JP6571346B2 - Medical information processing apparatus and medical information processing method - Google Patents

Description

  Embodiments described herein relate generally to a medical information processing apparatus and a medical information processing method.

  When a doctor makes a diagnosis, the diagnosis is made based on the doctor's own experience and information investigated from medical books, case reports, case databases, and the like. The same applies to determining a treatment plan other than diagnosis. When planning treatment with radiation therapy, the patient's age, patient status, primary lesion, stage, histopathology, extent of lesion development, risk organ location, radical or symptomatic, were done in the past It is necessary to consider the contents of treatment and the presence or absence of complications. Therefore, the types of treatment plans determined by these factors are numerous.

  It is impossible for a physician to have all the experience in radiotherapy, or to perform a thorough investigation, and there is a need for a device that assists doctors in diagnosing and determining treatment plans due to their lack. In general, such devices assist physicians in making diagnosis and treatment plan decisions by presenting potential diagnosis or treatment plan candidates to the target patient.

  However, the above-described apparatus presents a diagnosis candidate or a treatment plan candidate each time a doctor determines a diagnosis or a treatment plan. The rarer the diagnosis, the less frequently the candidate is presented by the above apparatus, and the presented diagnosis candidate and treatment plan candidate are generally similar each time.

JP 2003-325458 A JP 2006-271541 A

  An object of the present invention is to provide a medical information processing apparatus that provides useful support information for diagnosis and treatment plan determination by a diagnostician.

Medical image information apparatus according to the present embodiment, the physician history information storage unit that stores the patient information storage unit that stores medical information relating to a plurality of patients, historical information about even one with fewer diagnostic or therapeutic for a plurality of doctors A candidate determination unit that determines a candidate for a disease name or a treatment plan based on the medical information about the plurality of patients, and a disease name or a treatment plan that is frequently responded by the plurality of doctors based on the history information. A response status detection unit, a candidate for determination determined by the candidate determination unit, a candidate obtained by removing the disease name or treatment plan determined by the response status detection unit as a display candidate, and the output of the display candidate an output unit for, from the medical information for a plurality of patients the storage, and the patient information acquisition unit for acquiring medical information for similar patients patients diagnosed, before A doctor history information acquisition unit that acquires history information about a doctor of a patient to be diagnosed from among the stored history information about a plurality of doctors, and the candidate determination unit includes the acquired medical information about the similar patient Then, the candidate of a disease name or a treatment plan is determined, and the response status detection unit obtains a disease name or a treatment plan that is frequently dealt with by the plurality of doctors from the acquired history information regarding the doctor .

The figure which shows the structure of the medical information processing apparatus which concerns on this embodiment The figure which shows the table which linked | related patient ID, doctor ID, a symptom, and the diagnostic result by a doctor which the doctor information storage part of FIG. 1 memorize | stores. The figure which shows the table which linked | related the treatment plan which patient ID, doctor ID, a disease name, and the doctor decided memorize | store in the doctor information storage part of FIG. The figure which shows the table which linked | related doctor ID, a doctor name, and a doctor's affiliation department which the doctor information storage part of FIG. 1 memorize | stores. The figure which shows the table which linked | related the medical knowledge, the calculation model, and the classification number which the medical knowledge memory | storage part of FIG. 1 memorize | stores. The figure for demonstrating an example which analyzes several patient information by the data analysis part of FIG. 1, and classify | categorizes into the medical knowledge containing several collections. FIG. 1 is a diagram for explaining an example in which the data analysis unit in FIG. 1 classifies a new patient into any existing set. The figure which shows the typical flow of diagnostic candidate and treatment plan candidate extraction based on this embodiment The figure for demonstrating the example of presentation of the diagnostic candidate to the doctor X based on this embodiment The figure showing the number of treatment plans determined in the past for similar patients in the set to which the target patient belongs, analyzed by the data analysis unit of FIG. The figure showing the number of diagnosis results determined in the past for similar patients in the set to which the target patient belongs, analyzed by the data analysis unit of FIG. The figure which shows the treatment plan and the number of the cases which the diagnostician determined in the past which the doctor number calculation part of FIG. 1 calculated The figure showing the result and the number of cases diagnosed by the diagnostician in the past, calculated by the number-of-doctors calculation unit in FIG. The figure which shows an example of the presentation candidate of the treatment plan extracted by the assistance information determination part of FIG. The figure which shows an example in which the display part of FIG. 1 displays the presentation candidate of a treatment plan The figure which shows another example in which the display part of FIG. 1 displays the presentation candidate of a treatment plan The figure which shows an example of the patient information acquired from the outside which concerns on this embodiment The figure which shows an example of the component which does not necessarily exist in the apparatus among the medical information processing apparatuses which concern on this embodiment The figure which shows the structure of the medical information processing apparatus which concerns on an application example The figure which shows the table which linked | related the symptom or disease name, threshold value, and the number of results which the performance threshold value memory | storage part of an application example memorize | stores

  The medical information processing apparatus according to this embodiment will be described below with reference to the drawings. In the following description, components having substantially the same function and configuration are denoted by the same reference numerals, and redundant description will be given only when necessary.

