Diagnostic Support Device

Abstract

To provide a medical examination support device and a program that enable even an inexperienced doctor to correctly examine a patient based on the knowledge and know-how of an experienced doctor. [Solution] The diagnostic support device 1 has a chief complaint group memory unit that stores a chief complaint group including multiple chief complaints, a clinical feature memory unit that stores multiple clinical features associated with the multiple chief complaints, a disease memory unit that stores multiple diseases associated with the multiple clinical features, and a disease selection unit that selects one or more diseases stored in the disease memory unit.When patient chief complaint information, which is the patient's medical history and physical findings, is input, the multiple clinical features corresponding to the patient chief complaint information input from the clinical feature memory unit are read out from the clinical feature memory unit, and the disease selection unit selects one or more diseases using the patient's clinical feature information, which is the clinical feature selected from the multiple clinical features read out, and suggests the selected one or more diseases as candidate diseases that the patient may be suffering from.

Description

[Technical Field]

[0001] The present invention relates to a diagnosis support device. [Background technology]

[0002] Japanese Patent No. 4801995 describes a medical computerization system. This system inputs chief complaint information about the symptoms complained by the patient and examination information about the interview results of the doctor who examined the patient. The system then uses a computer to digitize medical information such as medical records.

[0003] Even if medical information is digitized using the above-mentioned systems, there are problems such as difficulty in quickly obtaining the necessary information for each complaint, and problems such as doctors having difficulty understanding the condition of the patient or being unable to determine the appropriate medical department. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Patent No. 4801995 Summary of the Invention [Problem to be solved by the invention]

[0005] Therefore, there is a need for a device that can quickly obtain the information necessary to understand the patient's condition based on the patient's main complaint, and that can obtain medical information that is easy for a doctor (or patient) to understand. There is also a need for a device that can obtain information on the appropriate medical department. [Means for solving the problem]

[0006] This invention is based on the finding that the above-mentioned problems can be basically solved by providing a device that can quickly obtain information corresponding to each chief complaint by appropriately guiding corresponding clinical characteristics, etc. based on the chief complaint, and that can obtain medical information regarding the pathological condition that is easy for many viewers, such as doctors and patients, to understand, and further by providing a device that can provide information regarding the appropriate medical department.

[0007] The present invention relates to a diagnosis support device for creating a medical information statement. The medical information statement is a statement for understanding the pathological condition of a patient. The medical information statement preferably includes information on a recommended medical department. The diagnosis support device includes a chief complaint group storage unit, a clinical feature storage unit, a disease storage unit, a chief complaint display unit, a clinical feature display unit, and a medical information statement creation unit. The chief complaint group storage unit is an element for storing a chief complaint group including multiple chief complaints. The clinical feature storage unit is an element for storing a plurality of clinical features associated with a plurality of chief complaints. The disease memory section is an element for storing a plurality of diseases associated with a plurality of clinical features. The chief complaint display section is an element for displaying a chief complaint selection to allow the patient to select one or more of a plurality of chief complaints. The clinical feature display unit is an element that, when patient's chief complaint information is input based on the chief complaint selection display, reads out multiple clinical features corresponding to the input patient's chief complaint information from the clinical feature memory unit and performs a clinical feature selection display to allow selection of multiple clinical features corresponding to the patient's chief complaint information. The medical information statement creation unit is an element for creating a medical information statement using selected clinical features, which are selected clinical features input into the diagnosis support device, and information including patient's chief complaint information. The medical information statement creation unit may have a trained model, and the medical information statement creation unit may obtain a medical information statement by inputting the selected clinical features and the patient's chief complaint information into the trained model.

