JP5186858B2 - Medical information processing system, medical information processing method, and program - Google Patents

Description

  The present invention relates to a medical information processing system, a medical information processing method, and a program. In particular, the present invention relates to a medical information processing system and a medical information processing method for processing medical information, and a program for the medical information processing system.

Radiology interpretation results, request department diagnosis results, CAD diagnosis results, and final confirmed diagnosis are recorded as diagnosis history information, and images of reference cases such as discrepancy diagnosis examples can be easily specified from the terminal. A reference case search method that can be displayed above is known (see, for example, Patent Document 1). There is also known a method of automatically collating contents read by a plurality of readers such as medical images (see, for example, Patent Document 2).
JP-A-5-101122 Japanese Patent Laid-Open No. 2004-199691

  In the search method described in Patent Document 1, information relating to the diagnosis performance of the diagnostician cannot be recorded, and thus the reliability of the diagnostic result cannot be calculated based on the diagnostic performance of the diagnostician. Further, according to Patent Document 2, different contents can be displayed in a pop-up when the diagnosis results by a plurality of doctors are different. However, as in the technique described in Patent Document 1, diagnosis based on the diagnosis results of the diagnostician The reliability of the result cannot be calculated.

  In order to solve the above-described problem, according to the first aspect of the present invention, in the medical information processing system, the definite diagnosis result and the initial diagnosis result coincide with each other corresponding to the diagnoser who made the initial diagnosis result. Diagnostic information storage unit for storing the frequency, diagnostic result acquisition unit for acquiring the initial diagnosis result made by each of a plurality of diagnosticians for the same diagnosis target, and reliability of the initial diagnosis result acquired by the diagnostic result acquisition unit Are calculated based on the frequency stored in the diagnostic information storage unit in association with the diagnoser who has made the initial diagnosis result, and the initial diagnosis result acquired by the diagnosis result acquisition unit And an output unit that outputs in accordance with the reliability calculated by the calculation unit.

  The reliability calculation unit increases the reliability of the initial diagnosis result acquired by the diagnosis result acquisition unit as the frequency of storing the diagnosis information storage unit in association with the diagnostician who made the initial diagnosis result is higher It may be calculated. The diagnostic information storage unit stores the frequency at which the final diagnosis result and the initial diagnosis result coincide with each other in association with the diagnostic result of the diagnostician and the diagnostician, and the reliability calculation unit is acquired by the diagnostic result acquisition unit The reliability of the initial diagnosis result may be calculated higher as the frequency stored in the diagnosis information storage unit in association with the initial diagnosis result and the diagnostician who has made the initial diagnosis result is higher.

  The diagnostic information storage unit stores the frequency at which the definitive diagnosis result is given to the diagnosis target for which the diagnostician has made the initial diagnosis result in association with the initial diagnosis result, the definitive diagnosis result, and the diagnostician, Based on the information stored in the diagnostic information storage unit in association with the diagnostician who made the initial diagnosis result, the medical information processing system uses the diagnosis indicating the validity of the initial diagnosis result as an index indicating the validity of the initial diagnosis result. An index calculation unit that calculates for each person may be further included, and the reliability calculation unit may calculate the reliability of the initial diagnosis result indicating that the index calculated by the index calculation unit is more appropriate. In addition, the diagnostic information storage unit stores the initial diagnosis result, the definitive diagnosis result, and the number of the definitive diagnosis result given to the diagnosis target for which the diagnostician has made the initial diagnosis result in association with the diagnoser. You can do it.

  According to the second aspect of the present invention, there is provided a medical information processing method for storing a frequency at which a definite diagnosis result and an initial diagnosis result coincide with each other in association with a diagnostician who has made an initial diagnosis result. For the same diagnosis object, the diagnosis result acquisition stage for acquiring the initial diagnosis result made by each of a plurality of diagnosticians, and the reliability of the initial diagnosis result acquired in the diagnosis result acquisition stage are expressed as the initial diagnosis result. A reliability calculation stage that is calculated based on the frequency stored in the diagnostic information storage stage in association with the diagnosed person and the initial diagnosis result acquired in the diagnosis result acquisition stage is the reliability calculated in the reliability calculation stage. And an output stage for outputting according to the degree.

  According to a third aspect of the present invention, there is provided a program for a medical information processing system, wherein the medical information processing system is associated with a diagnostician who has made an initial diagnosis result, and the confirmed diagnosis result and the initial diagnosis result coincide with each other. The diagnostic information storage unit for storing the frequency of the diagnosis, the diagnosis result acquisition unit for acquiring the initial diagnosis result obtained by each of a plurality of diagnosticians for the same diagnosis target, the reliability of the initial diagnosis result acquired by the diagnosis result acquisition unit A reliability calculation unit that calculates the diagnosis result based on the frequency stored in the diagnostic information storage unit in association with the diagnostician who has made the initial diagnosis result, and the reliability calculation unit that obtains the initial diagnosis result acquired by the diagnosis result acquisition unit. Is made to function as an output unit that outputs in accordance with the calculated reliability.