  FIG. 1 is a diagram illustrating a configuration of a medical information processing apparatus according to the present embodiment. As shown in FIG. 1, the medical information processing apparatus 10 according to the present embodiment includes a patient information storage unit 20, a doctor history information storage unit 30, and a medical knowledge storage unit 40.

  The patient information storage unit 20 stores medical information related to a plurality of patients (hereinafter, medical information related to patients is referred to as patient information). Medical information about a patient includes an image data set and a non-image data set. An image data set is image data generally output from a medical image diagnostic apparatus or the like. Data output from the medical image diagnostic apparatus is normally stored on the medical image storage system PACS, and can be managed and communicated according to the DICOM standard or the like. The non-image data set includes, for example, age, sex, race, medical history, and the like. The non-image data set is data other than image data managed on the medical information system. The medical information system is an information system used in the hospital, and representative examples thereof include an electronic medical record system, a receipt computer processing system, an ordering system, a reception system, a medical assistance system, and a department system.

  Hereinafter, in order to simplify the description, a series of processes in the present embodiment will be described using a non-image data set as medical information. Of course, a series of processes may be performed using an image data set as patient information.

  The doctor history information storage unit 30 stores history information related to diagnosis and treatment plan determination regarding a plurality of doctors. Specifically, the doctor history information storage unit 30 stores, for example, history information regarding diagnosis and treatment plan determination of a plurality of doctors in a hospital. The doctor history information storage unit 30 stores, for example, tables as shown in the following FIGS. 2A, 2B, and 2C. FIG. 2A is a diagram showing a table in which the doctor history information storage unit 30 in FIG. 1 stores the patient ID, the doctor ID, the symptom, and the diagnosis result by the doctor. Specifically, referring to FIG. 2A, for example, in the first row, it means that the doctor ID 10 who is a diagnostician for the patient ID 100 has diagnosed chest pain as a symptom and angina as a diagnostic result. FIG. 2B is a diagram illustrating a table associated with a patient ID, a doctor ID, a disease name, and a treatment plan determined by a doctor, which is stored in the doctor history information storage unit 30 of FIG. FIG. 2B specifically refers to, for example, the first row, meaning that the patient ID 100 is the diagnostic doctor ID10, and the decision is colon cancer as the disease name and Sanmon irradiation (50 Gy / 25 times / 4 weeks) as the treatment plan. To do. FIG. 2C is a diagram illustrating a table in which the doctor history information storage unit 30 in FIG. 1 stores the doctor ID, the doctor name, and the department to which the doctor belongs. Specifically, referring to FIG. 2C, for example, in the first line, the doctor ID 10 means that the doctor name is Taro and the department to which the department belongs is internal medicine. As will be described later, the doctor history information storage unit 30 may add or update information in the tables of FIGS. 2A, 2B, and 2C. The doctor history information storage unit 30 may store a table relating to doctor information having a format other than those shown in FIGS. 2A, 2B, and 2C.

  The medical knowledge storage unit 40 stores medical knowledge based on the similarity of attribute values of the medical information from a plurality of patient information to a plurality of sets for each of a plurality of calculation models. Furthermore, the medical knowledge storage unit 40 stores medical knowledge generated by the data analysis unit 110 described later based on the similarity of a plurality of patient information stored in the patient information storage unit 20. Specifically, the similarity includes similarity regarding patient medical information, similarity regarding cases, similarity regarding medical provision forms, and the like. FIG. 3 is a diagram illustrating a table in which the medical knowledge storage unit 40 in FIG. 1 stores medical knowledge, a calculation model, and the number of classifications. The medical knowledge A is medical knowledge regarding a similar patient generated by the data analysis unit 110. The calculation model is a model for calculating medical knowledge. In medical knowledge A, the calculation model is indicated by “model X f (x, y) = {1, 2, 3}”. This means that the model X is represented by the function f, and outputs 1, 2, or 3 are obtained by two arguments of medical information items x and y. Outputs 1, 2, or 3 indicate that they are classified into sets 1, 2, or 3, respectively. For example, the medical knowledge of FIG. 3 is medical knowledge mainly related to similar patients having similar patient information items.

  The medical information processing apparatus 10 includes a control unit 100. The control unit 100 includes a data analysis unit 110, a patient information acquisition unit 120, a doctor history information acquisition unit 130, a doctor number calculation unit 140, and a support information determination unit 150.

