to the importance and sensitivity of the content.

[0037] Beneficial noise reduction algorithms for the speech-to-text module 110 include spectral subtraction, Wiener filtering, beamforming and microphone arrays, noise suppression using deep neural networks (DNNs), recurrent neural networks (RNNs) and long short-term memory (LSTM), non-negative matrix factorization (NMF), statistical model-based methods, binaural processing, and end-to-end noise reduction using autoencoders.

[0038] Spectral subtraction is a classic technique that involves subtracting an estimate of the noise spectrum from the noisy speech spectrum. It works by estimating the noise spectrum of non-speech segments and then subtracting this estimated noise from the entire signal.

[0039] Wiener filtering is a statistically based approach. The Wiener filter minimizes the mean squared error between the estimated clean signal and the actual clean signal. It is particularly effective when the characteristics of the noise are known.

[0040] Beamforming and microphone arrays are techniques for reducing noise by using multiple microphones to capture sound. The signals from the multiple microphones are combined so that signals from certain directions are amplified and signals from other directions are suppressed. This allows for a focus on the speaker's voice and minimizes background noise.

[0041] Deep neural network (DNN) noise suppression is a method of noise reduction using a deep learning model. By applying a DNN model, it can be trained to distinguish between clean speech and noise. Once trained, the model can suppress noise from new speech data. With enough training data, DNNs can perform very well even in complex noisy environments.

[0042] Techniques that apply recurrent neural networks (RNNs) and long short-term memory (LSTMs) have the ability to memorize past information and can be used for noise reduction. They are particularly effective for time-series information such as audio signals, where context and time-series information are useful for identifying and removing noise.

[0043] Non-negative matrix factorization (NMF) decomposes the spectrogram of a noise signal into a set of basis spectra and their activity indices. Speech and noise signals can be separated by assuming that they can be approximated by different sets of basis spectra.

[0044] Statistical model-based methods estimate the statistical properties of speech and noise. Based on statistical models, noise is suppressed and speech is enhanced. Voice activity detection (VAD) is a technique that detects the presence or absence of human speech. During periods when speech is not detected, the system captures samples of background noise to more effectively suppress noise during speech segments. Binaural processing is a technique based on human auditory processing, using two or more microphones (simulating ears) to identify and suppress noise sources. End-to-end denoising using an autoencoder is a technique that uses an autoencoder trained to directly map noisy speech to clean speech. The encoder compresses the noisy speech into a compact representation, and the decoder reconstructs clean speech from this representation.

[0045] In addition, sensitive information such as personal information may be anonymized from text information indicating medical information. Sensitive information may be defined as appropriate. For example, any type of information that is considered sensitive or capable of identifying an individual may be defined as sensitive information. In medical records, sensitive information typically includes a patient's name, address, telephone number, social security number, and other unique identifiers.

[0046] One way to de-identify this sensitive information is through data tokenization, which breaks down text into individual words or tokens and then applies a substitution process to recognize and de-identify specific data points. De-identification can also be achieved through named entity recognition (NER), a natural language processing technique that identifies named entities in text and classifies them into predefined categories, such as names, organizations, or locations. Because traditional NER models may not capture certain medical terminology and nuances, a pre-trained NER model specifically fine-tuned for medical records is preferable.

[0047] After identifying sensitive information, several methods can be applied to anonymize it. One anonymization method is to completely delete the sensitive information by re-editing the text information. Another anonymization method is to pseudonymize the sensitive information contained in the text information. When pseudonymizing a person's name contained in the text information, for example, if a word contained in the text information, such as "Taro Tanaka," is recognized as a person's name, the word "Taro Tanaka" is replaced with information indicating the person's name, such as "PERSON_NAME." This method is particularly effective when it is necessary to maintain the data structure of the text information. Another anonymization method is to generalize the sensitive information contained in the text information. For example, this method replaces specific items of information with more general information, such as replacing dates contained in the text information with only the year.

[0048] Specifically, regular expressions can be used to detect typical patterns in specific sensitive information contained in text information, such as dates of birth, telephone numbers, etc. Alternatively, a dictionary containing known names, place names, or specific terms used within an organization such as a medical institution can be created, and the dictionary can be used to extract known terms and replace them with other specific terms.

[0049] Machine learning and deep learning may also be used in the anonymization process. By applying machine learning and deep learning models, text information related to medical information and information in which specific sensitive information contained in the text information, such as dates of birth and phone numbers, has been anonymized are combined as training data, and machine learning is performed so that when the text information is input, the training data is output. This makes it possible to build an AI model for anonymization processing that outputs anonymized specific sensitive information contained in the text information, such as dates of birth and phone numbers. By applying machine learning and deep learning, it becomes possible to detect and anonymize inconspicuous text patterns and relationships contained in text information that may lead to deanonymization.