  The above summary of the invention does not enumerate all the necessary features of the present invention, and sub-combinations of these feature groups can also be the invention.

  Hereinafter, the present invention will be described through embodiments of the invention. However, the following embodiments do not limit the invention according to the scope of claims, and all combinations of features described in the embodiments are included. It is not necessarily essential for the solution of the invention.

  FIG. 1 shows an example of a usage environment of a medical information processing system 100 according to an embodiment. The medical information processing system 100 includes a medical information processing apparatus 110, a diagnosis support information database 120, a plurality of image reader terminals 150a-d (hereinafter collectively referred to as an image reader terminal 150), and a plurality of medical image imaging apparatuses 170a-c. (Hereinafter collectively referred to as a medical image capturing apparatus 170). The diagnosis support information database 120 includes a medical information database 130, a case database 140, and a diagnosis result database 145.

  The radiogram interpreter terminal 150a and the medical image capturing device 170a are provided in the diagnosis site 102a, and the radiogram interpreter terminal 150a causes the radiogram interpreter 190a to interpret the image to be interpreted captured by the medical image capturing device 170a. The radiogram interpreter terminal 150b and the medical image capturing apparatus 170b are provided in the diagnosis site 102b, and cause the radiographer 190b to interpret the image to be interpreted captured by the medical image capturing apparatus 170b. The radiogram interpreter terminal 150c and the medical image capturing apparatus 170c are provided in the diagnosis site 102c, and cause the radiographer 190c to interpret the image to be interpreted captured by the medical image capturing apparatus 170c. As an example, the diagnostic site 102a, the diagnostic site 102b, and the diagnostic site 102c may be medical / medical institutions such as different hospitals.

  The medical information processing apparatus 110, the diagnostic support information database 120, and the radiogram interpreter terminal 150d are provided in a diagnostic site 102d different from the diagnostic sites 102a-c. The medical information processing apparatus 110 is connected to the medical image capturing apparatus 170 through the communication line 160, collects interpretation target images from the medical image capturing apparatus 170 and stores them in the case database 140. Then, the radiogram interpreter terminal 150 displays the radiogram interpretation target image stored in the case database 140. The diagnostic site 102d may be an organization that provides a remote interpretation service. Note that the radiogram interpreter terminals 150a-c can also acquire the radiogram interpretation target images stored in the case database 140 through the communication line 160 and cause the radiogram interpreters 190a-c to interpret the images.

  The medical information database 130 stores medical information that supports diagnosis in a recording medium that can be electronically recorded, such as a hard disk or a nonvolatile semiconductor memory. Medical information includes commentary data of a disease, points to be noted when interpreting an image including a lesion, points to be described as findings, and representative case images. The case database 140 stores a plurality of case information in a recording medium that can be electronically recorded, such as a hard disk or a nonvolatile semiconductor memory. The case information includes medical records such as case images, findings, confirmed diagnosis results, and medical record data.

  The medical information processing apparatus 110 is connected to the diagnosis support information database 120. The medical information processing apparatus 110 can electronically read medical information stored in the medical information database 130 and case information stored in the case database 140. The medical information processing apparatus 110 is connected to the radiographer terminal 150 via the communication line 160 and the like, and the medical information and case information read from the medical information database 130 and the case database 140 are read via the communication line 160 and the like. 150 electronically.

  The radiogram interpreter terminal 150 acquires the radiogram interpretation target image captured by the medical image imaging device 170 and presents it to the radiogram interpreter 190. At this time, the radiogram interpreter terminal 150 presents related medical information and case information acquired from the medical information processing apparatus 110 together with the image to be interpreted. At this time, the medical information processing apparatus 110 searches the case database 140 for similar case images similar to the image to be interpreted that the image interpreter 190 interprets, and causes the image interpreter terminal 150 to display the similar case image.

  Then, the radiogram interpreter terminal 150 receives input of initial diagnosis results, definitive diagnosis results, and the like by the interpreter 190, and receives information indicating the input initial diagnosis results, definitive diagnosis results, and image to be interpreted as the medical information processing apparatus 110. To the communication line 160 or the like. The medical information processing apparatus 110 stores case data including information received from the radiogram interpreter terminal 150 in the case database 140 as new case information. In addition, the medical information processing apparatus 110 stores the initial diagnosis result and the final diagnosis result received from the radiogram interpreter terminal 150 in the diagnosis result database 145 in association with the misdiagnosis frequency. Then, the medical information processing apparatus 110 identifies the misdiagnosis rate of the disease indicated by the similar case image from the diagnosis result database 145 and emphasizes and displays the similar case image having a low misdiagnosis rate.