  The control unit 100 comprehensively controls various components included in the control unit 100. The control unit 100 is realized by a central processing unit (Central_Processing_Unit: hereinafter referred to as CPU). The control unit 100 centrally controls each component as the center of the medical information processing apparatus 10.

  The data analysis unit 110 classifies medical knowledge according to each of a plurality of classification models based on a plurality of patient information stored in the patient information storage unit 20. Specifically, the data analysis unit 110 includes a plurality of sets of patient information stored in the patient information storage unit 20 according to each of a plurality of classification models and a plurality of patient information based on similarity of attribute values. Generate medical knowledge. The plurality of patient information includes, for example, race, age, sex, medical history, and the like. As values of medical information items (hereinafter referred to as medical information item values), for age, for example, age values such as 1 year old and 90 years old are expressed as they are. The medical information item value is expressed by using, for example, a value of 1 when there is a history of osteoporosis and 0 when there is no history. The data analysis unit 110 classifies sets by data analysis. Specifically, for example, data mining is used for medical knowledge classification. Data mining, for example, is a technique for searching for correlations, patterns, and the like between information items lurking in a plurality of pieces of medical information possessed by many patients in the present embodiment. In this embodiment, for example, a pattern between items such as “an older person tends to have osteoporosis in the medical history” is obtained. For data analysis, other clustering methods may be used instead of data mining.

  The patient information acquisition unit 120 acquires similar patient information of a patient to be diagnosed from medical information regarding a plurality of patients stored in the patient information storage unit 20. Specifically, the patient information acquisition unit 120 inputs the patient to be diagnosed into one of the medical knowledge stored in the medical knowledge storage unit 40, so that the patient to be diagnosed from among the plurality of stored patient information. Get patient information about similar patients. In order to make the description more specific, the patient information acquisition unit 120 performs the following description on the assumption that the medical knowledge A (similar patient) has been acquired.

  The doctor history information acquisition unit 130 acquires history information about a diagnostician from among history information about a plurality of doctors stored in the doctor history information storage unit 30. For example, the doctor history information acquisition unit 130 acquires history information related to the diagnostic doctor from the tables of FIGS. 2A and 2B. For example, the doctor history information acquisition unit 130 acquires history information related to the diagnostic doctor based on the doctor ID and the disease name input by the operator via the input unit 300. Note that the doctor history information acquisition unit 130 may acquire history information related to the treatment plan related to the diagnostician based on the doctor ID and the treatment plan input by the operator via the input unit 300. The doctor history information acquisition unit 130 may acquire history information related to the diagnosis result related to the diagnostician based on the doctor ID and the diagnosis result input to the operator via the input unit 300.

  The number-of-doctors calculation unit 140 acquires a history regarding treatment plan determination from the history information of the diagnostician, and calculates the number of treatment plan determinations.

  The support information determination unit 150 determines support information for diagnosis and treatment plan determination by the diagnostician for the patient to be diagnosed. Specifically, the support information determination unit 150 performs diagnosis and diagnosis by the diagnostic doctor on the diagnosis target patient based on the medical knowledge acquired from the patient information acquisition unit 120 and the history information acquired from the doctor history information acquisition unit 130. Determine support information for treatment planning decisions. Specifically, the support information determination unit 150 includes a candidate determination unit 152, a response status detection unit 154, a presentation candidate determination unit 156, and an output unit 158. The candidate determination unit 152 determines a disease name or a treatment plan candidate based on medical information regarding a plurality of patients. Based on the history information, the response status detection unit 154 obtains a disease name or treatment plan that is frequently dealt with by a plurality of doctors. The presentation candidate determination unit 156 determines a candidate obtained by removing the disease name or treatment plan obtained by the response status detection unit 154 from the candidates determined by the candidate determination unit 152 as a presentation candidate. The output unit 158 outputs the presentation candidates to the display unit 200. As shown in FIG. 1, the medical information processing apparatus 10 includes a display unit 200, an input unit 300, a main storage unit 400, and an interface unit 500 in addition to the control unit 100.

  The display unit 200 displays various information on the monitor. As the display unit 200, for example, a CRT display, a liquid crystal display, an organic EL display, or the like can be used as appropriate.

  The input unit 300 receives various commands and information input from an operator by an input device. As an input device, a keyboard, a mouse, various switches, and the like can be used.

  The main storage unit 400 stores a base image to be displayed and a sub image in which the base image and the imaging region are substantially coincident. The main storage unit 400 is a storage device that stores various types of information. This is a large-capacity storage device realized by an HDD (Hard_Disk_Drive) or the like. In recent years, SSD (Solid_State_Drive) may be used due to an increase in capacity.

  The interface unit 500 is connected to a not-shown (Picture_Archiving_and_Communication_System: PACS) or other computer via a network. The interface unit 500 performs data communication conforming to a DICOM (Digital_Imaging_and_Communication_in_Medicine) standard with a connected external device.

