Diagnosis and treatment assistance method, diagnosis and treatment assistance program, and diagnosis and treatment assistance system

Abstract

A medical assistance system acquires one or more items in patient information associated with a patient (steps S30, S40), and inputs the one or more items into an estimation model, thereby acquiring candidates for a disease that the patient is likely to suffer from from a plurality of diseases (step S50). Then, the medical assistance system acquires additional request information to be additionally acquired from the patient on the basis of the disease candidates (step S70). Then, the medical assistance system presents the additional request information (step S90).

Description

Technical Field

[0001] The present invention relates to the assistance in diagnosis and treatment. Background Art

[0002] Especially in small and medium-sized hospitals or clinics, sometimes non-specialist doctors of infectious diseases need to handle infectious diseases. It is difficult for non-specialist doctors of infectious diseases to be proficient in all infectious diseases, and there is a great demand for the diagnostic and treatment support from specialist doctors of infectious diseases. In addition, with the increase in drug-resistant pathogens, the importance of the rational use of antimicrobial drugs has been recognized, and from this point of view, there is also a great expectation for the diagnostic and treatment support for non-specialist doctors of infectious diseases.

[0003] Compared with the number of medical institutions, there is an absolute shortage of specialist doctors of infectious diseases who are proficient in the diagnosis and treatment of infectious diseases, and there is a great demand for tools for supporting the diagnosis and treatment of infectious diseases. Conventionally, as an example of a tool for supporting the diagnosis and treatment of infectious diseases, a system using a database that aggregates information related to infectious diseases has been proposed (see Non-Patent Document 1). In this system, if patient information such as the presence or absence of cough and the presence or absence of fever is input, a list of infectious diseases that the patient may have is displayed, and in addition, information related to the infectious disease is also provided.

[0004] Prior Art Documents

[0005] Non-Patent Documents

[0006] Non-Patent Document 1: "Advancing the global effort against Infectious Diseases", [Online], [Searched on May 10, 2022], Internet <URL: https: / / www.gideononline.com / > Summary of the Invention

[0007] Technical Problem to be Solved by the Invention

[0008] According to the above system, even if a doctor is not a specialist in a given disease such as an infectious disease, he or she can know the candidates for the disease that the patient has and can easily perform diagnosis and treatment.

[0009] However, the disease information output in this system is determined based on limited patient information, and there is a need for a system that can output more reliable information.

[0010] The present invention has been studied in view of such actual situations, and its purpose is to provide a technology for assisting non-specialist doctors of a disease in diagnosing and treating the disease.

[0011] Solution for Solving the Above Technical Problem

[0012] One embodiment of the diagnostic and treatment assistance method of the present disclosure is a diagnostic and treatment assistance method including the following steps: a step of obtaining one or more items in patient information associated with a patient; a step of inputting the one or more items into an inference model to obtain candidates for diseases that the patient is likely to suffer from among multiple diseases; a step of obtaining additional requirement information that should be additionally obtained from the patient based on the disease candidates; and a step of presenting the additional requirement information.

[0013] A program for diagnostic and treatment assistance according to one embodiment of the present disclosure is executed by one or more processors of a computer, whereby the computer implements the above-described diagnostic and treatment assistance method.

[0014] One embodiment of the diagnostic and treatment assistance system of the present disclosure includes: a processor; a communication interface; and a storage device that stores an inference model configured to output candidates for diseases that a patient is likely to suffer from and that are associated with the patient information based on the input of one or more items of patient information. The processor obtains an input of the values of one or more items in the patient information associated with the patient, inputs the values of the one or more items into the inference model, thereby obtaining candidates for diseases that the patient is likely to suffer from, and obtains additional requirement information that should be additionally obtained from the patient based on the disease candidates. The communication interface presents the additional requirement information.

[0015] Advantageous Effects of the Invention

[0016] According to one embodiment of the present disclosure, as additional requirement information, an indication of what values of what items should be further studied is provided to a non-specialist doctor of a disease. Thereby, assistance related to the diagnosis and treatment of the disease is provided to the non-specialist doctor of the disease. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 is a diagram for explaining the provision of information of the diagnostic and treatment assistance system 100.

[0018] Figure 2 is a diagram showing an example of the generated structured data.

[0019] Figure 3 is a diagram showing the hardware configuration of the diagnostic and treatment assistance system 100.

[0020] Figure 4 is a diagram showing a mathematical formula as an example of an inference model.

[0021] Figure 5 is a flowchart of a process performed to present candidates for infectious diseases that a patient is likely to suffer from to the attending doctor.

[0022] Figure 6 is a diagram showing an example of an initial screen.

[0023] Figure 7 This is a diagram showing an example of a screen that displays the electronic medical record data input by the attending physician.

[0024] Figure 8 This is a diagram showing an example of an updated screen based on the instruction of step S90.

[0025] Figure 9 This is a diagram showing an example of a screen after a new physical observation result is input.

[0026] Figure 10 This is a diagram showing an example of a new display screen. DETAILED DESCRIPTION

[0027] Hereinafter, embodiments of the present disclosure will be described in detail with reference to the accompanying drawings. In addition, the same or corresponding parts in the drawings are denoted by the same reference numerals, and their repeated description will be omitted.

[0028] [OVERALL CONFIGURATION]

[0029] Figure 1 This is a diagram for explaining the provision of information of the medical treatment support system 100. In one implementation example, the medical treatment support system of the present disclosure provides information related to the diagnosis and / or treatment of diseases. Hereinafter, as an example of a disease, "infectious disease" will be described. In addition, the object of the disease in the present disclosure is not limited to infectious diseases, and may also be non-infectious diseases.

[0030] In Figure 1 the situation shown, the attending physician is a non-specialist in infectious diseases who examines a patient who does not know whether they have an infectious disease, and operates the information terminal 500. The information terminal 500 can communicate with the medical treatment support system 100.

[0031] When electronic medical record data is input from the information terminal 500, the medical treatment support system 100 generates structured data by analyzing the electronic medical record data. The structured data includes one or more items related to the patient and the values of each of the one or more items.

[0032] Figure 2 This is a diagram showing an example of the generated structured data. In Figure 2 the example shown, the structured data includes at least 5 items. And the structured data includes Figure 2 the values of each of the 5 items shown.