  The medical imaging apparatus 170 may be an imaging apparatus that captures an image of a subject using electromagnetic waves including radiation, such as an X-ray CT apparatus, a tomosynthesis apparatus, and an MRI apparatus. Here, the medical information processing system 100 has been described as a system extending over a plurality of diagnosis sites 102, but the medical information processing system 100 may be a system closed within one diagnosis site 102. In this case, the diagnostic site 102 may be a different department within one medical institution. For example, the diagnostic site 102 may be a different department within a medical institution.

  As described above, when the medical information processing system 100 is applied to an actual system, it functions as a remote interpretation agent system that acts as an interpretation agent for a hospital that has a shortage of interpreters, or a second opinion for a diagnosis result in one hospital. It may function as a second opinion providing system that can be requested to other hospitals. Further, when the medical information processing system 100 is applied to an actual system, the medical information stored in the medical information database 130 and the case information stored in the case database 140 can be electronically provided to the radiogram interpreter terminal 150. In other words, it may function as an electronic medical book in which cases can be added dynamically as needed. The diagnosis result database 145 is an example of a diagnostic information storage unit in the present invention, and the medical information database 130 and the case database 140 are examples of a case image storage unit in the present invention.

  FIG. 2 shows an example of a block configuration of the medical information processing apparatus 110. The medical information processing apparatus 110 includes a diagnostic result acquisition unit 200, a diagnostic result input unit 205, an output unit 210, a case image selection unit 220, a diagnostic image acquisition unit 230, a reference case image specification unit 240, a misdiagnosis frequency update unit 245, a matching degree. The calculation unit 250, the other diagnosis result specifying unit 260, the reliability calculation unit 265, the other person diagnosis determination unit 270, the diagnosis person selection unit 275, and the index calculation unit 280 are included. The output unit 210 includes a disease information presentation unit 212. The operation of each component of the medical information processing apparatus 110 will be described below together with data stored in the diagnosis support information database 120.

  The diagnosis result database 145 stores the misdiagnosis frequency indicating the frequency at which the initial diagnosis result that is the diagnosis result in the initial diagnosis and the definitive diagnosis result that is the diagnosis result in the definitive diagnosis are associated with the initial diagnosis result and the definitive diagnosis result. To do. The diagnosis result input unit 205 inputs a diagnosis result at the time of diagnosis. Then, the disease information presentation unit 212 correlates with the initial diagnosis result or the definite diagnosis result that matches the diagnosis result input from the diagnosis result input unit 205, and the definite diagnosis result or initial diagnosis result stored in the diagnosis result database 145. Are presented to the image interpreter 190 according to the misdiagnosis frequency stored in the diagnosis result database 145 in association with the initial diagnosis result and the confirmed diagnosis result.

  Specifically, the diagnosis result database 145 indicates the misdiagnosis frequency indicating the frequency of the difference between the initial diagnosis disease indicating the disease diagnosed in the initial diagnosis and the definite diagnosis disease being the disease diagnosed in the definite diagnosis. And stored in association with the definite diagnosis disease. The diagnosis result input unit 205 inputs a disease indicated by the disease information to be presented to the radiogram interpreter 190 at the time of diagnosis. The disease information presentation unit 212 then sets the disease information indicated by the definite diagnosis disease stored in the diagnosis result database 145 in association with the initial diagnosis disease that matches the disease input by the diagnosis result input unit 205 to the misdiagnosis frequency. In response, it is presented to the image interpreter 190.

  The medical information database 130 and the case database 140 store case images to be presented to the radiogram interpreter 190 at the time of diagnosis. Then, the case image selection unit 220 selects a case image to be referred to by the image interpreter 190 at the time of diagnosis from the case images stored in the medical information database 130 and the case database 140. More specifically, the case image selection unit 220 determines that the diagnosis target image to be diagnosed and the image feature amount in the image are predetermined matches from the case images stored in the medical information database 130 and the case database 140. A matching case image is selected with a higher degree of coincidence.

  The diagnosis result input unit 205 inputs a disease indicated by the case image selected by the case image selection unit 220. As described above, the diagnosis result input unit 205 inputs a disease related to the diagnosis target diagnosed by the image interpreter 190 at the time of diagnosis.

  The reference case image specifying unit 240 specifies a case image referred to when the image interpreter 190 makes an initial diagnosis. The diagnosis result database 145 stores the reference case image specified by the reference case image specifying unit 240, the misdiagnosis frequency in association with the initial diagnosis disease and the definite diagnosis disease. Then, the disease information presentation unit 212 presents the case image selected by the case image selection unit 220 to the image interpreter 190, and the initial diagnosis disease or definite diagnosis disease and the case that match the disease input by the diagnosis result input unit 205 The confirmed diagnosis disease or the initial diagnosis disease stored in the diagnosis result database 145 in association with the case image selected by the image selection unit 220 is presented to the interpreter 190 according to the misdiagnosis frequency.