[0050] It is desirable for the anonymization process to maintain consistency. That is, for example, if "John Doe" is replaced with "Patient A" in one record of text information, the same replacement should be consistent across all records. Ideally, the anonymization process should be validated through a human-supervised review process. This allows for the identification of oversights and further refinement of the anonymization process. Continuously updating the model also ensures proper anonymization, especially when new data types or structures are introduced into records.

[0051] [Medical form generation module] As shown in Figure 9, the medical form generation module 112 receives as input text information containing medical information related to medical care and form information related to a medical form into which the medical information is embedded, and performs processing to output a medical form into which the medical information contained in the text information is appropriately embedded.

[0052] The text information input to the medical form generation module 112 is text information including medical-related information obtained from medical professionals. The text information may be text data in the form of a chat by a medical professional, as shown in Fig. 10. Alternatively, the text information may be text data extracted from voice information by the voice-to-text conversion module 110, as described above.

[0053] Furthermore, the form information input to the medical form generation module 112 is information indicating the configuration of a medical form into which medical information extracted from text information is embedded. Medical forms can be, for example, medical records, documents that need to be filled out for each disease or symptom, and the like. Medical forms can be, for example, diagnosis forms, treatment forms, patient follow-up forms, research data collection forms, and the like. Therefore, as shown in FIG. 11, for each type of medical form, the form information can be information that combines a question ID, the content of the question, and the type of answer to the question (information type such as numerical value or text). The medical form generation module 112 extracts appropriate types of information corresponding to the questions included in the form information from the input text information, embeds the information in an appropriate medical form, and outputs the extracted information.

[0054] The form information may be selected and input by a user such as a medical professional based on the patient's condition, medical procedure, etc. Also, the medical form generation module 112 may automatically select appropriate form information from a plurality of pieces of form information set in advance based on text information input to the medical form generation module 112.

[0055] The output data of the medical form generation module 112 is information including question IDs corresponding to the input form information and answers to each question. The format of the output medical form depends on the type of related question. For example, if the question is multiple choice, single choice, input number, free text, etc., the output answer is formatted accordingly. It is preferable that the output data be output in a predetermined format, such as JSON format, which is easy to process later.

[0056] The natural language understanding module 112a included in the medical form generation module 112 is configured to extract and output medical information to be included in the medical form based on the input text information and form information. Fig. 12 shows an example of output data output from the natural language understanding module 112a. The natural language understanding module 112a receives input text information and form information, extracts medical information from the text information for the medical information items required for the medical form specified in the form information, and outputs the medical information. Fig. 12 shows an example of extracted information required to fill in a medical form related to stroke, such as the patient's age, gender, blood pressure, and heart rate.

[0057] <Natural Language Understanding Module> The natural language understanding module 112a included in the medical form generation module 112 can be configured by applying an AI module based on deep learning. The natural language understanding module 112a can be an AI module that applies a large-scale language model (LLM) such as GPT-4 or LLaMA2. The large-scale language model (LLM) is a natural language processing model whose accuracy has been improved by machine learning using a large data set.

[0058] There are several steps involved in training a language model for automatically generating medical forms based on medical information. The training process for building the natural language understanding module 112a will be described below with reference to the flowchart in Figure 13.

[0059] In step S10, a process is performed to prepare training data to be used for training the natural language understanding module 112a. As the training data, a large amount of text information indicating medical records must be collected. In addition, it is necessary to prepare appropriate filled-out medical forms as output for the text information input to the medical form generation module 112.

[0060] The natural language understanding module 112a receives as input data text information containing medical information and form information specifying the medical information to be extracted from the text information required to embed the medical information in a desired medical form, and prepares supervised learning data by combining the medical information to be output when the text information and form information are input as supervised data.

[0061] For example, as shown in Fig. 14, text information including medical information and form information indicating information (items) to be output from the natural language understanding module 112a are used as input data, and the medical information to be output from the natural language understanding module 112a is used as training data, and these are combined to form training data. In the example of Fig. 14, the form information is instruction information that specifies the patient's medical history as an output item. Furthermore, the training data is "diabetes," which is the patient's medical history to be extracted from the text information included in the training data.

[0062] The instruction information may be, for example, information instructing to output items indicating the patient's condition, such as the patient's age, sex, medical history, blood pressure, pulse rate, oxygen saturation level, presence or absence of dyspnea, state of consciousness, changes in electrocardiogram, blood test results, etc. The instruction information may also be information instructing to output items indicating medical procedures, diagnostic and treatment methods, and test items to be performed on the patient, such as injections, diagnostic imaging, and thrombolytic therapy. However, the instruction information is not limited to these, and may be any information indicating the content of information to be extracted from the text information input to the natural language understanding module 112a and output.