[0033] An example of an item is "fever", and an example of a value is "yes" for the item "fever". In one implementation example, if the patient's body temperature is 37.0 °C or higher, the value of the item "fever" in the structured data of a certain patient is "yes", and if the body temperature is less than 37.0 °C, it is "no".

[0034] Return Figure 1 , the diagnosis and treatment assistance system 100 uses one or more values in the structured data to determine candidates for infectious diseases that the patient is likely to have. The diagnosis and treatment assistance system 100 obtains diagnostic reference information about each of the determined candidates for infectious diseases from the diagnosis and treatment assistance information database. Then, the diagnosis and treatment assistance system 100 sends the candidates for infectious diseases and the diagnostic reference information for each candidate to the information terminal 500.

[0035] The diagnosis and treatment assistance system 100 determines items that require their values to improve the accuracy of the determined candidates as "insufficient items". The diagnosis and treatment assistance system 100 determines the insufficient items based on the above-mentioned structured data, and then sends the determined insufficient items to the information terminal 500.

[0036] The information terminal 500 displays the insufficient items. Accordingly, the attending physician inputs the values of the insufficient items into the information terminal 500. Thus, the values of the insufficient items are appended to the electronic medical record data.

[0037] The information terminal 500 sends the electronic medical record data after appending the values of the insufficient items to the diagnosis and treatment assistance system 100. The diagnosis and treatment assistance system 100 uses the electronic medical record data after appending the values of the insufficient items to generate structured data, and uses the generated structured data to re-determine candidates for infectious diseases that the patient may be infected with.

[0038] [Hardware Configuration]

[0039] Figure 3 is a diagram showing the hardware configuration of the diagnosis and treatment assistance system 100.

[0040] The diagnosis and treatment assistance system 100 is configured based on, for example, a personal computer. The diagnosis and treatment assistance system 100 may also be composed of a server that can be accessed from one or more information terminals through a network such as the Internet.

[0041] The diagnosis and treatment assistance system 100 includes a processor 101, a memory 200, and an input / output port 300. A mouse 400, a keyboard 120, and a display device 130 are connected to the input / output port 300.

[0042] The input / output port 300 may also be a communication interface for data communication via a network. Patient data is input to the input / output port 300. The patient data is, for example, electronic medical record data. In one implementation example, structured data is generated based on the electronic medical record data. Additionally, structured data may be input to the input / output port 300. That is, the attending physician may also extract the values of one or more items that make up the structured data from the electronic medical record data and input the values of the one or more items to the input / output port 300 via the information terminal 500.

[0043] The patient data 210, teacher data 220, inference model 230, correspondence database 240, medical treatment assistance information database 250, and auxiliary program 260 are stored in the memory 200.

[0044] The teacher data 220 is used for machine learning of the inference model 230 and is data related to multiple people. In the teacher data 220, structured data regarding each of multiple people is annotated with information on the incidence of infectious diseases, and the structured data is the data as described with reference to Figure 2 The incidence information includes whether or not an infectious disease has been contracted and the type of infectious disease contracted.

[0045] The inference model 230 is a program for performing calculations according to a model. An example of the algorithm of the inference model 230 is logistic regression, but the algorithm followed by the inference model 230 is not limited thereto. In one implementation example, inference models 230 are prepared for two or more types of infectious diseases respectively, and each inference model for an infectious disease derives whether or not a patient has contracted the infectious disease. In addition, each inference model for an infectious disease may also derive a probability (likelihood).

[0046] The correspondence database 240 is an example of correspondence information that establishes a correspondence between two or more types of infectious diseases and items (one or more items included in the structured data) used in the inference of the inference model 230. For example, the correspondence database 240 establishes a correspondence between the infectious disease "pyelonephritis" and three items (fever, frequent urination, and flank pain).

[0047] The medical treatment assistance information database 250 establishes a correspondence between two or more types of infectious diseases and medical treatment assistance information (information for assisting a doctor in medical treatment) for each infectious disease. The medical treatment assistance information is an example of additional information for an infectious disease.

[0048] The auxiliary program 260 includes a data analysis unit 261, a model generation unit 262, a first information generation unit 263, a data collection unit 264, a second information generation unit 265, and an output unit 266 as its functions. In one implementation example, the functions of the data analysis unit 261, the model generation unit 262, the first information generation unit 263, the data collection unit 264, the second information generation unit 265, and the output unit 266 are implemented by a processor 101 executing a given program.

[0049] The data analysis unit 261 generates structured data as described with reference to Figure 2 based on the patient data 210.

[0050] In one implementation example, the data parsing unit 261 extracts, from the text of the electronic medical record data stored as patient data 210, parts that constitute a pre-registered pattern by applying parsing processes such as morpheme analysis to the text. The "pre-registered pattern" refers to a pattern corresponding to the value of structured data. Then, the data parsing unit 261 identifies the meaning of the parts that constitute the pattern extracted from the electronic medical record data, and determines the value corresponding to the identified meaning. Then, the data parsing unit 261 generates structured data by combining one or more of the determined values.

[0051] An example of the registered pattern is "body temperature is xx °C", and the part "xx" represents a numerical value. When the electronic medical record data contains text such as "body temperature is 37.5 °C", the data parsing unit 261 extracts the text "body temperature is 37.5 °C" as a part that constitutes the pre-registered pattern. Then, the data parsing unit 261 identifies that the patient's body temperature is 37.5 °C as the meaning of the text. On the other hand, in the data parsing unit 261, it is set to determine the items and their values in the structured data based on the identified meaning. For example, it is set that if the body temperature is 37.0 °C or higher, the item "fever" and its value "yes" are determined, and if the body temperature is less than 37.0 °C, the item "fever" and its value "no" are determined. Thus, when the electronic medical record data contains text such as "body temperature is 37.5 °C", the data parsing unit 261 generates structured data with the value "yes" for the item "fever".

[0052] The items of the structured data include at least one of medical interview information, physical observation results, and examination results. The medical interview information, physical observation results, and examination results may also be included in the electronic medical record data.

[0053] The medical interview information includes the information obtained by the attending physician from the patient during the medical interview of the patient. The medical interview information may also include the information obtained by other past physicians from the patient during the medical interview of the patient. An example of the items included in the medical interview information is the patient's chief complaint.