  The case image selection unit 220 matches the diagnosis target image to be diagnosed and the image feature amount in the image higher than a predetermined matching degree from the case images stored in the medical information database 130 and the case database 140. The case images stored in association with the misdiagnosis frequency in the diagnosis result database 145 lower than a predetermined value may be selected. In addition, the disease information presenting unit 212 has a misdiagnosis frequency when the diagnosis result database 145 stores a plurality of definite diagnosis diseases in association with the initial diagnosis disease that matches the disease input by the diagnosis result input unit 205. Larger disease information may be presented to the image interpreter 190 with more emphasis.

  The diagnosis result acquisition unit 200 acquires a new confirmed diagnosis result. When the diagnosis result acquisition unit 200 acquires a new confirmed diagnosis result, the misdiagnosis frequency update unit 245 calculates the above-described frequency and associates the determined diagnosis result and the initial diagnosis result with the misdiagnosis frequency indicating the calculated frequency. The misdiagnosis frequency stored in the diagnosis result database 145 may be updated.

  The diagnosis result database 145 stores the frequency at which the confirmed diagnosis result matches the initial diagnosis result in association with the interpreter 190 who made the initial diagnosis result. The diagnosis result acquisition unit 200 acquires an initial diagnosis result given by each of the plurality of image interpreters 190 for the same diagnosis target. Then, the reliability calculation unit 265 associates the reliability of the initial diagnosis result acquired by the diagnosis result acquisition unit 200 with the frequency stored in the diagnosis result database 145 in association with the image interpreter 190 who made the initial diagnosis result. Calculate based on Then, the output unit 210 outputs the initial diagnosis result acquired by the diagnosis result acquisition unit 200 according to the reliability calculated by the reliability calculation unit 265.

  The reliability calculation unit 265 associates the reliability of the initial diagnosis result acquired by the diagnosis result acquisition unit 200 with the image interpreter 190 who has made the initial diagnosis result, and the diagnosis result database 145 stores the frequency more frequently. You may calculate higher.

  The diagnosis result database 145 stores the frequency at which the confirmed diagnosis result matches the initial diagnosis result in association with the image interpreter 190 and the diagnosis result by the image interpreter 190. Then, the reliability calculation unit 265 stores the reliability of the initial diagnosis result acquired by the diagnosis result acquisition unit 200 in the diagnosis result database 145 in association with the initial diagnosis result and the interpreter 190 who made the initial diagnosis result. The higher the frequency, the higher the calculation.

  The diagnosis result database 145 stores the initial diagnosis result, the definitive diagnosis result, and the frequency with which the definitive diagnosis result is given to the diagnosis target for which the image interpreter 190 gave the initial diagnosis result in association with the interpreter 190. To do. Then, the index calculation unit 280 displays an initial diagnosis result as an index indicating the validity of the initial diagnosis result based on information stored in the diagnosis result database 145 in association with the interpreter 190 who made the initial diagnosis result. It is calculated for each of the read image interpreters 190. Then, the reliability calculation unit 265 calculates the reliability of the initial diagnosis result indicating that the index calculated by the index calculation unit 280 has a higher validity.

  The diagnosis result database 145 is associated with the initial diagnosis result, the definitive diagnosis result, and the radiogram interpreter 190, and the number of the definitive diagnosis results for the diagnosis target that the radiographer 190 gave the initial diagnosis result. Is stored. Then, the index calculation unit 280 is based on the information stored in the diagnosis result database 145 in association with the interpreter 190 who made the initial diagnosis result, the specificity for the initial diagnosis result, the positive predictive value for the initial diagnosis result, The negative predictive value for the initial diagnosis result, and the probability that the diagnostic result indicated by the definitive diagnosis result will be given when the image interpreter 190 gives the diagnosis result indicated by the initial diagnosis result (hereinafter may be referred to as the correct diagnosis rate). Is calculated for each radiogram interpreter 190 who gave the initial diagnosis result. Then, the reliability calculation unit 265 calculates the reliability of the initial diagnosis result having a higher specificity, positive prediction value, negative prediction value, or probability calculated by the index calculation unit 280.

  Based on the reliability of the initial diagnosis result calculated by the reliability calculation unit 265, the other person diagnosis determination unit 270 determines whether or not the other radiographers 190 should diagnose the diagnosis target. . Then, when the other person diagnosis determination unit 270 determines that the other radiographer 190 should diagnose the diagnosis target, the output unit 210 further outputs that the other radiographer 190 should make a diagnosis.

  When the other person diagnosis determining unit 270 determines that the other image interpreter 190 should diagnose the diagnosis target, the diagnosis person selecting unit 275 includes the initial diagnosis result given by each of the plurality of image interpreters 190, and Other interpretations that can diagnose a diagnosis object with high validity based on information stored in the diagnosis result database 145 in association with an initial diagnosis result or a definite diagnosis result that matches the diagnosis result indicated by the initial diagnosis result The user 190 is selected with priority. The output unit 210 further outputs the other radiographer 190 selected by the diagnostician selection unit 275 when the other person diagnosis determination unit 270 determines that the other radiographer 190 should diagnose the diagnosis target. .