[0063] Furthermore, as shown in Fig. 15, the training data may include form information that simultaneously specifies multiple items included in a medical form. Fig. 15 shows an example of form information that includes instruction information that specifies outputting items necessary to generate a medical form related to stroke, such as the patient's age ("Age"), gender ("Gender"), blood pressure ("Blood Pressure"), and pulse rate ("Heart Rate"). The training data includes the items specified in the form information and information that should be extracted from the text information included in the training data corresponding to those items.

[0064] The learning data can be divided into a training set, a validation set, and a test set. The training set is used for learning the natural language understanding module 112a. The validation set is used to validate the learned natural language understanding module 112a. The test set is used for final testing of the validated natural language understanding module 112a. This makes it possible to evaluate the performance of the natural language understanding module 112a on unknown data.

[0065] Furthermore, it is preferable that sensitive information contained in the text information be anonymized as described above in order to ensure privacy, etc.

[0066] In step S12, data preprocessing is performed. The text information input to the natural language understanding module 112a is preferably formatted in a consistent, machine-readable format. Therefore, in this step, the text information is converted into a machine-readable format as needed. For example, depending on the complexity of the medical form, the text information may be structured as pairs of items and their values ​​or as a more complex schema.

[0067] Furthermore, it is preferable that medical terms and abbreviations are converted as pairs of items and values. For example, text information may contain various different words that indicate the same "myocardial infarction," such as "acute myocardial infarction," "AMI," "shinkin kousoku," and "Heart Attack." These words are then used as keys, and these words are associated with "myocardial infarction" and stored in a medical terminology database. Then, when these words are included in the text information, a process is performed to replace them with "myocardial infarction." Similar processes can be applied to other medical terms and abbreviations, replacing words with the same meaning with a common word.

[0068] In step S14, a learning process is performed for the natural language understanding module 112a. The natural language understanding module 112a is trained using the training data prepared in steps S10 and S12. Specifically, text information and form information included in the training data are input to the AI ​​module of the natural language understanding module 112a, and machine learning is performed so that output data, which is training data included in the training data, is output from the natural language understanding module 112a.

[0069] Here, the AI ​​module constituting the natural language understanding module 112a may be a module including a pre-trained AI model such as GPT-3 or GPT-4, for example, which may be fine-tuned to constitute the natural language understanding module 112a.

[0070] 16 shows a learning method for the natural language understanding module 112a. By inputting supervised learning data into the AI ​​module of the natural language understanding module 112a, output data corresponding to the input text information and form information (including instruction information) is output from the natural language understanding module 112a. The output data is compared with the supervised learning data, and a loss assessment is performed to adjust the evaluation function and weighting in the AI ​​module of the natural language understanding module 112a so that the output data matches the supervised learning data. By preparing a large amount of supervised learning data and repeating the learning process on the large amount of supervised learning data, the natural language understanding module 112a is constructed to output output data that matches the supervised learning data.

[0071] FIG. 17 shows another example of a training method for the natural language understanding module 112a. In this example, the natural language understanding module 112a is trained as a module that simultaneously extracts and outputs multiple items included in a medical form. In the example of FIG. 17, the supervised training data includes form information including instruction information specifying the output of items necessary to generate a medical form related to stroke, such as the patient's age ("Age"), gender ("Gender"), blood pressure ("Blood Pressure"), and pulse rate ("Heart Rate"). By inputting the supervised training data into the AI ​​module of the natural language understanding module 112a, output data corresponding to the input text information and form information (including instruction information) is output from the natural language understanding module 112a. The output data is compared with the training data in the supervised training data, and a loss evaluation is performed to adjust the evaluation function and weighting in the AI ​​module of the natural language understanding module 112a so that the output data for each item included in the output data matches the data for each item included in the training data. A large amount of supervised learning data is prepared, and the natural language understanding module 112a is constructed so that output data that matches the supervised learning data is output by repeating the learning process on the large amount of supervised learning data.

[0072] It is more preferable to perform training so that items that are considered important in the output data are preferentially extracted. For example, the items included in the form information are weighted by importance, and the natural language understanding module 112a is trained so that the more important an item is, the more correctly medical information related to that item is extracted.

[0073] Specifically, a weight is assigned to each item for the loss used for evaluation when training the AI ​​module of the natural language understanding module 112a. That is, the higher the importance of an item, the greater the evaluation of the loss when the natural language understanding module 112a outputs something different from the training data. Then, the natural language understanding module 112a is trained so that the loss is small.