[0008] The diagnosis support device may further include a medical facility storage unit that stores one or more pieces of medical facility information related to a medical department or a disease, and a referral letter creation unit that creates a referral letter including the medical facility information related to the medical department or the disease and a medical information statement. This specification also provides a program for causing a computer to function as a diagnosis support device for creating a medical information statement. [Effects of the Invention]

[0009] According to the present invention, it is possible to provide a device that can quickly obtain the information necessary to understand the pathological condition based on the patient's chief complaint, and that can obtain medical information that makes the pathological condition easy to understand when viewed by a doctor (or patient). Furthermore, according to the present invention, it is possible to provide a device that can obtain information on the appropriate medical department. [Brief explanation of the drawings]

[0010] [Figure 1] FIG. 1 is a block diagram illustrating a diagnosis support device. [Figure 2] FIG. 2 is a conceptual diagram showing an example of the chief complaint input mode. [Figure 3] FIG. 3 is a conceptual diagram showing an example of clinical features displayed on the display unit when the chief complaint is numbness. [Figure 4] FIG. 4 is a conceptual diagram showing how a trained model is constructed and how a predetermined output is obtained using the trained trained model. [Figure 5] The example in Figure 5 is a conceptual diagram showing a diagnostic example for a patient whose chief complaint is "sore throat." [Figure 6-1] Figure 6 is a conceptual diagram showing a diagnostic example for a case where the main complaint is coughing. [Figure 6-2] Figure 6-2 shows a continuation of Figure 6-1. DETAILED DESCRIPTION OF THE INVENTION

[0011] The diagnostic support device of the present invention aims to provide a device that can quickly obtain information necessary to understand a patient's pathology based on the patient's chief complaint and that can generate medical information that is easy for a doctor (or patient) to understand. The diagnostic support device 1 includes, for example, a chief complaint group storage unit 11, a clinical feature storage unit 13, a disease storage unit 15, a chief complaint display unit 21, a clinical feature display unit 23, and a medical information statement creation unit 31. The diagnostic support device 1 may also include one or more of a disease selection unit 17, a medical facility storage unit, and a referral letter creation unit. Details of each component are described below. By storing and displaying information corresponding to a diagnosis in stages in this way, it is possible to appropriately suggest candidate diseases while utilizing the doctor's knowledge. The diagnostic support device 1 performs various processes using a computer.

[0012] The computer has an input unit, an output unit, a control unit, a calculation unit, and a memory unit, and each element is connected by a bus or the like to enable information exchange. For example, a control program or various information may be stored in the memory unit. When predetermined information is input from the input unit, the control unit reads the control program stored in the memory unit. Then, the control unit reads the information stored in the memory unit as appropriate and transmits it to the calculation unit. The control unit also transmits the input information to the calculation unit as appropriate. The calculation unit performs calculation processing using the received various information and stores it in the memory unit. The control unit reads the calculation results stored in the memory unit and outputs them from the output unit. In this manner, various processes and steps are performed. The various units and means execute these various processes. The computer may have a processor, and the processor may realize various functions and steps. The computer may be standalone. Some of the functions of the computer may be distributed between a server and a terminal. In this case, it is preferable that the server and the terminal can exchange information via a network such as the Internet or an intranet. The computer may include a processor and a memory coupled to the processor. The memory may store instructions that, when executed by the processor, cause the computer to perform various processes or function as various elements. The computer may be provided with various training data to build a learning model and perform various calculations through machine learning. In this case, the computer may perform various analyses using the learning model created through machine learning and deep learning of AI (artificial intelligence).

[0013] A computer typically has a display unit (not shown). The display unit is an element for displaying various types of information based on the computer. A monitor or display, which is a type of computer output unit, or a touch panel of a smartphone functions as the display unit. The display unit may also be a projector.

[0014] Chief complaint group memory section 11 The chief complaint group memory unit 11 is an element for storing chief complaint groups including multiple chief complaints. Chief complaints include the patient's medical history and physical findings. Chief complaint groups may also include test findings. For example, by recording chief complaints such as "fever," "cough," and "abdominal pain" in the chief complaint group memory unit 11, it becomes easier to organize information at the early stage of diagnosis. It is preferable that chief complaints be categorized into multiple categories. The chief complaint group memory unit 11 may store multiple candidate associated diseases (and therefore diseases to be differentiated) associated with multiple chief complaints. Examples of chief complaint categories include those with a predetermined number or more of associated diseases (diseases to be differentiated) (first category), those with a predetermined number or more of associated diseases (diseases to be differentiated) that can be grouped based on clinical characteristics (second category), and those with a number of associated diseases (diseases to be differentiated) less than a predetermined number (third category). The first category may be a category in which the number of related diseases (diseases to be differentiated) is equal to or greater than a predetermined number and cannot be grouped based on clinical characteristics. Here, the predetermined number may be changed by inputting it into the device as appropriate. The categories may be automatically assigned by machine learning. In this way, the categories can be automatically changed according to the times and the prevalence of specific diseases.