  Specifically, the index calculation unit 280 is based on the information stored in the diagnosis result database 145 when the other person diagnosis determination unit 270 determines that the other radiographer 190 should diagnose the diagnosis target. The diagnostic result indicated by the definitive diagnostic result when the specificity, positive predictive value, negative predictive value, and interpreter 190 give the diagnostic result indicated by the initial diagnostic result for the diagnostic result for which the reliability calculating unit 265 has calculated the reliability. At least one of the probabilities of being calculated is calculated for each image interpreter 190 stored in the diagnosis result database 145. Then, the diagnostician selection unit 275 determines the specificity, the positive predictive value, and the negative value calculated by the index calculation unit 280 when the other person diagnosis determination unit 270 determines that the other radiographer 190 should diagnose the diagnosis target. The interpreter 190 having a higher predicted value or higher probability is selected with higher priority.

  Based on the information stored in the diagnostic result storage unit, the other diagnostic result specifying unit 260 determines the frequency of the final diagnosis obtained when the initial diagnostic result obtained by the diagnostic result obtaining unit 200 is given in the initial diagnosis. Identify other diagnostic results that are higher than a predetermined frequency. Then, when the other person diagnosis determination unit 270 determines that the other radiographer 190 should diagnose the diagnosis target, the index calculation unit 280 determines the other diagnosis based on the information stored in the diagnosis result database 145. Specificity, positive predictive value, negative predictive value for the diagnosis result specified by the result specifying unit 260, and the probability that the diagnosis result indicated by the definitive diagnosis result is given in the definitive diagnosis when the diagnosis result is made in the initial diagnosis At least one of them is calculated for each radiogram interpreter 190 stored in the diagnosis result database 145. Then, when the other person diagnosis determining unit 270 determines that the other radiogram interpreter 190 should diagnose the diagnosis target, the diagnoser selecting unit 275 provides an index for the diagnosis result specified by the other diagnosis result specifying unit 260. The interpreter 190 with higher specificity, positive predicted value, negative predicted value, or higher probability calculated by the calculation unit 280 is selected with higher priority.

  The other diagnosis result specifying unit 260 is made in the definitive diagnosis when the initial diagnosis result obtained by the diagnosis result obtaining unit 200 is made in the initial diagnosis based on the information stored in the diagnosis result storage unit. Another diagnosis result having a frequency higher than a predetermined frequency may be specified for each image interpreter 190 who made the initial diagnosis result. The index calculation unit 280 extracts the confirmed diagnosis result and the frequency stored in the diagnosis result database 145 in association with the image interpreter 190 and the initial diagnosis result made by the image interpreter 190, and the extracted frequency is set to a plurality of frequencies. You may calculate the total value totaled over the radiogram interpreter 190 for every definite diagnosis result. Then, the reliability calculation unit 265 may calculate the reliability of the definitive diagnosis result having a larger total value calculated by the index calculation unit 280 to a greater degree.

  In the present embodiment, it is assumed that the initial diagnosis is a misdiagnosis when the disease diagnosed by the initial diagnosis is different from the disease diagnosed by the definitive diagnosis. For certain diseases, the presence or absence of disease and the type of disease are finally determined by cytodiagnosis or the like. In some cases, the presence or absence of a disease and the type of disease may be finally determined by a consultation between a plurality of doctors. In the present embodiment, such final determination of the presence or absence of a disease and the type of disease is referred to as a definitive diagnosis. On the other hand, a diagnosis before a definitive diagnosis or an initial finding diagnosed by a radiologist or the like with respect to a diagnosis target is referred to as an initial diagnosis in this embodiment.

  When the medical information processing apparatus 110 is applied as an actual system as in the medical information processing system 100 described above, similar case images are presented to the interpreter 190 according to the misdiagnosis frequency while dynamically updating the misdiagnosis frequency. There are cases where it is possible. In addition, based on the result of interpretation by the image interpreter 190, it may be possible to appropriately determine another image interpreter 190 that should interpret the same diagnostic target image.

  FIG. 3 shows an example of data stored in the medical information database 130 in a table format. The medical information database 130 stores medical information data stored in a so-called electronic medical dictionary.

  Specifically, the medical information database 130 stores a disease ID for identifying a disease, an image ID for identifying a case image indicating the disease, and disease information data in association with each other. The case image stored in the medical information database 130 may be a representative case image of a disease. For example, the case image stored in the medical information database 130 may be a case image published in a so-called medical book, or may be a case image having characteristic features of a disease. The disease information data is provided to the interpreter 190 in relation to the disease, such as the characteristics of the disease, points to be noted at the time of interpretation, points to be described as findings, examples of findings, names of other diseases that are easily misdiagnosed, etc. Power information and image information may be included.

  FIG. 4 shows an example of data stored in the case database 140 in a table format. The case database 140 stores an image captured by the medical image capturing apparatus 170 as a case image together with the diagnosis result.