  Heretofore, each component of the medical information processing apparatus 10 according to the present embodiment has been described. Next, a typical flow of creating medical knowledge from a plurality of patient information stored in the patient information storage unit 20 and storing it in the medical knowledge storage unit 40 will be described.

  FIG. 4 is a diagram for explaining an example in which a plurality of pieces of patient information are analyzed by the data analysis unit 110 of FIG. 1 and classified into medical knowledge including a plurality of sets. The patient information is one of the patient information as described above, and is stored in the patient information storage unit 20. For example, the data analysis unit 110 analyzes patient information on five pieces of patient information, that is, patient A, patient B, patient C, patient D, and patient E, and classifies them into three sets. Actually, the patient information storage unit 20 stores more patient information, and the data analysis unit 110 classifies them into a plurality of sets. More patient information is, for example, patient information about a patient diagnosed so far in a certain hospital. The data analysis unit 110 uses the medical information item I (race) and the medical information item II (age) included in the five patient information to group the five non-image data sets into a plurality of sets (3 in the example of FIG. 3). Class). In data mining and other clustering methods, a set may be called a segment.

  FIG. 5 is a diagram for explaining an example in which the data analysis unit 110 in FIG. 1 classifies a new patient a into any existing set. FIG. 5 shows that when a medical information item I (race) and medical information item II (age) of a new patient a are input to the model function f of the medical knowledge A, the new patient a is classified into the set 3. Means that results are obtained. When a new patient a is classified into the set 3 in the medical knowledge A, the medical knowledge storage unit 40 updates the medical knowledge A. The number of classifications is the number of classifications obtained by the calculation model, and the medical knowledge A is output 1, 2 or 3, that is, the number of classifications is 3. In FIG. 2 and FIG. 3, data analysis is performed on two medical information items of race and age, but actually there are more medical information items.

  In addition, the apparatus which supports the diagnosis and treatment plan which concerns on a prior art example shows the diagnosis candidate and treatment plan candidate which a doctor knows well. It is meaningless for doctors to present diagnosis candidates and treatment plan candidates that he / she knows well many times in the determination of diagnosis and treatment plan, which may be troublesome information. Therefore, the medical information device according to the present embodiment presents diagnosis candidates and treatment plan candidates that are more difficult for a doctor to conceive than such annoying information. FIG. 6 is a diagram for explaining a presentation example of diagnosis candidates to the doctor X according to the present embodiment. In the case of the patient 1, since the doctor X does not have a diagnosis history of aortic divergence among a plurality of diagnoses, the medical information processing apparatus in the present embodiment presents the aortic divergence as a diagnostic candidate. On the other hand, in the case of the patient 2, since the doctor X has a diagnosis history for all of the plurality of diagnoses, the medical information processing apparatus in the present embodiment does not present candidates.

  Below, the typical flow of the diagnostic candidate and treatment plan candidate extraction which concerns on this embodiment is demonstrated. FIG. 7 is a diagram showing a typical flow of diagnostic candidate and treatment plan candidate extraction according to the present embodiment. Note that this embodiment does not present a possible diagnosis or treatment plan candidate for the target patient as in the conventional example.

  First, the control unit 100 causes the input unit 300 to acquire patient information of the target patient (step S11). In step S <b> 11, the operator inputs patient information of the target patient via the input unit 300. Note that the operator may read out patient information of the target patient from another existing database and input it through the input unit 300.

  When step S11 is performed, the control unit 100 causes the patient information acquisition unit 120 to acquire patient information of a set to which the target patient belongs from any medical knowledge (step S12). In step S <b> 12, any medical knowledge is selected from medical knowledge stored in the medical knowledge storage unit 40 by, for example, a doctor's instruction via the input unit 300. The patient information acquisition unit 120 classifies target patients into any set included in the selected medical knowledge by inputting the patient information of the target patient into the selected medical knowledge. The details of the classification of the target patient into any set are as described above. In order to simplify the following description, it is assumed that the target patient is classified into the collection 3 of medical knowledge A. Here, FIG. 8A is a diagram illustrating the number of treatment plans determined in the past for similar patients in the set to which the target patient belongs, analyzed by the data analysis unit 110 of FIG. The diagram of FIG. 8A shows that patients in medical knowledge set A 3 have previously experienced 130 treatment plans of triple irradiation (50 Gy / 25 times / 4 weeks) for colorectal cancer. Moreover, the treatment plan of four-port irradiation (45 Gy / 28 times / 4 weeks) for colorectal cancer has been determined only 7 times. The data analysis unit 110 calculates the number of cases only when the treatment plan is adopted and the result is improved. FIG. 8B is a diagram illustrating the number of diagnosis results determined in the past for similar patients in the set to which the target patient belongs, analyzed by the data analysis unit 110 in FIG. 1. The diagram of FIG. 8B shows that 150 patients have been diagnosed with angina pectoris for chest pain so far in the collection 3 of medical knowledge A. In addition, the diagnosis result of aortic divergence of chest pain has been experienced only 4 times. The patient information acquisition unit 120 can also acquire medical knowledge for a predetermined period as a target. When medical knowledge for three months from the present is targeted, specifically, for example, items in the number of cases in the diagram of FIG. In the case of targeting medical knowledge for three months retroactively, specifically, for example, in the item of the number of cases in the diagram of FIG.