[0054] The physical observation results include the information collected by the attending physician from the patient during the examination (visual inspection, auscultation, palpation, percussion, tendon reflex, light reflex, etc.).

[0055] The examination results include the results of examinations performed on the patient (blood test, CT (Computed Tomography) scan, body temperature measurement, weight measurement, blood pressure measurement, heart rate measurement, etc.).

[0056] The items of the structured data may also include items related to the patient's basic information (age, gender, degree of need for care, smoking history, etc.), allergy information, and / or past medical history.

[0057] The model generation unit 262 performs machine learning processing on the inference model 230 using the teacher data 220. In this specification, the inference model 230 on which the machine learning processing has been performed is also referred to as the "learned model".

[0058] The first information generation unit 263 applies the structured data generated from the patient data to the learned model to derive a determination result regarding an infectious disease that the patient is likely to have. In one implementation example, the first information generation unit 263 uses the learned model for each type of infectious disease to derive the probability (likelihood) that the patient has each infectious disease, and determines an infectious disease whose probability is equal to or greater than a given value as a candidate for the infectious disease.

[0059] The data collection unit 264 collects the diagnostic assistance information for each candidate determined by the first information generation unit 263 from the diagnostic assistance information database 250.

[0060] The second information generation unit 265 determines, for each candidate determined by the first information generation unit 263, an item with insufficient value in the structured data generated from the patient data 210 as a shortage item by referring to the correspondence database 240.

[0061] The output unit 266 generates image information for displaying the information determined in the assistance program 260, and instructs the information terminal 500 to display the image information.

[0062] [Determination of Candidates for Infectious Diseases]

[0063] An implementation example of determining candidates for infectious diseases by the first information generation unit 263 will be described. Referring to Figure 3 , as described above, in one implementation example, the inference model 230 is prepared for each type of infectious disease, and the first information generation unit 263 applies the structured data to each inference model 230 to thereby derive the probability that the patient has each infectious disease.

[0064] Figure 4 is a diagram showing a mathematical formula as an example of the inference model. Figure 4 The formula (1) shown follows logistic regression analysis. In the formula (1), the subscript represents the type of item included in the structured data. In the formula (1), the i-th item is defined as "1" to "i". The value of "i" may also vary depending on the type of infectious disease. That is, the type and number of items used in the inference of the probability of having each infectious disease may also vary for each infectious disease. The correspondence database 240 associates each infectious disease with the i types of items included in the formula (1) prepared for each infectious disease.

[0065] In formula (1), x is an explanatory variable, corresponding to the values of each item of structured data. For example, regarding the item "fever", when the value is "yes", x is 1, and when the value is "no", x is 0.

[0066] In formula (1), b is a coefficient. The values of each of b0 to b i are set in the machine learning for each infectious disease.

[0067] In formula (1), y is a target variable, representing the calculation result for each infectious disease. The possibility of suffering from each infectious disease is represented by the value of y.

[0068] The first information generation unit 263 applies the values of each item included in the structured data to the inference model for each infectious disease, thereby deriving the possibility of the patient suffering from each infectious disease.

[0069] Then, the first information generation unit 263 determines, as candidates for the infectious diseases suffered by the patient, the infectious diseases whose "possibility" value is equal to or higher than a given threshold.

[0070] In addition, the method for determining candidates for infectious diseases described with reference to Figure 4 is just an example. The determination of candidates for infectious diseases can also be achieved by any other method such as multivariate analysis using one or more types included in the structured data as variables.

[0071] Furthermore, when the structured data is applied to the inference model, for some infectious diseases, sometimes the structured data does not include the values of some items specified in the inference model. That is, in the structured data, sometimes some of the items specified in the inference model are missing.

[0072] For example, suppose the inference model for the infectious disease "pyelonephritis" specifies three items (fever, frequent urination, and flank pain). When the attending physician adds information about fever and frequent urination to the electronic medical record data but does not add information about flank pain, the structured data includes the values of the items "fever" and "frequent urination" but does not include the value of the item "flank pain". In such a case, the first information generation unit 263 performs missing value processing in the derivation of the possibility using the inference model. An example of missing value processing is to randomly supplement the value of the missing item with the values of other items used in the formula of the same inference model. Another example is to supplement the value of the missing item with the average value of the values of other items used in the formula of the same inference model. For example, when the value of the item "flank pain" is missing, the value is supplemented with the average value of the values of the items "fever" and "frequent urination".

[0073] [Determination of Missing Items]

[0074] The second information generation unit 265 determines, for each infectious disease identified as a candidate by the first information generation unit 263, whether there is a missing item as described above. More specifically, for each infectious disease identified as a candidate by the first information generation unit 263, if there is an item associated with the infectious disease in the correspondence database 240 that is not included in the structured data, the second information generation unit 265 determines that there is a missing item. In addition, when all the items associated with the infectious disease in the correspondence database 240 are included in the structured data, the second information generation unit 265 determines that there is no missing item.

[0075] Moreover, when there is a missing item, the second information generation unit 265 designates the item as a lacking item.

[0076] [Important factor]

[0077] In the memory 200, it is also possible to store, for each infectious disease, the score of the importance of each item calculated when generating the inference model. The second information generation unit 265 can also, for each infectious disease, designate as important factors the items among the items included in the structured data that have a score of importance equal to or greater than a given value.

[0078] "Equal to or greater than a given value" may also mean that the score of the importance has a "positive" value equal to or greater than a certain value.

[0079] "Equal to or greater than a given value" may also mean that the absolute value of the score of the importance is equal to or greater than a given value. When the score of the importance of a certain item is a positive value, the magnitude of the absolute value of the score indicates the likelihood of contracting the infectious disease. When the score of the importance of a certain item is a negative value, the magnitude of the absolute value of the score indicates the likelihood of not contracting the infectious disease. When "equal to or greater than a given value" means that the absolute value of the score of the importance is equal to or greater than a given value, the items designated as important factors characterize the likelihood of contracting the infectious disease or the likelihood of not contracting the infectious disease.