  Specifically, the case database 140 includes a user ID that uniquely identifies the image interpreter 190, an image ID that identifies a case image captured by the medical image capturing apparatus 170, a disease ID that identifies a disease, and diagnosis result data. Stored in association. The user ID may be information for identifying a user who can use a case image stored in the case database 140.

  Thus, since the case database 140 stores the case image in association with the user ID, the case images that should be stored by the image interpreter 190 can be accumulated for each image interpreter 190. Thereby, the case database 140 can function as a case database unique to the radiogram interpreter 190. In addition, the medical information processing apparatus 110 may use the user ID as permission information for limiting the radiogram interpreter 190 who can use the case image. Note that the case database 140 may store information for identifying the diagnostic site 102 instead of the user ID. Thereby, the case database 140 can function as, for example, a hospital-specific case database.

  The image ID may be information for identifying a case image stored in the case database 140. The image ID may be information for identifying a case image stored in the case database 140 and a case image stored in the medical information database 130. The disease ID may be a disease ID that identifies the disease indicated by the definitive diagnosis result for the diagnosis target indicated by the corresponding case image.

  The diagnosis result data includes data recorded as a medical record, for example, information related to a patient acquired in a medical process such as name, sex, age, diagnosis result, medication information, etc. Contains information called medical information. Specifically, the diagnosis result data may be medical chart data. In addition, the case database 140 may store diagnosis result data indicating a diagnosis result given by the image interpreter 190 for each image interpreter 190 identified by the user ID.

  FIG. 5 shows an example of data stored in the diagnosis result database 145 in a table format. The diagnostic result database 145 stores an image ID, a diagnostic result, a diagnostician ID, a reference case image ID, and a confirmed diagnosis. The diagnosis result database 145 stores the diagnosis result, the diagnosing ID, and the reference case image ID for each of a plurality of diagnosing persons (for example, a radiology department and a requested department).

  The image ID may be identification information that identifies an image to be interpreted. The diagnoser ID may be a disease ID that identifies a disease included in a diagnosis result made by the image interpreter 190 by interpreting an image to be interpreted including an image identified by the image ID. The diagnostician ID may be identification information for identifying the radiogram interpreter 190 who made the diagnosis. Further, the reference case image ID may be an image ID for identifying a case image referred to by the image interpreter 190 who made the diagnosis among the case images stored in the medical information database 130 and the case database 140. . The definitive diagnosis may be a disease ID that identifies a disease included in a diagnosis result made as a definitive diagnosis.

  FIG. 6 shows an example of other data stored in the diagnosis result database 145 in a table format. The diagnosis result database 145 stores a diagnostician ID, an initial diagnosis result, a definitive diagnosis result, and the number of cases.

  The diagnostician ID may be identification information for identifying the radiogram interpreter 190 as described with reference to FIG. The initial diagnosis result may be a disease ID that identifies a disease included in a diagnosis result made before a definitive diagnosis result is obtained. The initial diagnosis result includes a diagnosis result diagnosed in the radiology department described with reference to FIG. 5 and a diagnosis result diagnosed in the request department.

  As described with reference to FIG. 5, the definitive diagnosis result may be a disease ID that identifies a disease included in the diagnosis result made as the definitive diagnosis. The number of cases may be the number of combinations of the initial diagnosis result and the definitive diagnosis result by the image interpreter 190 identified by the diagnoser ID. The diagnosis result database 145 calculates and stores the data shown in FIG. 6 by counting the data described in relation to FIG. 5 for each radiogram interpreter 190 when the final diagnosis result is obtained.

  Thus, the diagnosis result database 145 stores the number of cases for each combination of the initial diagnosis result and the definite diagnosis result made by the image interpreter 190 for each image interpreter 190. For this reason, as will be described later, the medical information processing apparatus 110 can calculate not only the misdiagnosis frequency but also the misdiagnosis frequency that each of the image interpreters 190 has made.

  FIG. 7 shows an example of a screen displayed on the radiogram interpreter terminal 150 by the medical information processing apparatus 110. The medical information processing apparatus 110 interprets a screen including an interpretation target image window 710, a similar case image window 720 and a similar case image window 951-954, a diagnostic report input window 730, a reference diagnostic report display window 740, and a misdiagnosis information button 760. Displayed on the operator terminal 150.

  The medical information processing apparatus 110 displays the image to be interpreted acquired by the image interpreter terminal 150 from the medical information processing apparatus 110 in the image interpretation target image window 710. Then, the medical information processing apparatus 110 displays similar case images in the similar case image windows 951 to 953 among the case images stored in the case database 140. In addition, the medical information processing apparatus 110 causes the similar case image window 954 to display the case images stored in the medical information database 130 in association with the disease indicated by the designated case image. When the image interpreter 190 specifies a case image by clicking any one of the similar case image windows 951 to 954 with a mouse or the like, the image interpreter terminal 150 displays the specified case image as a similar case. The image is enlarged and displayed in the image window 750.