[0074] For example, if the weights for the gender, illness, time, and MRI items are 1, 5, 3, and 3, respectively, and the output data for the gender, illness, time, and MRI items from the natural language understanding module 112a are correct, incorrect, incorrect, and correct, respectively, relative to the training data, the weighted loss error is calculated as (5 + 3) / (1 + 5 + 3 + 3) = 0.67. In contrast, if no weighting is applied to each item, the loss error is calculated as (1 + 1) / (1 + 1 + 1 + 1) = 0.50. Therefore, by performing machine learning on the natural language understanding module 112a to minimize the weighted loss error, the natural language understanding module 112a can be trained to output more correct output data for items with larger weighting values.

[0075] Furthermore, if the input text information does not contain necessary medical information, it is preferable to train the system to output special information such as "not included." This is because when medical information is unknown, the fact that the information is unknown itself carries information.

[0076] In step S16, the natural language understanding module 112a is evaluated. The learning data is divided into a training set, a validation set, and a test set, and the natural language understanding module 112a that has been trained using the training set is evaluated using the validation set and the test set. This makes it possible to evaluate the performance of the AI ​​model of the natural language understanding module 112a for unknown data.

[0077] For example, the accuracy of output data to be filled in a medical form based on the transcription can be evaluated using metrics such as accuracy rate and mean absolute error, or domain-specific accuracy metrics such as perplexity and BLEU.

[0078] As described above, the natural language understanding module 112a included in the medical form generation module 112 can be constructed by machine learning.

[0079] <Form Embedding Module> In the medical form generation module 112, the medical information extracted from the text information and output by the natural language understanding module 112a is embedded in a desired medical form by the form embedding module 112b and output.

[0080] The form embedding module 112b can be constructed as a machine-learned AI module. The training of the form embedding module 112b involves several steps. The training process for constructing the form embedding module 112b will be described below with reference to the flowchart in FIG. 18.

[0081] In step S20, a process is performed to prepare training data to be used for training the form embedding module 112b. As the training data, supervised training data is prepared by combining the medical information output from the natural language understanding module 112a and a medical form in which the medical information is appropriately embedded, as training data.

[0082] For example, as shown in FIG. 19, medical information output from the natural language understanding module 112a is used as input data, a medical form with the medical information appropriately embedded is used as training data, and these pieces of information are combined to prepare supervised training data.

[0083] The learning data can be divided into a training set, a validation set, and a test set. The training set is used to train the form embedding module 112b. The validation set is used to validate the trained form embedding module 112b. The test set is used to finally test the validated form embedding module 112b. This makes it possible to evaluate the performance of the form embedding module 112b on unknown data.

[0084] In step S22, a learning process is performed for form embedding module 112b. Form embedding module 112b is trained using the training data prepared in step S20. Specifically, input data included in the training data is input to the AI ​​module of form embedding module 112b, and machine learning is performed so that the medical form, which is training data included in the training data, is output from form embedding module 112b.

[0085] FIG. 20 illustrates a learning method for the form embedding module 112b. In the example of FIG. 20, medical information related to stroke, such as the patient's age ("Age"), gender ("Gender"), blood pressure ("Blood Pressure"), and pulse rate ("Heart Rate"), is used as input data in the supervised learning data, and a medical form with this information embedded is used as training data. By inputting the input data to the AI ​​module of the form embedding module 112b, a medical form with the input data embedded is output as output data from the form embedding module 112b. The output data is compared with the training data in the supervised learning data, and a loss assessment is performed to adjust the evaluation function and weighting in the AI ​​module of the form embedding module 112b so that the medical information for each item included in the output data matches the medical information for each item in the medical form included in the training data. A large amount of supervised learning data is prepared, and the learning process is repeated on the large amount of supervised learning data. This configures the form embedding module 112b to output output data that matches the training data.

[0086] In step S24, the form embedding module 112b is evaluated. The learning data is divided into a training set, a validation set, and a test set, and the form embedding module 112b, which has been trained using the training set, is evaluated using the validation set and the test set. This allows the performance of the AI ​​model of the form embedding module 112b for unknown data to be evaluated.

[0087] As described above, the form embedding module 112b included in the medical form generation module 112 can be constructed by machine learning.

[0088] The medical form actually generated by the form embedding module 112b is stored in the memory unit 12 of the server 102. Furthermore, the generated medical form is presented to a user (such as a medical professional) by the output unit 16 of the server 102 or the output unit 26 of the client 104, as necessary. The generated medical form may be displayed on a display or the like as electronic data, or may be printed out on paper or the like and presented to the user.