[0080] [Flow of processing]

[0081] Figure 5 is a flowchart of the processing implemented to present to the attending physician the candidates for infectious diseases that the patient may have contracted. In one implementation example, Figure 5 the processing is implemented by the processor 101 of the medical treatment assistance system 100 executing a given program (medical treatment assistance program). Hereinafter, Figure 5 the processing will be described. In one implementation example, Figure 5The processing starts when the medical treatment support system 100 is requested to start providing information from the information terminal 500. In one implementation example, the information terminal 500 starts according to the application program for medical treatment support and performs an operation for requesting the start of information provision in the application program, and requests the start of information provision from the medical treatment support system 100.

[0082] In step S10, the medical treatment support system 100 instructs the information terminal 500 to display the initial screen. Thereby, the information terminal 500 displays the initial screen on the display of the information terminal 500.

[0083] Figure 6 is a diagram showing an example of the initial screen. In Figure 6 the screen 600 includes a first communication box 610, columns 621 to 627, 630, and a second communication box 640.

[0084] The first communication box 610 displays information for the attending physician to communicate with the medical treatment support system 100. The first communication box 610 includes a column 611 and a button 619. In Figure 6 the example of, a message for the attending physician (please enter the medical record) is displayed in the column 611. Operating the button 619 is used to instruct the medical treatment support system 100 to provide information for the data input to the screen.

[0085] The electronic medical record data of the patient is respectively input in the columns 621 to 627. More specifically, the basic information of the patient (age, gender, degree of need for care, smoking history, etc.) is input in the column 621. The past medical history of the patient is input in the column 622. The allergy information of the patient is input in the column 623. The vital sign information in the results of the examinations performed on the patient is input in the column 624. The results of examinations other than the vital sign information in the results of the examinations performed on the patient (results of blood tests, CT scan results, etc.) are input in the column 625. The physical observation results are input in the column 626. The interview information (chief complaint, etc.) is input in the column 627. In addition, the input method of the electronic medical record data in the above description is only an example. The medical treatment support system 100 can also accept the input of electronic medical record data described in text form, extract the data displayed in the columns 621 to 627 from the electronic medical record data, and thereby generate Figure 6 the screen data as shown.

[0086] In the column 630, information related to candidates for infectious diseases that the medical treatment support system 100 provides to the attending physician is displayed.

[0087] The second communication box 640 is used for the attending physician to communicate with the infectious disease specialist.

[0088] The attending physician who visually confirmed the screen 600 inputs electronic medical record data in columns 621 to 627. Figure 7 It is a figure showing an example of a screen that displays the electronic medical record data input by the attending physician. In Figure 7 , on the screen 601, the electronic medical record data input by the attending physician is respectively displayed in columns 621 to 627.

[0089] After that, the attending physician operates the button 619. Thereby, the input electronic medical record data is sent from the information terminal 500 to the medical treatment assistance system 100.

[0090] Return Figure 5 , in step S20, the medical treatment assistance system 100 determines whether information provision is indicated by the information terminal 500. The medical treatment assistance system 100 repeats the control of step S20 (no in step S20) until it is determined that information provision is indicated. If the medical treatment assistance system 100 determines that information provision is indicated (yes in step S20), the control proceeds to step S30.

[0091] In step S30, the medical treatment assistance system 100 reads the electronic medical record data sent from the information terminal 500.

[0092] In step S40, the medical treatment assistance system 100 uses the electronic medical record data read in step S30 to generate structured data as described with reference to Figure 2 .

[0093] In step S50, the medical treatment assistance system 100 uses the structured data generated in step S40 to determine one or more candidates for infectious diseases that the patient may have.

[0094] In step S60, the medical treatment assistance system 100 obtains the medical treatment assistance information for each of the one or more candidates for infectious diseases determined in step S50 by referring to the medical treatment assistance information database 250.

[0095] In step S70, the medical treatment assistance system 100 respectively determines the missing items for the one or more candidates for infectious diseases determined in step S50.

[0096] In step S80, the medical treatment assistance system 100 respectively determines the important factors for the one or more candidates for infectious diseases determined in step S50.

[0097] In step S90, the medical treatment assistance system 100 uses the results of steps S50 to S80 to generate a display screen that displays information such as candidates for infectious diseases, and instructs the information terminal 500 to display the generated display screen. Thereby, the display screen is updated in the information terminal 500. After that, the medical treatment assistance system 100 returns the control to step S20.

[0098] Figure 8 This is a diagram showing an example of an updated screen based on the indication in step S90. In Figure 8 it, screen 602 is Figure 7 an example of the updated screen of screen 601.

[0099] Compared with Figure 7 the screen 601, in Figure 8 the screen 602, one or more candidates that the patient may have are shown in column 630. Each of the one or more labels displayed in column 630 corresponds to one or more candidates for infectious diseases determined in step S50, respectively.

[0100] In Figure 8 column 630, two labels representing two infectious diseases (pyelonephritis and urinary tract infectious diseases) are shown, respectively.

[0101] More specifically, column 630 displays two labels in a way that the label content can be switched. In Figure 8 it, the state of the content of the label showing "pyelonephritis" among the label showing "pyelonephritis" and the label showing "urinary tract infectious diseases" is shown. The attending physician can operate the information terminal 500 to display the content of the label showing "urinary tract infectious diseases" in column 630.

[0102] The content of each label in column 630 is the diagnostic assistance information for the corresponding infectious disease. The diagnostic assistance information for each candidate for infectious diseases obtained in step S60 is displayed in each label in column 630.

[0103] Compared with Figure 7 it, column 612 is added to the first communication box 610 in Figure 8 . When the diagnostic assistance system 100 determines a shortage item in step S70, a column is added to the first communication box 610, and a message reminding to input the value of the shortage item is displayed in the added column. Columns can also be added for each shortage item. For example, when two candidates are determined and one shortage item is determined for each of the two candidates, two columns can be added. A message reminding to input the value of the shortage item determined for the first candidate can be displayed in one of the two added columns, and a message reminding to input the value of the shortage item determined for the second candidate can be displayed in the remaining one of the two added columns.

[0104] In Figure 8In the example, a message "Please enter observations related to flank pain" is displayed in column 612. This message corresponds to "flank pain" being determined as a deficient item. More specifically, the medical treatment assistance system 100 displays this message in order to remind the attending physician to further input the value of the determined deficient item (as a physical observation).

[0105] Compared with Figure 7 in Figure 8 the example, a part of the text displayed in columns 621 to 627 is emphasized by being surrounded by a frame. The emphasized text corresponds to the important factors determined in step S80. In one implementation example, the important factors are emphasized as the determined bases related to the infectious diseases determined as candidates.