  When step S12 is performed, the control unit 100 causes the doctor history information acquisition unit 130 to acquire doctor information from the doctor history information storage unit 30 (step S13). In step S <b> 13, the operator inputs doctor information of a diagnostician via the input unit 300. The operator may read the doctor information of the diagnostician from another existing database and input it via the input unit 300. Here, the doctor number calculation unit 140 calculates the treatment plan and the number of cases determined in the past by the diagnostician. The number-of-doctors calculation unit 140 calculates the results diagnosed in the past by the diagnostician and the number of cases. FIG. 9A is a diagram showing a treatment plan determined by a doctor and the number of cases calculated by the doctor number calculation unit 140 of FIG. The diagram in FIG. 9A shows that the diagnostician has experienced 80 treatment plans of triple portal irradiation (50 Gy / 25 times / 4 weeks) for colorectal cancer so far. In addition, the treatment plan of four-port irradiation (45 Gy / 28 times / 4 weeks) for colorectal cancer has been experienced only once, that is, the treatment plan is very rare for the doctor. . In other words, the treatment plan of Sanmon irradiation (50 Gy / 25 times / 4 weeks) for colorectal cancer is a treatment plan that can be easily conceived for the diagnostician. On the other hand, the treatment plan of four-port irradiation (45 Gy / 28 times / 4 weeks) for colorectal cancer is a treatment plan that is very difficult to conceive for diagnosticians. FIG. 9B is a diagram showing a result of diagnosis by a doctor and the number of cases calculated by the doctor number calculation unit 140 of FIG. The diagram in FIG. 9B shows that the diagnostician has made 150 diagnoses of chest pain angina so far. Moreover, it has shown that the diagnosis of chest embolism of chest pain was performed 20 times so far.

  If step S13 is performed, the control part 100 will make the candidate determination part 152 determine the candidate of a disease name or a treatment plan based on the medical information regarding the some patient acquired by step S12 (step S14). In step S14, a disease name or a treatment plan candidate is determined based on medical information related to similar patients in the set to which the target patient belongs.

  When step S14 is performed, the control unit 100 causes the response status detection unit 154 to obtain a disease name or a treatment plan with a large number of times corresponding to a plurality of doctors based on the history information about the doctor acquired in step S13 (step S15). ).

  When step S14 is performed, the control unit 100 causes the presentation candidate determination unit 156 to determine, as candidates for presentation, candidates obtained by removing the disease name or treatment plan obtained by the response status detection unit 154 from the candidates determined by the candidate determination unit 152. (Step S16). In step S <b> 16, the presentation candidate determining unit 156 determines the candidate candidate by comparing the patient information of the set to which the target patient belongs and the doctor information. The presentation candidate determination unit 156 compares the patient information and the doctor information, and outputs a diagnosis or treatment plan likely to be overlooked by the diagnostician as a presentation candidate.

  The operation from step S14 to step S16 will be described using a treatment plan as an example. The presentation candidate determination unit 156 compares FIG. 8A (patient information) with FIG. 9A (doctor information). First, the presentation candidate determination unit 156 extracts a treatment plan that is in the patient information but not in the doctor information as a presentation candidate. Moreover, the presentation candidate determination unit 156 extracts a treatment plan in which the number of pieces of doctor information is less than a predetermined threshold as a presentation candidate. Here, FIG. 10 is a diagram illustrating an example of treatment plan presentation candidates extracted by the support information determination unit 150 of FIG. In the case of FIG. 8A (patient information) and FIG. 9A (doctor information), the support information determination unit 150 performs four-port irradiation (45 Gy / 28 times / 4 weeks) for colon cancer and four-port irradiation (40 Gy) for colon cancer shown in FIG. / 28 times / 5 weeks) is extracted as a presentation candidate. The extraction of the diagnosis result is similarly performed based on FIG. 8B (patient information) and FIG. 9B (doctor information).

  Step S14 may be omitted. In that case, the presentation candidate determination unit 156 extracts all treatment plans in FIG. 8A (patient information) and FIG. 9A (doctor information) as presentation candidates. The extraction of the diagnosis result is similarly performed based on FIG. 8B (patient information) and FIG. 9B (doctor information).