[0089] The medical information output from the natural language understanding module 112a is output in association with each item of the medical form specified by the form information. Therefore, the form embedding module 112b may embed the medical information output from the natural language understanding module 112a in association with each item of the medical form, and generate and output the medical form.

[0090] Here, the form embedding module 112b may refer to a medical terminology database stored in advance in the storage unit 12 of the server 102 and determine whether or not the terms included in the medical information extracted by the natural language understanding module 112a match the terms included in the medical terminology database. For example, if the terms included in the medical information extracted by the natural language understanding module 112a do not match the terms included in the medical terminology database, an alert may be output to a user such as a medical professional. Furthermore, for example, if the terms included in the medical information extracted by the natural language understanding module 112a do not match the terms included in the medical terminology database, the terms may be replaced with terms from the medical terminology database that are predicted to be correct, and the replaced terms may be embedded in the medical form and output.

[0091] Furthermore, an alert (warning) may be output when a specific performance evaluation value for a medical form output from the medical support system 100 deteriorates below a predetermined reference value. For example, the importance of each of the above items may be weighted, and a performance evaluation value may be calculated according to the accuracy of the medical information included in the medical form output from the medical form generation module 112, and an alert (warning) may be output when the performance evaluation value deteriorates below the reference value.

[0092] Specifically, for example, if the weights for the gender, illness, time, and MRI items are 1, 5, 3, and 3, respectively, and the output data for the gender, illness, time, and MRI items from the natural language understanding module 112a are correct, incorrect, incorrect, and correct, respectively, the weighted loss error is calculated as (5 + 3) / (1 + 5 + 3 + 3) = 0.67. Therefore, if the weighted loss error exceeds a predetermined threshold, an alert (warning) is output. Furthermore, for example, an alert (warning) may be output if certain required medical information items are missing from the output medical form.

[0093] Furthermore, the form embedding module 112b may detect potential contradictions or inconsistencies in the medical information based on the interrelationships between the medical information output from the natural language understanding module 112a. For example, when specific medical information should be combined for multiple items, the specific combination of medical information for the multiple items is stored in advance in the storage unit 12 of the server 102 as a medical information database, and an alert (warning) is output when a combination of medical information other than the specific combination of medical information for the multiple items is output from the natural language understanding module 112a.

[0094] 21, the medical form generation module 112 may be constructed by combining the natural language understanding module 112a and the form embedding module 112b into a single AI module. In this case, text information and form information are used as input data, and the medical form to be output corresponding to that information is used as training data, with these pieces of information being combined and used as supervised training data. The input data included in the training data is input to an AI module combining the natural language understanding module 112a and the form embedding module 112b, and machine learning is performed so that the medical form, which is training data included in the training data, is output from the AI ​​module.

[0095] Furthermore, when the natural language understanding module 112a and the form embedding module 112b are trained by machine learning, they can also be trained to predict the type of medical form appropriate for text information based on the text information.

[0096] Furthermore, as shown in FIG. 22, it is preferable to analyze errors and mistakes in the output data output from the natural language understanding module 112a. This provides insight into the need for improvement, and makes it possible to improve the AI ​​model and training data of the natural language understanding module 112a based on this insight. It is also preferable to continuously evaluate the natural language understanding module 112a when it is actually used. Feedback from users such as doctors and medical professionals is useful for identifying problems and areas for improvement in the natural language understanding module 112a. It is also preferable to continuously train the AI ​​model of the natural language understanding module 112a (active learning) based on user feedback.

[0097] In addition, the medical support system 100 may be provided after the natural language understanding module 112a and the form embedding module 112b have been trained in advance, or scratch learning may be performed to build the natural language understanding module 112a and the form embedding module 112b in-house by training them from public data or the company's own service data.

[0098] Additionally, medical forms should be categorized based on the type of illness, type of treatment, medical condition, surgical procedure, and other medically relevant factors. For medical forms, metadata or tags can be associated with each form type to enable the medical form generation module 112 to quickly identify and select the appropriate form.

[0099] The medical support system 100 can identify a patient when the patient arrives or logs in using a patient-specific identifier, biometric authentication, near-field communication (NFC) technology, or the like, and output the treatment plan and history of the identified patient in an appropriate medical form. In addition, it can synchronize with an electronic health record (EHR) and input patient data into the medical support system 100 in real time. This allows the medical support system 100 to predict and select an appropriate medical form based on recent diagnoses, medications, and test results, and embed medical information in the medical form and output it.