[0106] More specifically, the medical treatment assistance system 100 emphasizes the part corresponding to the important factors determined in step S80 in the text (electronic medical record data) displayed in columns 621 to 627. The part corresponding to the important factors refers to the part in the electronic medical record data that is used to generate the value of the item determined as an important factor in the generation of structured data.

[0107] For example, consider the following situation: The infectious disease "pyelonephritis" is determined as a candidate, and the correspondence database 240 specifies three items (fever, frequent urination, flank pain) as the items used in the inference of the infectious disease "pyelonephritis", and "fever" and "frequent urination" are determined as important factors. In addition, in this case, the structured data includes the value "yes" of the item "fever" generated based on the text "body temperature 37.2°C" in the electronic medical record data, and also includes the value "yes" of the item "frequent urination" generated based on the text "increased frequency of urination" in the electronic medical record data.

[0108] In the above situation, the part of the electronic medical record data corresponding to the important factor "fever" is "body temperature 37.2°C", and the part corresponding to the important factor "frequent urination" is "increased frequency of urination". Therefore, in the above situation, "body temperature 37.2°C" and "increased frequency of urination" in the electronic medical record data are emphasized in columns 621 to 627.

[0109] In the processing described above Figure 5 when the attending physician inputs the electronic medical record data of the patient, candidates for infectious diseases that the patient may have are presented to the attending physician, and messages for reminding the input of the values of deficient items are presented, which are the items considered deficient for determining the presented candidates as the infectious diseases suffered by the patient. Thereby, information about what kind of items are deficient items is provided to the attending physician, which is used for the final diagnosis using the presented candidates.

[0110] In addition, the attending physician can also observe the above information and append the values of the insufficient items. The medical treatment assistance system 100 can also use the initially input values and the values of the insufficient items appended by the attending physician to re-determine the candidates for infectious diseases that the patient may be infected with.

[0111] More specifically, in step S90, the medical treatment assistance system 100 instructs the update of the screen, whereby the information terminal 500 Figure 8 as shown in the middle column 612, displays a message reminding the input of the values of the insufficient items. Based on the message displayed in column 612, the attending physician inputs new physical observation results on the screen.

[0112] Figure 9 is a diagram showing an example of the screen after the new physical observation results are input. Compared with the Figure 8 screen 602, Figure 9 the screen 603 in

[0113] further includes in column 626 the text "Abdomen_Inflated_Crying 'Ouch!' when only gently pressing the left and right sides of the abdomen" corresponding to the new physical observation results. After that, the attending physician operates the button 619.

[0114] After the medical treatment assistance system 100 instructs the update of the screen in step S90, it returns the control to step S20. Based on the operation of the attending physician on the button 619, the medical treatment assistance system 100 determines that information provision has been instructed (Yes in step S20), and the control proceeds to step S30.

[0114] In step S30, the medical treatment assistance system 100 reads the electronic medical record data including the above new physical observation results. Then, in step S40, the medical treatment assistance system 100 uses the electronic medical record data read out in step S30 to generate new structured data. Then, in step S50, the medical treatment assistance system 100 determines the candidates for infectious diseases using the new structured data, in step S60, obtains the medical treatment assistance information of the candidates determined in step S50, in step S70, determines the insufficient items of the candidates determined in step S50, and in step S80, determines the important factors of the candidates determined in step S50. Then, in step S90, the medical treatment assistance system 100 uses the results of steps S50 to S80 implemented after reading out the electronic medical record data including the new physical observation results to generate a new display screen and instructs the information terminal 500 to display the new display screen. As a result, the display screen is updated in the information terminal 500.

[0115] Figure 10 is a diagram showing an example of the new display screen. Compared with the Figure 9 screen 603, in Figure 10 the screen 604, column 630 only includes "Pyelonephritis" as a candidate for infectious diseases. In addition, compared with Figure 9Compared with screen 603, in Figure 10 In screen 604, as emphasized text, the text "It hurts just by gently pressing the left and right sides of the abdomen!" in column 626 was added.

[0116] In reference to Figure 9 and Figure 10 In the example described, according to the text "It hurts just by gently pressing the left and right sides of the abdomen!", the value "Yes" of the item "Pain in the side abdomen" was added to the structured data. And, in Figure 8 Among the two candidates (pyelonephritis and urinary tract infectious diseases) shown, the possibility of developing a urinary tract infectious disease decreased. That is, in this example, by inputting the values of the missing items, the number of candidates for infectious diseases decreased. Thus, more accurate information can be provided to the attending physician as information on the candidates for infectious diseases that the provided patient may be infected with.

[0117] In the present embodiment, diagnostic assistance information regarding each candidate is also provided. Thus, even when the attending physician is not a specialist in infectious diseases, useful information can be provided in the diagnosis and treatment of the patient.

[0118] In the present embodiment, for each candidate, the part corresponding to the important factor in the electronic medical record data is presented by highlighting display. Thus, useful information can be provided when the attending physician makes a final diagnosis using the presented candidate.

[0119] In addition, in the present embodiment, the diagnostic assistance information database 250 is stored in the memory 200 of the diagnostic assistance system 100, but it may also be stored in a storage device other than the diagnostic assistance system 100. The diagnostic assistance system 100 may also access this storage device to obtain the diagnostic assistance information for each of one or more candidates for infectious diseases.

[0120] (Modification example)

[0121] In the present embodiment, a diagnostic assistance system has been described, which determines the missing items that can improve the accuracy of candidates for infectious diseases when input into the inference model, and presents these missing items to the attending physician. In addition, in the present embodiment, the missing items are an example of the information (additional requirement information) that should be additionally obtained from the patient. When obtaining the missing items, a correspondence database (an example of correspondence information) is referred to.

[0122] In the present embodiment, in step S90, instead of, or in addition to, the lacking items, information based on the requirements of a specialist doctor may also be presented. The information based on the requirements of a specialist doctor refers, for example, to information of the type required by a specialist doctor regarding an infectious disease of the type determined as a candidate, which is information of a type not included in the structured data (electronic medical record data read in step S30) generated in step S40. In this context, the information based on the requirements of a specialist doctor is another example of additional requirement information.