[0015] The first category is the most difficult to accurately diagnose. Therefore, when the patient's chief complaint is entered into the device, it is preferable to display a hierarchical list of diseases that the patient may be suffering from. For example, fever of unknown origin is included in the first category.

[0016] The second category can be grouped by clinical features. If the chief complaint is in the second category, information regarding the grouping can be displayed and a group can be selected, allowing the user to select an appropriate clinical feature. An example of the second category is numbness. Numbness can be grouped based on the distribution of symptoms. For example, clinical features can be selected appropriately based on the part of the body where numbness occurs. In this case, a group can be selected by displaying the part of the body on the display and allowing the user to select it. If the chief complaint includes a localized chief complaint, the clinical feature memory unit 13 may store the localized chief complaint together with an image including the local area of ​​the body where the chief complaint may occur.

[0017] The third category is those in which the number of associated diseases (diseases that must be differentiated) is less than a specified number.

[0018] If the associated diseases (diseases to be differentiated) related to the input chief complaint include an extremely high-risk disease (which occurs with a certain frequency), a diagnosis of whether or not the patient has that disease may be made at an early stage. In this case, the chief complaint or clinical characteristics related to the high-risk disease may be appropriately displayed and selected to determine whether or not the patient has that high-risk disease. Then, after the high-risk disease (which occurs with a certain frequency) has been excluded, the chief complaint and clinical characteristics of other candidate diseases may be displayed and selected.

[0019] Clinical feature memory section 13 The clinical feature memory unit 13 is an element for storing multiple clinical features associated with multiple chief complaints. Clinical features are medical features derived based on the chief complaints. For example, the clinical feature memory unit 13 stores a clinical feature such as "possible viral infection" associated with the chief complaints of "fever" and "cough." Clinical features may include, for example, one or more of necessary precautions, matters to be confirmed in a medical interview, and tests to be performed, associated with the chief complaint. Necessary precautions are, for example, points that a doctor or patient should be aware of regarding a disease associated with the chief complaint. When patient complaint information is input to the device, corresponding clinical features are extracted from the clinical feature memory unit 13. Based on the input information, features likely to be associated with a specific disease are preferentially extracted. For example, candidate clinical features, such as sudden onset of numbness, may be extracted and displayed. The patient may then select a clinical feature associated with the chief complaint from the multiple clinical features displayed.

[0020] Disease Memory Section 15 The disease memory unit 15 is an element for storing multiple diseases associated with each clinical feature. This makes it possible to provide a list of possible diseases for a patient with specific clinical features. For example, in the case of "fever + cough," "influenza," "pneumonia," "bronchitis," etc. are listed as candidates. The disease memory unit 15 may store multiple diseases in association with their severity. The disease memory unit 15 may store multiple diseases in association with their frequency. Furthermore, the disease memory unit 15 may store information about the corresponding medical department in association with each disease.

[0021] Disease Selection Section 17 The disease selection unit 17 is an element that selects the feature most relevant to the patient's symptoms from among the multiple clinical features read from the clinical feature storage unit 13, and selects the corresponding disease from the disease storage unit 15. This makes it possible to identify appropriate disease candidates based on the patient's symptoms. For example, if a patient complains of "fever," "cough," and "fatigue," the disease selection unit 17 acquires clinical features corresponding to these chief complaints and presents diseases such as "influenza," "COVID-19," and "cold" as candidate diseases based on the acquired clinical features. The disease selection unit 17 may propose a group of diseases with high severity as candidate diseases before a group of diseases with low severity. The disease selection unit 17 may select a group of diseases with low severity so that they are displayed in descending order of frequency.