  Further, the medical information processing apparatus 110 acquires diagnosis result data such as findings of the designated case image or disease information data from the case database 140 or the medical information database 130 and displays the contents in the reference diagnosis report display window 740. Let At this time, the medical information processing apparatus 110 displays the disease name included in the diagnosis result data on the reference diagnosis report display window 940 as a link 742 to the disease information data stored in the medical information database 130. The link 742 may be a hyperlink in hypertext described by HTML, for example. When the image interpreter 190 clicks the link 742 displayed on the image interpreter terminal 150 by a mouse operation or the like, the medical information processing apparatus 110 acquires the linked disease information data from the medical information database 130, and the image interpreter terminal 150. To be displayed on the radiogram interpreter terminal 150. The disease information data may be described in a markup language such as HTML or XML.

  When the misdiagnosis information button 760 is clicked by a mouse operation or the like, the medical information processing apparatus 110 interprets the case image displayed in the similar case image window 750 and the misdiagnosis information related to the diagnosis result on the case image. It is displayed on the terminal 150. The misdiagnosis information will be described with reference to FIG. Then, the image interpreter 190 refers to the image to be interpreted displayed in the image to be interpreted image interpretation window 710, the similar case image displayed in the similar case image window 750, the diagnosis result of the similar case image, and the misdiagnosis information. The diagnosis result including the disease is input to the diagnosis report input window 730.

  The radiogram interpreter terminal 150 identifies the disease ID that identifies the disease included in the diagnosis result input to the diagnostic report input window 730 by the radiogram interpreter 190, and the case image that the radiogram interpreter 190 displays in the similar case image window 750. The information is transmitted to the medical information processing apparatus 110 together with the case image ID and the diagnostician ID for identifying the image interpreter 190. Then, the medical information processing apparatus 110 stores the disease ID, the diagnostician ID, and the case image ID received from the radiogram interpreter terminal 150 in the diagnostic result database 145.

  Specifically, as described with reference to FIG. 5, the medical information processing apparatus 110 stores the disease ID received from the radiogram interpreter terminal 150 in the diagnosis result database 145 as a diagnosis result. Further, the radiogram interpreter terminal 150 stores the case image ID received from the radiogram interpreter terminal 150 in the diagnosis result database 145 as a reference case image ID. The medical information processing apparatus 110 stores the diagnostician ID received from the radiogram interpreter terminal 150 in the diagnostic result database 145 as the diagnostician ID.

  FIG. 8 shows an example of misdiagnosis information displayed on the radiogram interpreter terminal 150 by the medical information processing apparatus 110. The medical information processing apparatus 110 displays misdiagnosis information 810, 820, and 830 indicating misdiagnosis contents on the radiogram interpreter terminal 150.

  The misdiagnosis information 810 includes the probability that the initial diagnosis is a misdiagnosis when the disease diagnosed for the selected similar case image is made in the initial diagnosis, and the confirmation for the diagnosis target in which the disease is made in the initial diagnosis. And the proportion of diseases other than the disease made in the diagnosis. Further, the misdiagnosis information 820 includes a misdiagnosis probability when referring to the selected similar case image and a ratio of diseases other than the disease made in the definitive diagnosis. Further, the misdiagnosis probability information 830 includes a ratio of a disease that is diagnosed with respect to the selected similar case image that is diagnosed in a definitive diagnosis and a disease that is different from the disease in the initial diagnosis, and an initial diagnosis. And the proportion of diseases other than the disease given in 1.

  As described above, according to the medical information processing system 100, various information can be provided to the image interpreter 190 from the diagnosis result regarding the selected similar case image. The medical information processing apparatus 110 can generate data for displaying the misdiagnosis content based on the information described with reference to FIG.

  FIG. 9 shows an example of an operation sequence of the radiogram interpreter terminal 150 and the medical information processing apparatus 110. The image interpreter terminal 150 requests an image to be interpreted by the image interpreter 190 to be interpreted (S902). The medical information processing apparatus 110 transmits the requested interpretation target image to the radiogram interpreter terminal 150 (S904). Then, the radiogram interpreter terminal 150 displays the radiogram interpretation target image acquired from the medical information processing apparatus 110 (S906).

  On the other hand, the medical information processing apparatus 110 searches for a similar case image similar to the image to be interpreted transmitted to the image interpreter terminal 150 (S908). The processing content of the medical information processing apparatus 110 in S908 will be described with reference to FIG. Then, the medical information processing apparatus 110 calculates misdiagnosis information regarding each searched similar case image (S910). The misdiagnosis information may be data for causing the radiogram interpreter terminal 150 to display the misdiagnosis content described in relation to FIG. Then, the medical information processing apparatus 110 transmits the similar case image searched in S908 and the misdiagnosis information calculated in S910 to the radiogram interpreter terminal 150 (S912).

  As described with reference to FIG. 7, the radiogram interpreter terminal 150 displays similar case images as thumbnails in comparison with the image to be interpreted (S914). When the misdiagnosis information button 760 is clicked with the similar case image selected (S916), the radiogram interpreter terminal 150 displays the misdiagnosis information described with reference to FIG. 8 (S918).