[0123] The types of information required by a specialist doctor may also be pre-stored, for example, as a requirement correspondence table in a storage device such as the memory 200 according to the types of infectious diseases. In this context, the requirement correspondence table constitutes an example of information describing the relationships between multiple infectious diseases and additional requirement information respectively. In one implementation example, when the medical treatment assistance system 100 infers a candidate for an infectious disease through an inference model, it reads from the storage device the requirements of the specialist doctor corresponding to the candidate for the infectious disease in the requirement correspondence table. Then, the information of the type not included in the structured data generated in step S40 among the one or more types of information included in the read requirements is determined as the information based on the requirements of the specialist doctor. Thereby, the attending doctor can obtain in advance (before directly contacting the specialist doctor) the types of information required by the specialist doctor for the infectious disease candidate. Therefore, the attending doctor can reduce the number of times and / or the time of communication with the specialist doctor, and thus respond to the patient more quickly.

[0124] In addition, in addition to, or instead of, the determination of the lacking items, the determination of the information based on the requirements of the specialist doctor may also be implemented in step S70.

[0125] The determination of the information based on the requirements of the specialist doctor may be implemented by using, in addition to, or instead of, referring to the above-mentioned requirement correspondence table, the response obtained by the attending doctor from the specialist doctor. More specifically, the attending doctor may also present the electronic medical record data and the infectious disease candidate to the specialist doctor. Accordingly, the specialist doctor may also require in the response to the attending doctor the information required for determining the infectious disease that the patient is likely to have. The medical treatment assistance system 100 may also determine the "required information" included in the above-mentioned response from the specialist doctor as the "information based on the requirements of the specialist doctor". The medical treatment assistance system 100 may also determine the "information based on the requirements of the specialist doctor" by performing natural language processing on the response from the specialist doctor.

[0126] In one implementation example, the medical treatment assistance system 100 may also receive an operation for sending information to a specialist doctor on a display screen that displays information such as candidates for infectious diseases in step S90. The medical treatment assistance system 100 may also send the electronic medical record data and the infectious disease candidates to the address associated with the specialist doctor according to this operation. After that, when the medical treatment assistance system 100 receives a reply from the specialist doctor, it may update the above display screen by adding the reply. Thus, the reply from the specialist doctor can be provided to the attending doctor. The information sent to the specialist doctor may also be the display screen generated in step S90. Thus, the patient information, the disease candidates, and the diagnosis result of the attending doctor are sent to the specialist doctor.

[0127] The medical treatment assistance system may also present both the information required for the inference model to infer infectious disease candidates and the requirements of the specialist doctor to the attending doctor. After obtaining the observation results of the patient while referring to the requirements of the specialist doctor, the attending doctor may additionally input the observation results to the medical treatment assistance system. The medical treatment assistance system may use the information (observation results) additionally input as described above in the re-inference of the inference model. The content of the "information required for the inference model to infer infectious disease candidates" may overlap with the content of the "additionally input information" as described above. The medical treatment assistance system may present these two types of information to the attending doctor in a distinguishable manner (for example, visually) or in the same manner.

[0128] (Reference data)

[0129] As the information referred to in the determination of the lacking items in step S70, "reference data" may also be used. Information for diagnosis or treatment of each disease is stored in the reference data. In step S70, the medical treatment assistance system 100 determines the lacking items for each of the one or more disease candidates determined in step S50 by referring to the reference data. In one implementation example, for the disease determined in step S50, the medical treatment assistance system 100 determines the types of information that are not included in the acquired information among the types of information stored in the reference data as the lacking items.

[0130] Among the "information for diagnosis or treatment", the information for diagnosis and the information for treatment may be distinguishable from each other. For example, regarding the disease "pyelonephritis", the types of information for diagnosis include "fever", "frequency of urination", "flank pain", "age", and the types of information for treatment include "past medical history", "age".

[0131] The instruction for information provision from the information terminal 500, which is the determination object in step S20, may also include the purpose of information provision (diagnosis or treatment). In the determination of the lacking items in step S70, the medical treatment support system 100 may also use, according to the purpose included in the above instruction, either the information for diagnosis or the information for treatment in the "information for diagnosis or treatment". More specifically, the medical treatment support system 100 may use the information for diagnosis when the purpose is "diagnosis", and use the information for treatment when the purpose is "treatment".

[0132] In the "information for diagnosis or treatment", one or more types of information may also be classified into "essential information" and "optional information". Among the information used in the diagnosis of the above disease "pyelonephritis" ("fever", "frequency of micturition", "flank pain", "age"), it may also be classified in such a way that "fever", "frequency of micturition", and "flank pain" are "essential information", and "age" is "optional information".

[0133] The lacking items determined by using the reference data may also be output to the above display screen regarding step S90.

[0134] Furthermore, in the above display screen, it may also include a display indicating whether the information about the patient obtained by the medical treatment support system 100 includes all the types determined as the above "essential information". An example of this display is a "qualified" mark displayed when the information about the patient obtained by the medical treatment support system 100 includes all the types of the above "essential information". Another example is an "unqualified" mark displayed when the information about the patient obtained by the medical treatment support system 100 does not include at least a part of the types of the above "essential information".

[0135] Another example of the above display may also represent the ratio of the types included in the information about the patient obtained by the medical treatment support system 100 to all the types of the above "essential information". In one example, the display representing this ratio is a numerical value (100%, 50%, etc.). In another example, when there are 3 types of essential information ("fever", "frequency of micturition", "flank pain"), the display representing the above ratio is "good" when all 3 types are included, "qualified" when 1 or 2 types are included, and "unqualified" when none are included at all.

[0136] (Display of guidelines)

[0137] The medical assistance system 100 may also add information extracted from a guideline to the display screen generated in step S90. In one implementation example, the medical assistance system 100 adds a diagnostic guideline or a treatment guideline corresponding to the disease determined in step S50 to the above display screen. An example of the added guideline is the JAID / JSC Infectious Disease Treatment Guideline (https: / / www.kansensho.or.jp / uploads / files / guidelines / gui de line_JAID-JSC_2015_urinary-tract.pdf).

[0138] The medical assistance system 100 may also extract a part corresponding to the determined disease from the guideline and add only the extracted part to the above display screen. In addition, the medical assistance system 100 may add only either the article of the above guideline itself or the access information (such as a URL) of the above guideline to the above display screen. Further, for a disease based on the observation results of the attending physician, the medical assistance system 100 may similarly add the information extracted from the guideline to the display screen.