[0022] Chief complaint display section 21 The chief complaint display unit 21 is an element for displaying a chief complaint selection to allow the user to select one or more of a plurality of chief complaints. The chief complaint selection display may be, for example, the first screen in a doctor's diagnosis. In this chief complaint selection display, multiple chief complaints may be read from the chief complaint group storage unit and displayed on the display unit. It is preferable that one or more of the multiple chief complaints related to the patient can be selected using radio buttons or the like. The chief complaint display unit may also read and display patient information, including the patient's medical history, from a patient master that stores the target patient's medical history, and if the target patient is the one displayed, information including the patient's medical history may be input into the device by confirming using a radio button or the like. When a chief complaint is selected, information related to the patient's chief complaint (patient's chief complaint information) may be input into the device.

[0023] Clinical characteristics display section 23 When patient complaint information is input to the diagnostic support device, the clinical feature display unit 23 reads out a plurality of clinical features corresponding to the input patient complaint information from the clinical feature storage unit and displays the plurality of clinical features corresponding to the read patient complaint information on the display unit. The plurality of clinical features are, for example, provided with radio buttons, and selecting one of them selects the clinical feature corresponding to the patient. The clinical feature display unit 23 is an element for performing such clinical feature selection and display.

[0024] Medical information creation section 31 The medical information statement creation unit 31 is an element for creating a medical information statement using selected clinical features, which are selected clinical features input into the diagnostic support device, and information including patient complaint information. For example, the device 1 may have a statement creation unit that creates a medical information statement using terms contained in the clinical features and patient complaint information. The medical information statement refers to text necessary for doctors and medical professionals to understand the patient's condition. Furthermore, the device 1 may analyze a target disease and then create a medical information statement using terms contained in the clinical features and patient complaint information to provide information necessary to treat the target disease. The medical information statement preferably includes information about a recommended medical department. For example, after the device 1 analyzes the patient's target disease, the medical information statement creation unit 31 may use the information about the target disease to read information about the medical department stored in association with the target disease from the disease memory unit 15 and add it to the medical information statement.

[0025] The medical information sentence creation unit 31 may have a trained model, and the medical information sentence creation unit may obtain a medical information sentence by inputting selected clinical features and patient complaint information into the trained model. For example, the trained model may be trained by inputting a large number of combinations of selected clinical features, patient complaint information, and medical information sentences as training data. Then, when the selected clinical features and patient complaint information are input into the trained model as input data, appropriate medical information sentences can be output. Clinical features that may cause embarrassment when seen by a patient (e.g., snoring) are preferably expressed as medical terms in the training data. In this way, such terms (clinical features that may cause embarrassment when seen by a patient) included in the obtained medical information sentence are written in medical terms (technical terms). The device 1 may have such a dictionary, and when the medical information sentence contains such terms, the corresponding medical terms are replaced and the medical information sentence after the replacement is output.

[0026] The medical information creation unit 31 allows for the rapid and appropriate provision of information necessary for treating various chief complaints and for doctors to understand the pathology. Furthermore, if it is possible to determine the recommended medical department, it can be effectively used as an outpatient interview app, and it will enable the patient to be referred to the appropriate medical department or to be seen by the appropriate medical department.

[0027] Referral Letter Writing Department The device may have a referral letter creation unit. The medical information statement described above contains sufficient medical information for a doctor to treat the patient. For this reason, it is preferable that the referral letter created using this device 1 includes the medical information statement described above. On the other hand, it is preferable that the referral letter be created for the target medical institution (although the section related to the target medical institution may be left blank). To create a referral letter for the target medical institution, it is preferable that the device 1 has a medical facility storage unit that stores information on one or more medical facilities associated with a medical department or disease. In this case, the referral letter creation unit may read and display medical facility information associated with a medical department or disease, allowing the user to select an appropriate medical institution (hospital, medical department, doctor). If an appropriate medical institution is selected, a referral letter addressed to that medical institution can be created. If an appropriate medical institution cannot be selected, a referral letter can be created with the destination section left blank. The created referral letter can be stored in a storage unit, for example, and output as appropriate.

[0028] Diagnostic support flow An example of the diagnostic support flow of this device is shown below. Patient complaint information (medical history and physical findings) is input to the device. For example, a doctor may input the information using a touch panel, keyboard, mouse, or the like. Patient information, such as the patient's medical history, may also be read from information such as an electronic medical record. Furthermore, diagnostic results may be input to this device from various diagnostic devices.