  When step S14 is performed, the control unit 100 causes the output unit 158 to output presentation candidates for diagnosis and treatment plan determination by the diagnostician (step S15). In step S15, the output unit 158 presents the treatment plan presentation candidates extracted in step S14. In addition, when there are a plurality of presentation candidates, priority may be given. The priority is calculated based on the number of treatment plans in patient information and doctor information. For example, the support information determination unit 150 increases the priority of a treatment plan with a large number of cases. The support information determination unit 150 compares the patient information and the doctor information with respect to the diagnosis result similarly to the treatment plan, and extracts the presentation candidates. Similarly to the treatment plan, the output unit 158 presents the treatment plan presentation candidate extracted in step S14 in the same manner as the treatment plan. For example, the output unit 158 may display the presentation candidates on the display unit 200. FIG. 11A is a diagram illustrating an example in which the display unit 200 in FIG. 1 displays treatment plan presentation candidates. FIG. 11A shows a case where there is one presentation candidate. The display unit 200 displays the treatment plan candidates as images and characters in the treatment plan candidate display area A1. The display unit 200 displays the treatment plan in stages in the treatment plan display area A2. The display unit 200 displays a patient name, a doctor name, a calendar, operation buttons, and the like in the other information display area A3. FIG. 11B is a diagram illustrating another example in which the display unit 200 in FIG. 1 displays treatment plan presentation candidates. FIG. 11B shows a case where there are two or more presentation candidates. 11B is different from FIG. 11A in that there is a button for selecting the next candidate in the treatment plan candidate display area A1 displayed on the display unit 200. FIG. When the operator inputs a button for selecting the next candidate by clicking the mouse or the like via the input unit 300, the treatment plan candidate next to the currently displayed treatment plan candidate is displayed. Moreover, the display part 200 displays the treatment plan regarding the following treatment plan candidate with the display change of a treatment plan candidate.

  Note that the medical information processing apparatus according to the present embodiment is a treatment plan determination unit (not shown) that determines a treatment plan by a diagnostician for a diagnosis target patient based on the determined support information for diagnosis and treatment plan determination. ) May be provided. The treatment plan determination unit causes the display unit 200 to display the diagnosis or treatment plan by limiting the diagnosis plan or treatment plan proposal presented in step S15 to one more diagnosis or treatment plan. One diagnosis and treatment plan is the rarest diagnosis and treatment plan for the target patient and the diagnostician. Presenting only the most rare diagnosis and treatment plan is useful for determining the diagnosis and treatment plan for physicians who do not need a diagnosis and treatment plan that is easy to come up with.

  The medical information processing apparatus according to the present embodiment has been described so far for candidate presentation of treatment plans and diagnosis results, but the medical information processing apparatus according to the present embodiment is not limited thereto. The information presented as candidates may be examination results and image interpretation results instead of treatment plans and diagnosis results.

  In the medical information processing apparatus according to the present embodiment, the patient information storage unit 20 has basically been described as storing patient information acquired by the medical information processing apparatus. However, the medical information according to the present embodiment is described. The processing device is not bound by it. The patient information storage unit 20 may store patient information acquired via the interface unit 500. Specifically, for example, the patient information storage unit 20 may store patient information accumulated at affiliated hospitals, regions, consortiums, academic societies, or national levels. FIG. 12 is a diagram illustrating an example of patient information acquired from the outside according to the present embodiment. In the diagram in FIG. 12, the patient attribute ID, the institution ID, the disease name, and the treatment plan are associated with each other. In the case of patient information acquired from the outside, the patient name or patient ID cannot be used because the ID may overlap for each institution. Accordingly, the information storage unit 20 stores patient information for each attribute such as sex, age, height, weight, habit, disease name, treatment status, tumor size, and side effects. The institution ID represents the ID of the institution that provided the information.

  Further, in the medical information processing apparatus according to the present embodiment, the information stored in the doctor history information storage unit 30 has been described with respect to information for each individual doctor, but the medical information processing apparatus according to the present embodiment is not limited thereto. Absent. The information stored in the doctor history information storage unit 30 may be stored in units such as medical departments, hospitals, regions, and countries.

  FIG. 13 is a diagram illustrating units that are not necessarily included in the medical information processing apparatus according to the present embodiment. In FIG. 13, units that are not necessarily in the apparatus are indicated by alternate long and short dash lines. That is, the patient information storage unit 20, the doctor history information storage unit 30, and the medical knowledge storage unit 40 are not necessarily provided in the medical information processing apparatus 10. Specifically, the above components may be in the cloud. For example, since the storage units such as the patient information storage unit 20, the doctor history information storage unit 30, and the medical knowledge storage unit 40 are on the cloud, not only a single hospital but also a plurality of hospitals or international medical institutions And share medical knowledge. As a result, since a lot of medical information is accumulated, the accuracy of data mining and the accuracy of medical knowledge generated thereby are improved.