[0100] Additionally, sensors such as wearable health devices may be integrated into medical support system 100. Real-time health data of a patient obtained from the wearable health device may be input as text information to medical support system 100, and when the health data indicates a specific condition (e.g., an abnormal condition) of the patient, medical support system 100 may output a medical form with medical information related to that condition embedded therein.

[0101] Furthermore, it is also possible to predict which format of medical form will be required next based on past medical forms. For example, if there is a history of a patient's condition being filled out into a monitoring form each month, form information may be input into the medical support system 100 so that the monitoring form is generated and output at each monthly examination of the patient. Furthermore, it is also possible to analyze the patient's medical history and input form information into the medical support system 100 so that a medical form in a format required for that patient is generated and output. For example, it is preferable to generate and output a medical form in a format specialized for cardiology for a patient with a history of heart disease.

[0102] Furthermore, a geolocation or indoor positioning system may be integrated into the medical support system 100, and location information indicating the location of a patient or medical equipment may be input to the medical support system 100, and a medical form to be generated and output by the medical support system 100 may be selected based on the location information. For example, if a patient is in the cardiology department of a medical institution (hospital, etc.), it is preferable to set the form information so that a medical form related to the cardiology department is generated and output from the medical support system 100.

[0103] Furthermore, medical forms of an appropriate format may be selected in the medical support system 100 depending on the time period, date and time, day of the week, week, month, or season. For example, during an influenza epidemic, it is preferable to set form information and input it into the medical support system 100 so that medical forms for evaluating influenza are generated and output preferentially.

[0104] Furthermore, by analyzing the behavior and preferences of medical professionals, if a medical professional frequently uses a particular medical form, the medical form may be preferentially selected. For example, if a doctor involved in a particular disease frequently uses a medical form related to that disease, it is preferable to set form information and input it into the medical support system 100 so that the medical form is preferentially generated and output.

[0105] Additionally, external data sources such as disease occurrence databases and local health alerts may be integrated into the medical support system 100. In this way, if a specific disease suddenly occurs in a local area, it is preferable to set form information and input it into the medical support system 100 so that medical forms related to that disease are generated and output preferentially.

[0106] Furthermore, natural language processing (NLP) may be used to implement a chatbot or a voice assistant in the medical support system 100. Using the chatbot or the voice assistant, patients or medical professionals may input their current needs (e.g., "I need a post-operative follow-up form") by voice, and form information may be set to generate and output a specified medical form and input to the medical support system 100.

[0107] Furthermore, by integrating the medical support system 100 of this embodiment with an existing clinical decision support system, details of a patient's treatment plan can be input as text information to the medical support system 100. This allows the medical support system 100 to generate and output a medical form in which appropriate information for the treatment plan is embedded, based on the patient's treatment plan input as text information.

[0108] It is also possible to provide feedback to users such as medical professionals as to whether the medical forms generated and output by the medical support system 100 are correct. Based on this feedback, the medical form generation module 112 may learn to generate and output more correct medical forms.

[0109] In addition, by expressing the learning data in multiple languages ​​and having the medical form generation module 112 perform machine learning, the medical support system 100 can be made multilingual and applied to generating medical forms that support multiple languages.

[0110] As described above, the medical support system 100 of this embodiment can generate text information containing medical information related to medical care from voice information containing medical information related to medical care spoken by medical professionals such as doctors, nurses, and emergency personnel. Furthermore, based on the text information containing medical information related to medical care, medical information for items to be filled in a desired medical form can be extracted from the text information. Furthermore, the medical information extracted from the text information containing medical information related to medical care can be automatically embedded in the desired medical form, and a completed medical form can be generated and output.

[0111] The medical support system 100 can be used in various medical settings, and is particularly effective for creating medical forms in emergency medical settings such as acute stroke and acute myocardial infarction (AMI).

[0112] For example, when a patient is brought to the emergency room with symptoms of acute cerebral infarction, a doctor can quickly initiate a stroke protocol. Using the medical support system 100, the doctor can verbally provide the patient's symptoms, time of onset, risk factors, and initial physical examination findings. The medical information can then be extracted from the audio information, accurately transcribed, and automatically entered into the desired medical form. This expedites the process of diagnosis, testing, treatment, and other procedures. Medical professionals can then focus on treating the patient, including imaging diagnosis, testing, and the possibility of thrombolytic therapy or thrombectomy.

[0113] For example, if a patient visits an emergency room with severe chest pain, a doctor may suspect acute myocardial infarction. The doctor immediately verbally communicates the patient's medical history, symptoms, electrocardiogram changes, and initial blood test results to the medical support system 100. The medical support system 100 quickly extracts the necessary medical information from the voice information, enters the medical information into a desired medical form, and outputs it. This makes it possible to quickly indicate, for example, the possibility of percutaneous coronary intervention (PCI), saving precious time when every minute counts in order to preserve the patient's myocardium and increase the patient's chances of survival.