[0029] FIG. 2 is a conceptual diagram illustrating an example of the chief complaint input mode. In the example of FIG. 2, medical history and physical findings are displayed as candidates, and a section prompting the user to enter the medical history and physical findings of the chief complaint by checking a box is displayed. Also, in the example of FIG. 2, a selection field for test findings is displayed. This selection field allows the doctor to understand what tests should be performed in relation to the chief complaint. After the doctor examines the patient (or after the patient performs the test themselves), the doctor can select a selection field for test findings to input information about the test findings into the device 1. The device 1 may have artificial intelligence or a language analysis function (a function for extracting keywords). When a sentence related to the chief complaint is entered in the free text entry field, the device 1 may analyze the entered sentence, acquire the patient's chief complaint information, and store it appropriately in a memory unit. The device may also have a camera, which may take a picture of the doctor or patient and use it to analyze the chief complaint. Since doctors' facial expressions change depending on the severity of the patient's condition, a trained model may be constructed using the doctor's facial expression photographed and the target disease as training data, and the target disease and the severity of the disease may be output by inputting the doctor's facial expression photographed into the trained model. In either case, the patient's chief complaint information is input to the device.

[0030] Based on the patient's chief complaint information, the device 1 acquires information corresponding to the chief complaint stored in association with the patient's chief complaint information in the chief complaint group storage unit 11. The device 1 may have a chief complaint information acquisition unit. The device may then display the acquired information about the chief complaint on a display unit (monitor, printed matter, etc.). For example, the chief complaint group display unit 21 of the device may output information for displaying the acquired information about the chief complaint to the display unit, and the display unit that receives the information may output the information about the chief complaint.

[0031] The device 1 extracts clinical features related to the relevant chief complaint from the clinical feature storage unit 13. For this purpose, the device 1 may have a clinical feature extraction unit (clinical feature acquisition unit). The device 1 may display disease candidates corresponding to the extracted clinical features from the disease storage unit 15. In this case, the clinical feature display unit 23 of the device 1 outputs information for displaying the clinical features related to the chief complaint acquired in this manner on the display unit, and the display unit receiving the information outputs disease candidates corresponding to the extracted clinical features.

[0032] After multiple clinical features related to the chief complaint are displayed on the display unit, the user (doctor, patient) may select an applicable clinical feature. When the user selects an applicable clinical feature, information about the selected clinical feature is input to the device 1. In this way, clinical feature information of the patient, which is the clinical feature selected from the multiple clinical features, is obtained. To achieve this function, the device 1 may have a clinical feature selection input unit. The selected clinical feature input to the device 1 is preferably stored in the memory unit as appropriate, and is read out when the disease selection unit 17 selects a disease candidate, or is used when the disease selection unit 17 selects a disease candidate.

[0033] FIG. 3 is a conceptual diagram showing an example of clinical features displayed on the display unit when the chief complaint is numbness. For example, when the chief complaint is numbness, an image prompting the user to input the location of the numbness is displayed on the display unit. The doctor or patient selects the location of the numbness. The color of the numb location on the screen may then change. Alternatively, the display screen may display multiple candidate locations of numbness in advance, allowing the user (doctor or patient) to select one. In the example of FIG. 3, clinical features such as sensory impairment localized to one limb, glove or stocking-type sensory impairment, polyneuropathy, sensory impairment due to spinal cord or cauda equina, and sensory impairment due to cerebral or brainstem involvement are displayed. Depending on the location input, corresponding clinical features may be automatically selected and input into the device 1. Alternatively, the diagnosing doctor may select the corresponding clinical features and input them into the device 1.

[0034] Next, the disease selection unit 17 selects the most appropriate disease candidate based on the patient's clinical feature information. The selected disease candidate may be output as a diagnosis result. The device 1 may display information about the disease candidate or the selected disease candidate on a display unit (monitor, printed matter, etc.). The disease display unit 27 of the device 1 outputs information about the disease candidate or the selected disease candidate to the display unit, and the display unit that receives the information outputs information about the disease candidate or the selected disease candidate.