  As described above, according to the medical information processing apparatus according to the present embodiment, medical knowledge can be stored so as to be searchable by performing data analysis on the stored patient information. A set to which a target patient belongs in an arbitrary medical knowledge can be acquired from a plurality of stored medical knowledge. In addition, it is possible to acquire past diagnosis results of the diagnostician and history information on treatment plan determination. It is possible to present a diagnosis and a treatment plan that are difficult for a diagnostician to come up with, based on the diagnosis result and treatment plan information regarding the set to which the target patient belongs and the past diagnosis result and treatment plan decision information of the diagnostician. Presenting diagnosis and treatment plans that are difficult for a diagnostician to come up with can provide useful support information in determining the diagnosis and treatment plan.

(Application examples)
The medical information processing apparatus in the embodiment includes a doctor history information storage unit 30 that stores doctor information. The medical information processing apparatus according to the applied example feeds back the presented support information and the accompanying information.

  FIG. 14 is a diagram illustrating a configuration of the medical information processing apparatus 10 ′ according to the application example 1. As illustrated in FIG. The medical information processing apparatus 10 ′ according to the application example 1 includes a performance feedback unit 610 and a performance threshold storage unit 620.

  The performance feedback unit 610 stores the support information determined by the support information determination unit 150 in the patient information storage unit 20 in association with the patient ID and the medical information related to the patient. The information to be fed back may be any of the tables shown in FIGS. 2A, 2B, and 2C. Also, the information to be fed back is not limited to the tables shown in FIGS. 2A, 2B, and 2C.

  The performance threshold value storage unit 620 stores a predetermined threshold value used when the support information determination unit 150 determines support information in association with the symptom and the disease name. FIG. 15 is a diagram illustrating a table that associates a symptom or disease name, a threshold value, and the number of results, which is stored in the result threshold value storage unit 620 of Application Example 1. The table of FIG. 15 shows that the number of achievements using threshold 5 is 10 to determine the candidate for presentation of colorectal cancer, the number of achievements using threshold 2 is 3, the number of achievements using threshold 1 is 0, and chest pain This indicates that the number of achievements using the threshold value 2 to determine the presentation candidate is 30.

  As described above, the history information in the doctor information storage unit is increased by feeding back the presentation candidates regarding the determined diagnosis and treatment result. By increasing the history information in the doctor information storage unit, the accuracy of the support information presentation is improved. In addition, by storing the actual number of thresholds used for determining the presentation candidates for each medical condition, it becomes reference information for threshold determination for the diagnosed doctor. The presence of reference information for threshold determination improves the accuracy of support information presentation. By improving the accuracy of support information presentation, it is possible to provide support information useful for determination of diagnosis and treatment plan.

  Although several embodiments of the present invention have been described, these embodiments are presented by way of example and are not intended to limit the scope of the invention. These novel embodiments can be implemented in various other forms, and various omissions, replacements, and changes can be made without departing from the scope of the invention. These embodiments and modifications thereof are included in the scope and gist of the invention, and are included in the invention described in the claims and the equivalents thereof.

  DESCRIPTION OF SYMBOLS 10 ... Medical information processing apparatus, 20 ... Patient information storage part, 30 ... Doctor history information storage part, 40 ... Medical knowledge storage part, 100 ... Control part, 110 ... Data analysis part, 120 ... Patient information acquisition part, 130 ... Doctor History information acquisition unit 140 ... Number of doctors calculation unit 150 ... Support information determination unit 152 ... Candidate determination unit 154 ... Correspondence status detection unit 156 ... Presentation candidate determination unit 158 ... Output unit 200 ... Display unit, 300 ... Input unit, 400 ... main storage unit, 500 ... interface unit, 610 ... results feedback unit, 620 ... results threshold value storage unit

Claims (18)