[0139] The medical assistance system 100 may also select the type of the additional guideline according to the purpose included in the information provision from the information terminal 500 which is the determination object in step S20. More specifically, the medical assistance system 100 may add information about a diagnostic guideline to the display screen when the purpose is "diagnosis", and add information about a treatment guideline to the display screen when the purpose is "treatment".

[0140] [Solution]

[0141] Those skilled in the art understand that the above multiple exemplary embodiments are specific examples of the following solutions.

[0142] (Item 1) It may be that a medical assistance method of one solution includes: a step of obtaining one or more items in patient information associated with a patient; a step of inputting the one or more items into an inference model to obtain candidates for diseases that the patient may have from multiple diseases; a step of obtaining additional requirement information that should be additionally obtained from the patient based on the candidates for the diseases; and a step of presenting the additional requirement information.

[0143] According to the medical assistance method of Item 1, an indication of what values of what items should be further studied is provided to a non-specialist doctor of a disease as additional requirement information. Thereby, assistance related to the diagnosis and treatment of the disease is provided to a non-specialist doctor of the disease.

[0144] (Item 2) Alternatively, in the diagnostic assistance method of Item 1, the step of obtaining the additional requirement information includes: referring to reference data associating diseases with information for diagnosis or treatment; and determining the additional requirement information based on the information for diagnosis or treatment associated with the candidate of the disease.

[0145] According to the diagnostic assistance method of Item 2, information can be provided to the attending physician for studying the appropriateness of disease candidates and treatment methods.

[0146] (Item 3) Alternatively, in the diagnostic assistance method of Item 2, the reference data includes essential information and optional information as the information for diagnosis or treatment, and the diagnostic assistance method further includes a step of presenting a judgment result on whether the one or more items include all the essential information.

[0147] According to the diagnostic assistance method of Item 3, information can be provided to the attending physician for studying whether additional requirement information actually needs to be input.

[0148] (Item 4) Alternatively, in the diagnostic assistance method of Item 2 or Item 3, the information for diagnosis or treatment includes diagnostic information and treatment information, and the diagnostic assistance method further includes a step of accepting diagnosis or treatment as the purpose of obtaining the candidate of the disease. Determining the additional requirement information includes: when the purpose is diagnosis, determining the additional requirement information based on the diagnostic information; and when the purpose is treatment, determining the additional requirement information based on the treatment information.

[0149] According to the diagnostic assistance method of Item 4, appropriate information can be provided to the attending physician according to the situation (treatment or diagnosis) faced by the attending physician.

[0150] (Item 5) Alternatively, the diagnostic assistance method of any one of Items 1 to 5 further includes a step of displaying a diagnostic guideline or a treatment guideline.

[0151] According to the diagnostic assistance method of Item 5, more information about the disease can be provided to the attending physician.

[0152] (Item 6) Alternatively, the diagnostic assistance method of any one of Items 1 to 5 further includes: a step of receiving the diagnosis result of the attending physician together with the patient information; and a step of sending the patient information, the candidate of the disease, and the diagnosis result of the attending physician to a specialist.

[0153] According to the diagnostic assistance method of Item 6, more information is given to the specialist, and thus, a more appropriate response can be made by the specialist.

[0154] (Item 7) Alternatively, the medical treatment assistance method according to any one of Items 1 to 5 may further include a step of emphasizing and displaying the basis for the inference model to determine the candidate for the disease.

[0155] According to the medical treatment assistance method of Item 7, it is possible to provide the basis for the judgment of the inference as a black box. (Item 8) Alternatively, in the medical treatment assistance method according to any one of Items 1 to 7, the additional required information includes information that can improve the accuracy of the candidate for the disease if input into the inference model, and the step of obtaining the additional required information includes: obtaining the additional required information by referring to the correspondence information that associates each of the multiple diseases with the items of the patient information.

[0156] According to the medical treatment assistance method of Item 8, a process for obtaining additional required information is provided more specifically.

[0157] (Item 9) Alternatively, in the medical treatment assistance method according to any one of Items 1 to 8, the additional required information includes the lacking items that are not included in the one or more items among the items of the patient information corresponding to the candidate for the disease, and the step of obtaining the additional required information includes determining the lacking items.

[0158] According to the medical treatment assistance method of Item 9, more specific information is provided as the additional required information.

[0159] (Item 10) Alternatively, in the medical treatment assistance method according to any one of Items 1 to 9, the step of obtaining the additional required information includes: obtaining the additional required information corresponding to the obtained candidate for the disease from a storage device that stores information describing the relationship between each of the multiple diseases and the additional required information; or obtaining the additional required information by using the response obtained by presenting the obtained candidate for the disease to a specialist doctor.

[0160] According to the medical treatment assistance method of Item 10, more specific information is provided as the additional required information.

[0161] (Item 11) Alternatively, in the medical treatment assistance method according to any one of Items 1 to 10, the step of obtaining the values of the one or more items includes: receiving the input of the electronic medical record data of the patient; and generating the values of the one or more items by using the electronic medical record data.

[0162] According to the medical treatment assistance method of Item 11, in order to receive the provision of information, a non-specialist doctor can input the patient's medical record data instead of inputting the values of the one or more items. Therefore, the work of non-specialist doctors becomes easier.

[0163] (Item 12) Alternatively, the method for assisting diagnosis and treatment according to any one of Items 1 to 11 may further include a step of suggesting candidates for the disease, and the step of suggesting candidates for the disease includes: suggesting items having an importance level equal to or higher than a given value in the determination of candidates for the disease among the above-mentioned one or more items, or additional information on candidates for the disease.

[0164] The method for assisting diagnosis and treatment according to Item 12 provides information helpful for understanding the reasons for which the suggested candidates are suggested.

[0165] (Item 13) Alternatively, in the method for assisting diagnosis and treatment according to any one of Items 1 to 12, the patient information includes at least one of interview information, physical observation results, and examination results.

[0166] According to the method for assisting diagnosis and treatment of Item 10, candidates for infectious diseases that the patient may have are provided based on the input of information that can be obtained by the attending physician of the patient.