[0035] The disease display unit 27 may have a trained model (diagnostic result model) for outputting diagnostic result information. This diagnostic result model may be constructed by inputting the chief complaint, clinical features, diagnostic result, and sentences for doctors to explain the disease to patients as training data. In this case, the training data may include sentences for doctors (which may include technical terms) and sentences for patients (in simple language that will not upset patients). In this way, when the chief complaint, clinical features, and diagnostic result are obtained through the above process, the disease display unit 27 inputs this information into the diagnostic result model, thereby obtaining sentences for doctors to explain the disease to patients as output. Obtaining such sentences not only promotes understanding for doctors but also provides sentences that are easy for patients to understand. Furthermore, inputting information regarding whether the information is intended for doctors or patients as input information enables output appropriate for the viewer. In this way, the device 1 may construct a trained model using the clinical feature information of a patient and one or more diseases associated with that patient's clinical feature information as training data (training data). In this way, the disease selection unit 17 can input the clinical feature information of the patient into the trained model, thereby obtaining one or more diseases as output information from the trained model.

[0036] In this way, the diagnosis support device of the present invention can support diagnosis by deriving clinical features based on the patient's chief complaint information and efficiently selecting disease candidates.

[0037] FIG. 4 is a conceptual diagram showing how a trained model is constructed and how a predetermined output is obtained using the trained trained model. As explained above, the medical information sentence creation unit inputs a large number of combinations of selected clinical features, patient complaint information, and medical information sentences as learning data to train the trained model. Then, when the selected clinical features and patient complaint information are input as input data to the trained model, it becomes possible to output appropriate medical information sentences. The device 1 may also have a medical information sentence display unit for displaying the medical information sentences. In this way, the doctor who made the diagnosis can obtain the medical information necessary for treating the patient in text form.

[0038] Furthermore, for example, if the consultation is with a general internal medicine specialist or if there are other specialists available, it may be possible to refer the patient to another specialist. The other specialist may be a different department within the same hospital or a department at another medical institution. This device 1 can include information about the department in the medical information text. This makes it possible to guide the patient to the appropriate department. Furthermore, if the device 1 has a referral letter creation unit, the referral letter creation unit may create the referral letter.

[0039] The program or application causes the computer to function as a diagnostic support device including a chief complaint group storage unit 11, a clinical feature storage unit 13, a disease storage unit 15, and a disease selection unit 17. The computer-readable information recording medium is a medium that non-temporarily stores the above-mentioned program. The computer may be realized by a processor having a storage unit and a calculation unit. This method also provides a diagnostic support method by a computer.

[0040] Another aspect of the present invention is as follows: There may be many diseases that can cause a chief complaint. For this reason, even if a medical computer system is used to store chief complaint information and examination information, it is difficult to provide reliable and efficient medical treatment. This invention provides a device or the like that can display one or more of necessary precautions, items to be confirmed in the medical interview, and tests to be performed, so that even inexperienced doctors can provide reliable, efficient, and thorough medical treatment.

[0041] The diagnostic support device of this invention is a device that proposes candidate clinical features necessary for diagnosing a disease that a patient may be suffering from based on information including findings on the patient, and outputs candidate diseases that the patient may be suffering from based on the proposed candidate clinical features. The diagnostic support device has a chief complaint group memory unit, a clinical feature memory unit, a disease memory unit, and a disease selection unit. The chief complaint group memory unit is an element for storing a chief complaint group including multiple chief complaints. The multiple chief complaints include the patient's medical history and physical findings. The clinical feature memory unit is an element for storing multiple clinical features associated with the multiple chief complaints. The disease memory unit is an element for storing multiple diseases associated with the multiple clinical features. The disease selection unit is an element for selecting one or more diseases stored in the disease memory unit.

[0042] The patient's medical history and chief complaint information, which is physical findings, are input into the diagnosis support device. A plurality of clinical features corresponding to the patient's chief complaint information input from the clinical feature storage unit are read out from the clinical feature storage unit and displayed on the display unit of the diagnosis support device. The disease selection unit selects one or more diseases using the patient's clinical feature information, which is a clinical feature selected from the read-out plurality of clinical features. The diagnosis support device outputs one or more diseases selected by the disease selection unit as candidates for diseases that the patient may be suffering from.