A patient information storage unit for storing medical information on a plurality of patients;
And doctors history information storage unit that stores history information about even one with fewer diagnostic or therapeutic for a plurality of doctors,
A candidate determining unit that determines a candidate for a disease name or a treatment plan based on medical information regarding the plurality of patients;
Based on the history information, a response status detection unit for obtaining a frequent disease name or treatment plan that the plurality of doctors have responded to,
From the candidates determined by the candidate determination unit, a presentation candidate determination unit that determines a candidate excluding the disease name or treatment plan obtained by the correspondence status detection unit as a presentation candidate;
An output unit for outputting the presentation candidates;
A patient information acquisition unit for acquiring medical information related to a similar patient of a patient to be diagnosed from among the stored medical information related to a plurality of patients;
Among the history information about the plurality of doctors stored, a doctor history information acquisition unit for acquiring history information about a doctor of a patient to be diagnosed, and
The candidate determination unit determines a disease name or a treatment plan candidate from the acquired medical information on the similar patient,
The response status detection unit obtains a disease name or treatment plan with a large number of times that the plurality of doctors corresponded from the acquired history information about the doctors,
Medical information processing device. The presentation candidate determination unit preferentially determines a diagnosis with a small diagnosis history as a presentation candidate from candidates obtained by removing the disease name or treatment plan obtained by the response status detection unit from the candidates determined by the candidate determination unit. The medical information processing apparatus according to claim 1. The presentation candidate determination unit preferentially selects a treatment plan with a small determination history from candidates obtained by removing the disease name or treatment plan obtained by the response status detection unit from the candidates determined by the candidate determination unit. The medical information processing apparatus according to claim 1, which is determined. For each of a plurality of classification models, a classification format storage unit that stores a classification format based on similarity of attribute values of the medical information from a plurality of medical information related to the plurality of patients to a plurality of sets,
The candidate deciding unit inputs the patient to be diagnosed into one of the stored classification forms, so that the medical information on the similar patient of the patient to be diagnosed is selected from the stored medical information on the patient. Get
The medical information processing apparatus according to claim 1, wherein the presentation candidate determination unit determines a presentation candidate for diagnosis and treatment plan determination for the diagnosis target patient based on the acquired medical information. A data analysis unit for classifying medical information regarding the plurality of patients according to each of the plurality of classification models and classifying the medical information regarding the plurality of patients into a classification format including a plurality of sets based on similarity of attribute values; The medical information processing apparatus according to claim 4 , further comprising: The medical information processing apparatus according to claim 5 , wherein the data analysis unit classifies the medical information related to the plurality of patients into a classification format including a plurality of sets by performing data mining. The medical information processing apparatus according to claim 5 , wherein the data analysis unit classifies into a classification style including the plurality of sets based on similarity with respect to patient attributes. The medical information processing apparatus according to claim 5 , wherein the data analysis unit classifies the data into a classification format including the plurality of sets based on similarity regarding cases. The medical information processing apparatus according to claim 5 , wherein the data analysis unit classifies the classification format including the plurality of sets based on similarity with respect to a medical provision form. Obtaining a history of treatment plan determination from the history information of the doctor, further comprising a doctor number calculation unit for calculating the number of each treatment plan determination,
The presentation candidate determination section, the number in the treatment plan included in the acquired not be in the medical information on similar patient exists in the history information about the patient's physician in the diagnosis target treatment regimen, or the history information given The medical information processing apparatus according to claim 1, wherein a treatment plan that is smaller than a threshold value is determined as the presentation candidate. A history of diagnosis results is acquired from the history information of the target doctor, and further includes a doctor number calculation unit that calculates the number of diagnosis results of each,
The presentation candidate determination section does not exist in the history information about the patient's physician in the diagnosis target In the the acquired medical information about similar patients diagnosed, or the history threshold number is given in the diagnostic included in the information The medical information processing apparatus according to claim 1, wherein fewer diagnostics are determined as the presentation candidates. The medical information processing apparatus according to claim 10 , further comprising: a treatment plan determination unit that determines a treatment plan by the doctor for the patient to be diagnosed based on the determined diagnosis and candidate for treatment plan determination. The presentation presented candidates determined by the candidate determination section, the medical information processing apparatus according to claim 1, in association with medical information about the patient ID and the patient further comprising a feedback unit to be stored in the patient information storage unit. Wherein the feedback unit, the presentation by the candidate determining unit a predetermined threshold value used in determining the presentation candidates, medical information processing apparatus according to claim 13, further comprising a threshold storage unit that stores in association with symptoms and disease name. The doctor history information storage unit further stores history information regarding examination results and interpretation results regarding a plurality of doctors ,
The presentation candidate determination unit obtains support information for determining examination results and interpretation results by the doctor for the diagnosis target patient based on the acquired medical information on the similar patient and the acquired history information on the doctor. The medical information processing apparatus according to claim 1, which is determined. An interface unit for inputting medical information related to the plurality of patients from the outside to the patient information storage unit;
The medical information processing apparatus according to claim 1, wherein the patient information storage unit stores medical information regarding the plurality of patients input from the outside for each predetermined attribute. The medical information processing according to claim 4 , wherein the candidate determination unit searches and acquires medical information corresponding to the diagnosis target patient stored within a predetermined period from the plurality of stored classification formats. apparatus. Obtain medical information related to a similar patient of a patient to be diagnosed from medical information related to a plurality of patients, and history related to a doctor of the patient to be diagnosed from historical information related to at least one diagnosis or treatment related to a plurality of doctors Information is obtained, a candidate for a disease name or a candidate for a treatment plan is determined from the acquired medical information on the similar patient , and a disease name or a number of times corresponding to the plurality of doctors from the acquired history information about the doctor seeking treatment planning, the candidate excluding the disease name or treatment plan obtained from said the determined candidate determined as presented candidates, to output the presentation candidates,
A medical information processing method.