[0167] (Item 14) Optionally, a program for assisting diagnosis and treatment of one aspect is executed by one or more processors of a computer, whereby the computer implements the method for assisting diagnosis and treatment according to any one of Items 1 to 13.

[0168] According to the program for assisting diagnosis and treatment of Item 14, hints on which item values should be further studied are provided to non-specialist doctors of the disease as additional requirement information. Thereby, assistance related to the diagnosis and treatment of the disease is provided to non-specialist doctors of the disease.

[0169] (Item 15) Optionally, a system for assisting diagnosis and treatment of one aspect includes: a processor; a communication interface; and a storage device that stores an inference model configured to output candidates for diseases that the patient may have associated with the patient information based on the input of one or more items of patient information. The processor obtains the input of the values of one or more items in the patient information associated with the patient, inputs the values of the one or more items into the inference model, thereby obtaining candidates for diseases that the patient may have, and based on the candidates for the disease, obtains additional requirement information that should be further obtained from the patient, and the communication interface presents the additional requirement information.

[0170] According to the system for assisting diagnosis and treatment of Item 15, hints on which item values should be further studied are provided to non-specialist doctors of the disease as additional requirement information. Thereby, assistance related to the diagnosis and treatment of the disease is provided to non-specialist doctors of the disease.

[0171] The embodiments disclosed this time shall be considered illustrative in all aspects and not restrictive. The scope of the present disclosure is not shown by the description of the above embodiments, but by the claims, and is intended to include all changes within the meaning and scope equivalent to the claims. In addition, each technology in the embodiments can be implemented independently, and can also be implemented in combination with other technologies in the embodiments as much as possible according to needs.

[0172] Description of Reference Numerals

[0173] 100 Diagnostic Assistance System

[0174] 101 Processor

[0175] 500 Information Terminal

[0176] 600, 601, 602, 603, 604 Screens

[0177] 610 First Communication Frame

[0178] 611, 612, 621, 622, 623, 624, 625, 626, 627, 630 Columns

[0179] 619 Button

[0180] 640 Second Communication Frame.

Claims

1. A method for assisting diagnosis and treatment, characterized in that, Comprising: a step of obtaining one or more items from patient information associated with a patient; a step of inputting the one or more items into an inference model to obtain candidates for diseases that the patient may have from multiple diseases; a step of obtaining additional requirement information that should be additionally obtained from the patient based on the candidates for the diseases; and a step of presenting the additional requirement information.

2. The diagnostic and treatment assistance method according to claim 1, wherein The step of obtaining the additional requirement information includes: referring to reference data that associates diseases with information for diagnosis or treatment; and determining the additional requirement information based on the information for diagnosis or treatment associated with the candidates for the diseases.

3. The diagnostic and treatment assistance method according to claim 2, wherein The reference data includes essential information and optional information as the information for diagnosis or treatment, and the medical treatment assistance method further comprises a step of presenting a judgment result on whether the one or more items include all of the essential information.

4. The diagnostic and treatment assistance method according to claim 2, characterized in that The information for diagnosis or treatment includes diagnostic information and treatment information, and the medical treatment assistance method further comprises a step of accepting diagnosis or treatment as the purpose of obtaining the candidates for the diseases, Determining the additional requirement information includes: in the case where the purpose is diagnosis, determining the additional requirement information based on the diagnostic information; and in the case where the purpose is treatment, determining the additional requirement information based on the treatment information.

5. The diagnostic and treatment assistance method according to claim 1, wherein It further includes a step of displaying a diagnostic guideline or a treatment guideline regarding the candidates for the diseases.

6. The diagnostic and treatment assistance method according to claim 1, wherein It further includes: a step of accepting the diagnosis result of the attending physician together with the patient information; and a step of sending the patient information, the candidates for the diseases, and the diagnosis result of the attending physician to a specialist.

7. The diagnostic and treatment assistance method according to claim 1, characterized in that It further includes a step of prominently displaying in the patient information the basis on which the inference model determines the candidates for the diseases.

8. The diagnostic and treatment assistance method according to claim 1, wherein The additional requirement information includes information that can improve the accuracy of the candidates for the diseases if input into the inference model, and the step of obtaining the additional requirement information includes: obtaining the additional requirement information by referring to correspondence information that associates each of the multiple diseases with items of the patient information.

9. The diagnostic and treatment assistance method according to claim 1, characterized in that, The additional requirement information includes deficiency items that are not included in the one or more items among the items of the patient information corresponding to the candidates for the diseases, and the step of obtaining the additional requirement information includes determining the deficiency items.

10. The diagnostic and treatment assistance method according to claim 1, characterized in that, The step of obtaining the additional requirement information includes: obtaining, from a storage device, additional requirement information corresponding to the obtained candidates for the diseases, where the storage device stores information describing the relationship between each of the multiple diseases and the additional requirement information; or obtaining the additional requirement information using the response obtained by presenting the obtained candidates for the diseases to a specialist.

11. The medical diagnosis and treatment assistance method according to claim 1, wherein The step of obtaining the values of the one or more items includes: accepting the input of the patient's electronic medical record data; and generating the values of the one or more items using the electronic medical record data.

12. The diagnostic and treatment assistance method according to claim 1, wherein It further comprises a step of presenting the candidates for the diseases, and the step of presenting the candidates for the diseases includes: presenting items among the one or more items that have an importance level equal to or higher than a given value in the determination of the candidates for the diseases or additional information about the candidates for the diseases.

13. The diagnostic and treatment assistance method according to claim 1, wherein, The patient information includes at least one of the medical interview information, physical observation results, and examination results.

14. A program for assisting diagnosis and treatment, characterized in that, Executed by one or more processors of a computer, thereby causing the computer to implement the medical treatment assistance method according to claim 1.

15. A diagnosis and treatment assistance system, characterized in that, Comprising: a processor; A communication interface; And A storage device that stores an inference model configured to output candidates for diseases that the patient is likely to suffer from, which are associated with the patient information, based on the input of one or more items of the patient information. The processor obtains the input of the values of one or more items in the patient information associated with the patient. Inputs the values of the one or more items into the inference model, thereby obtaining candidates for diseases that the patient is likely to suffer from. Based on the candidates for the diseases, obtains additional requirement information that should be additionally obtained from the patient. The communication interface prompts the additional requirement information.

Images