[0114] In this way, the medical support system 100 enables the generation and output of quick and accurate medical forms in medical emergencies, thereby minimizing the time medical professionals spend on paperwork and leading to prompt diagnosis, testing, and treatment in time-sensitive situations such as stroke and AMI, thereby enabling the efficient provision of medical services.

[0115] <Reasons for extracting medical information> The natural language understanding module 112a can be configured to output the reason why medical information extracted from text information containing medical information was extracted.

[0116] There are several steps involved in training the natural language understanding module 112a, which outputs the reason why extracted medical information was extracted. Hereinafter, the training process for building the natural language understanding module 112a, which outputs the reason why medical information was extracted, will be described with reference to the flowchart in Figure 23.

[0117] In step S30, a process is performed to prepare training data to be used for training the natural language understanding module 112a. A large amount of text information indicating medical records needs to be collected as the training data. Form information is also prepared to identify medical information to be extracted from the text information. Reason information indicating the reason for the extraction of the medical information to be extracted is also prepared.

[0118] The natural language understanding module 112a receives input data of text information containing medical information and form information specifying the medical information to be extracted from the text information, and uses training data of medical information to be output when this information is input and reason information indicating why the medical information was extracted, and prepares supervised learning data that combines these.

[0119] For example, text information containing medical information and form information indicating information (items) to be output from the natural language understanding module 112a are used as input data, and the medical information to be output from the natural language understanding module 112a and the reason for extracting the medical information are used as training data, which are then combined to form training data. Figure 24 shows an example of text information containing medical information acquired as a chat message. Preferably, the form information is configured by combining instruction information for instructing the extraction of medical information and reason information with medical forms that specifically identify the medical information to be extracted.

[0120] The form information may be the same as the form information in the above embodiment. For example, the form information may include instruction information instructing to output items indicating the patient's condition, such as the patient's age, sex, medical history, blood pressure, pulse rate, oxygen saturation level, presence or absence of dyspnea, state of consciousness, electrocardiogram changes, blood test results, etc.

[0121] The learning data can be divided into a training set, a validation set, and a test set. The training set is used for learning the natural language understanding module 112a. The validation set is used to validate the learned natural language understanding module 112a. The test set is used for final testing of the validated natural language understanding module 112a. This makes it possible to evaluate the performance of the natural language understanding module 112a on unknown data.

[0122] In step S32, data preprocessing is performed. The text information input to the natural language understanding module 112a is preferably formatted in a consistent, machine-readable format. Therefore, in this step, the text information is converted into a machine-readable format as needed. For example, depending on the complexity of the medical form, the text information may be structured as pairs of items and their values ​​or as a more complex schema.

[0123] In step S34, a learning process is performed for the natural language understanding module 112a. The natural language understanding module 112a is trained using the training data prepared in steps S30 and S32. Specifically, the text information and form information included in the training data are input to the AI ​​module of the natural language understanding module 112a, and machine learning is performed so that output data, which is training data included in the training data, is output from the natural language understanding module 112a.

[0124] Here, the AI ​​module constituting the natural language understanding module 112a may be a module including a pre-trained AI model such as GPT-3 or GPT-4, as in the above embodiment, and such an AI model may be fine-tuned to constitute the natural language understanding module 112a.

[0125] 25 shows a learning method of the natural language understanding module 112a. By inputting supervised learning data (Prompt) into the AI ​​module of the natural language understanding module 112a, medical information (Answer) corresponding to the input text information and form information (including instruction information) is output as output data from the natural language understanding module 112a. Furthermore, reason information (reason) indicating why the medical information was extracted is output from the natural language understanding module 112a.

[0126] In machine learning, the output data is compared with teacher data in supervised learning data, and a loss assessment is performed to adjust the evaluation function and weighting in the AI ​​module of the natural language understanding module 112a so that the output data matches the teacher data. By preparing a large amount of supervised learning data and repeating the learning process on the large amount of supervised learning data, the natural language understanding module 112a is constructed so that output data that matches the teacher data is output.

[0127] Fig. 26 shows an example of output of medical information and reason information extracted by the natural language understanding module 112a. Fig. 26 shows both the output information and its Japanese translation. As shown in Fig. 26, the natural language understanding module 112a is machine-trained to extract medical information related to items specified in the form information.