[0043] The purpose of this diagnostic support device is to suggest candidate diseases that a patient may be suffering from, based on information including findings about the patient. This diagnostic support device 1 includes a chief complaint group storage unit 11, a clinical feature storage unit 13, a disease storage unit 15, and a disease selection unit 17. In this way, by storing and displaying information corresponding to a diagnosis in stages, it is possible to make use of the doctor's knowledge and appropriately suggest candidate diseases. [Example]

[0044] For example, if a 50-year-old male patient complains of "fever of unknown origin," the device operates as follows: "fever of unknown origin" is entered into the device as the patient's chief complaint. Then, chief complaints related to the fever of unknown origin (e.g., multiple chief complaints belonging to any of the patient's medical history, physical findings, and test results) are displayed on the display. The physician selects a relevant chief complaint from the displayed chief complaints. Then, multiple clinical features related to the selected multiple chief complaints (including "fever of unknown origin") are displayed. Then, a corresponding clinical feature is selected from the displayed candidate clinical features (e.g., "prolonged fever," "night sweats," "weight loss," etc.). In this way, candidate clinical features related to the patient are selected and displayed, and appropriately selected. After a clinical feature related to the patient is selected, a candidate disease related to it (e.g., "infectious disease (tuberculosis, CMV infection, etc.)," ​​"autoimmune disease (SLE, polymyalgia rheumatica, etc.)," ​​or "malignant tumor (lymphoma, etc.)") is selected. The selected candidate disease may be displayed on the display. The acquired disease candidates may also be compared with additional patient information (blood test results, diagnostic imaging results, etc.) to narrow down more appropriate candidates. The device may output a list of disease candidates and recommendations for subsequent detailed examinations as the diagnostic result. The diagnostic support device of the present invention contributes to faster and more accurate diagnoses and also contributes to reducing the burden on medical professionals. Although not specifically illustrated, an appropriate medical information statement may be created in this example as well. [Example]

[0045] If a 40-year-old male patient complains of fever, cough, and fatigue, the device operates as follows: A doctor inputs the patient's chief complaint information. The patient's chief complaint information is then input into the device. The device then acquires clinical features related to "fever," "cough," and "fatigue." The device then acquires disease candidates (e.g., "influenza," "COVID-19," and "cold") that correspond to the clinical features. The device then ranks and displays the acquired disease candidates based on the latest clinical data, etc. Although not specifically illustrated, an appropriate medical information statement may be created in this example as well. [Example]

[0046] If a 60-year-old female patient complains of chest pain, shortness of breath, and cold sweats, the device will output possible illnesses such as myocardial infarction, angina pectoris, and anxiety neurosis, and recommend that she visit a medical institution immediately. The diagnosis support device of the present invention contributes to speeding up and improving the accuracy of diagnosis, and also contributes to reducing the burden on medical professionals. Although not specifically shown, an appropriate medical information statement may also be created in this example. [Example]

[0047] The example in Figure 5 is a conceptual diagram showing a diagnostic example for a patient whose chief complaint is "sore throat." The device 1 selects multiple candidate clinical features as shown in Figure 4 in relation to the patient's information and the chief complaint of "sore throat," and displays them on the display unit. The doctor interviews the patient and selects the clinical features that the patient complains of. In the example shown in Figure 5, a radio button is provided for each clinical feature, and the corresponding clinical feature is selected by clicking the radio button. Although not shown, in this example, the following medical information text was created and displayed on the display unit. "Suggestive signs of bacterial tonsillitis include a fever of 30°C or higher, suspected swelling of the anterior cervical lymph nodes, fever within 24 hours, and onset of symptoms within three days. The departments we recommend you visit are ENT or internal medicine. The doctor who saw this medical information recommended that the patient visit an ENT or internal medicine department at the hospital, and recorded the above medical information in the hospital's medical record. When the patient visits an ENT or internal medicine department at the same hospital, the above medical information will be used and the doctor will provide appropriate treatment. [Example]