[0128] Furthermore, the reason information can be information indicating a text portion containing medical information extracted from text information containing medical information. As illustrated in FIG. 26, when the medical information "NIHSS_score" is extracted as "15", the natural language understanding module 112a is machine-learned to also extract reason information indicating that the text information contains information such as "NIHSS is 15." For other medical information, machine learning is also performed to extract reason information indicating the reason why the medical information was extracted together with the medical information.

[0129] It is preferable that the medical information is also output in JSON format, as shown in Fig. 26. That is, it is preferable that the natural language understanding module 112a includes output data in JSON format as training data in the supervised training data, and is trained so that the medical information is output in JSON format.

[0130] In addition, if the input text information does not contain the necessary medical information, it is preferable to train the system to output special information such as "not included." When medical information is unknown, the fact that the information is unknown itself carries information. Similarly, the system can be trained to display reasons such as "not stated in the text." It is preferable to make it so.

[0131] In step S36, the natural language understanding module 112a is evaluated. The learning data is divided into a training set, a validation set, and a test set, and the natural language understanding module 112a, which has been trained using the training set, is evaluated using the validation set and the test set. This allows the performance of the AI ​​model of the natural language understanding module 112a for unknown data to be evaluated. This process can be performed in the same way as the machine learning of the natural language understanding module described above.

[0132] As described above, in the medical form generation module 112, a natural language understanding module 112a can be constructed by machine learning that can output medical information corresponding to form information (including instruction information) from text information containing medical information, and also output the reason for extracting the medical information.

[0133] [Configuration of the invention] [Configuration 1] A medical support system that automatically fills in medical forms related to medical care, A medical support system characterized by comprising a natural language understanding module that has been machine-learned to output the reason for extracting specified medical information from text information containing medical information related to medical care. [Configuration 2] The medical support system according to configuration 1, The natural language understanding module is a medical support system characterized by being machine-learned using text information containing medical information related to medical care as input data, and the reason for extracting the specified medical information to be output from the text information as training data, combining these pieces of information to use supervised learning data so that when the input data is input, the reason for extracting the specified medical information, which is the training data, is output. [Configuration 3] The medical support system according to configuration 1 or 2, A medical support system comprising a form embedding module that embeds medical information output from the natural language understanding module into the medical form and outputs the medical information. [Configuration 4] The medical support system according to any one of configurations 1 to 3, A medical support system comprising a speech-to-text conversion module that generates text information from speech information containing medical information related to medical care. [Configuration 5] A medical assistance program for automatically filling out medical forms relating to medical care, comprising: A medical support program that causes a computer to function as a machine-learned natural language understanding module that outputs the reason for extracting specified medical information from text information containing medical information related to medical care. [Configuration 6] A medical support method for automatically filling out medical forms relating to medical care, comprising: A medical support method characterized by having a computer functioning as a natural language understanding module that has been machine-learned to output reasons for extracting specified medical information from text information containing medical information related to medical care, outputting reasons for extracting the specified medical information from text information containing medical information related to medical care. [Explanation of symbols]

[0134] 10 processing unit, 12 memory unit, 14 input unit, 16 output unit, 18 communication unit, 20 processing unit, 22 memory unit, 24 input unit, 26 output unit, 28 communication unit, 100 medical support system, 102 server, 104 client, 106 information communication network, 110 speech-to-text conversion module, 112 medical form generation module, 112a natural language understanding module, 112b form embedding module.

Claims

1. A medical support system that automatically fills in medical forms related to medical care, A medical support system characterized by comprising a natural language understanding module that has been machine-learned to output the reason for extracting specified medical information from text information containing medical information related to medical care.

2. The medical support system according to claim 1, The natural language understanding module is a medical support system characterized by being machine-learned using text information containing medical information related to medical care as input data, and the reason for extracting the specified medical information to be output from the text information as training data, combining these pieces of information to use supervised learning data so that when the input data is input, the reason for extracting the specified medical information, which is the training data, is output.

3. 3. The medical support system according to claim 1 or 2, A medical support system comprising a form embedding module that embeds medical information output from the natural language understanding module into the medical form and outputs the medical information.

4. 3. The medical support system according to claim 1 or 2, A medical support system comprising a speech-to-text conversion module that generates text information from speech information containing medical information related to medical care.

5. A medical assistance program for automatically filling out medical forms relating to medical care, comprising: A medical support program that causes a computer to function as a machine-learned natural language understanding module that outputs the reason for extracting specified medical information from text information containing medical information related to medical care.

6. A medical support method for automatically filling out medical forms relating to medical care, comprising: A medical support method characterized by having a computer functioning as a natural language understanding module that has been machine-learned to output reasons for extracting specified medical information from text information containing medical information related to medical care, outputting reasons for extracting the specified medical information from text information containing medical information related to medical care.

Images