[0048] The example in Figure 6 is a conceptual diagram showing a diagnosis example for a case where the chief complaint is coughing. Figure 6-1 shows a part of the screen, and Figure 6-2 shows a continuation of Figure 6-1. In this example, various candidate clinical features are displayed for the chief complaint of cough. For example, for "cough," device 1 stores "cough sounds" and "breathing sounds" as clinical features, and by clicking the audio button, the clinical features "cough sounds" and "breathing sounds" are reproduced. This makes it easier to analyze symptoms. Device 1 may also automatically collect the patient's breathing and cough sounds using a sound collection unit such as a microphone, and compare them with the clinical features "cough sounds" and "breathing sounds" stored in the memory unit to analyze the clinical features of the target patient. Although not shown, in this example, the following medical information text was created and displayed on the display unit. "I've had a cough for about 8 days to 2 weeks, and I've been expectorating a small amount of clear phlegm. There is a light breathing sound. Associated symptoms included nighttime aggravation and feeling asthma or rhoncus. Possibly related diseases and lifestyle habits include chronic respiratory disease. I'm eating less but still drinking water. The department we recommend you visit is internal medicine (if possible, respiratory medicine). Other possibilities include gastroenterology and cardiology." [Industrial Applicability]

[0049] The present invention can be used in the field of medical devices. [Explanation of symbols]

[0050] 1. Diagnostic support device 11 Main complaint group memory section 13 Clinical Features Memory Section 15 Disease Memory Section 17 Disease Selection Department 21 Main complaint group display section 23 Clinical Features Display Section 27 Disease display section

Claims

1. A diagnostic support device for creating a medical information statement, The diagnosis support device includes: a chief complaint group storage unit that stores a chief complaint group including a plurality of chief complaints; a clinical feature storage unit that stores a plurality of clinical features associated with the plurality of chief complaints; a chief complaint display unit for displaying a chief complaint selection for allowing the user to select one or more of the plurality of chief complaints; a clinical feature display unit for, when patient's chief complaint information is input to the diagnosis support device based on the chief complaint selection display, reading out a plurality of clinical features corresponding to the input patient's chief complaint information from the clinical feature storage unit, and performing a clinical feature selection display for allowing a user to select a plurality of clinical features corresponding to the patient's chief complaint information; a medical information statement creation unit that creates a medical information statement, which is a statement for understanding the pathology of a patient, using selected clinical features that are selected clinical features input to the diagnosis support device and information including the patient's chief complaint information; The medical information statement creation unit creates a medical information statement including terms for doctors or a medical information statement including terms for patients based on information regarding whether the medical information statement is for doctors or for patients. Diagnostic support device.

2. 2. The diagnosis support device according to claim 1, wherein the medical information text includes information about a recommended medical department within the hospital.

3. 2. A diagnostic support device according to claim 1, wherein the medical information statement creation unit has a trained model, and the medical information statement creation unit obtains the medical information statement by inputting the selected clinical features and the patient's chief complaint information into the trained model.

4. A program for causing a computer to function as a diagnostic support device for creating a medical information statement, The diagnosis support device includes: a chief complaint group storage unit that stores a chief complaint group including a plurality of chief complaints; a clinical feature storage unit that stores a plurality of clinical features associated with the plurality of chief complaints; a chief complaint display unit for displaying a chief complaint selection for allowing the user to select one or more of the plurality of chief complaints; a clinical feature display unit for, when patient's chief complaint information is input to the diagnosis support device based on the chief complaint selection display, reading out from the clinical feature storage unit a plurality of clinical features corresponding to the patient's chief complaint information input from the clinical feature storage unit, and performing a clinical feature selection display for allowing a user to select a plurality of clinical features corresponding to the patient's chief complaint information; a medical information statement creation unit that creates a medical information statement, which is a statement for understanding the pathology of a patient, using selected clinical features that are selected clinical features input to the diagnosis support device and information including the patient's chief complaint information; The medical information sentence creation unit is a diagnostic support device that creates a medical information sentence including terms for doctors or a medical information sentence including terms for patients based on information on whether the medical information sentence is for doctors or for patients. program.

Images