  As described with reference to FIGS. 7 and 8, when a disease name link is clicked, information on the link destination is transmitted to the medical information processing apparatus 110 (S922). Based on the link destination information received from the radiogram interpreter terminal 150, the medical information processing apparatus 110 reads the link destination disease information from the medical information database 130 and transmits it to the radiogram interpreter terminal 150 (S924). The radiogram interpreter terminal 150 displays the link destination disease information received from the medical information processing apparatus 110 (S926). The disease information displayed on the radiogram interpreter terminal 150 by the medical information processing apparatus 110 will be described with reference to FIG.

  When the radiogram interpreter terminal 150 detects that a diagnostic result is input to the diagnostic report input window 730 (S928), the diagnostic interpreter terminal 150 transmits a diagnostic result including a disease to the medical information processing apparatus 110 (S930). The medical information processing apparatus 110 calculates the disease included in the input diagnosis result in the same manner as S910 (S932). Note that the process in S910 and the process in S932 differ depending on whether the disease is a disease diagnosed on a similar case image or a disease indicated by a diagnosis result made by the image interpreter 190. The medical information processing apparatus 110 transmits the misdiagnosis information calculated in S932 to the radiogram interpreter terminal 150 (S934). The radiogram interpreter terminal 150 displays misdiagnosis information as in S918 (S936).

  FIG. 10 shows an example of the processing flow of the medical information processing apparatus 110 in S908. The coincidence calculation unit 250 calculates the image feature amount of the image to be interpreted (S1002). The image feature amount may include the circularity, density, and size of the abnormal shadow. The abnormal shadow may be an image of an affected part or an image of a part suspected of being an affected part in an image obtained by imaging a subject.

  The degree of coincidence calculation unit 250 calculates the degree of coincidence of image feature amounts between the abnormal shadow included in the image to be interpreted and the abnormal shadow included in the case images stored in the medical information database 130 and the case database 140 (S1004). ). Then, the case image selection unit 220 selects a case image including an abnormal shadow that matches the abnormal shadow included in the image to be interpreted with a higher matching degree than a predetermined matching degree (S1006).

  At this time, the case image selection unit 220 calculates the misdiagnosis frequency when the selected case image is referred to (S1008). The misdiagnosis frequency calculated here may be the misdiagnosis probability described in relation to the misdiagnosis information 820 in FIG. Then, the case image selection unit 220 selects a case image whose misdiagnosis frequency calculated in S1008 is smaller than a predetermined frequency as a similar case image to be transmitted to the radiogram interpreter terminal 150 (S1010).

  FIG. 11 shows an example of a processing flow after the medical information processing apparatus 110 acquires an initial diagnosis result. The diagnosis result acquisition unit 200 acquires a disease diagnosed as an initial diagnosis in the radiogram interpreter terminal 150 (S1102). The diagnosis result acquisition unit 200 may acquire a disease directly from the radiogram interpreter terminal 150, or may acquire a disease stored in the diagnosis result database 145 or the like.

  Then, the index calculation unit 280 calculates the specificity, positive predictive value, negative predictive value, and correct diagnosis rate for the disease acquired in S1102 (S1104). The index calculation unit 280 determines the specificity, positive predictive value, and negative prediction for each reader 190 who made the initial diagnosis based on the data stored in the diagnosis result database 145 described with reference to FIG. Values and correct diagnosis rates can be calculated.

  The reliability calculation unit 265 calculates the reliability based on the specificity, the positive prediction value, the negative prediction value, and the correct diagnosis rate calculated in S1104 (S1106). Specifically, the reliability calculation unit 265 calculates the reliability according to the specificity, the positive prediction value, the negative prediction value, and the magnitude of the correct diagnosis rate. For example, the reliability calculation unit 265 uses the specificities, positive prediction values, negative prediction values, and weighting factors determined in advance for the correct diagnosis rate, the specificity calculated by the index calculation unit 280 in S1104, positive A total value of the predicted value, the negative predicted value, and the value obtained by multiplying the correct diagnosis rate may be calculated as the reliability.

  The other person diagnosis determination unit 270 determines whether or not to request diagnosis from another person (S1108). The other person diagnosis determination unit 270 determines that the other person is requested to make a diagnosis when the reliability calculated in step S1106 is smaller than a predetermined value.

  If it is determined that the diagnosis is requested from another person, the diagnostician selection unit 275 specifies a disease having a probability that a diagnosis result different from the initial diagnosis is obtained in the definitive diagnosis is higher than a predetermined value (S1110). . Specifically, the other person diagnosis determination unit 270 correlates with the interpreter 190 who made the initial diagnosis, the initial diagnosis result, and the definite diagnosis result different from the initial diagnosis result, and the number of cases in which the diagnosis result database 145 indicates a higher frequency. The disease indicated by the definitive diagnosis result stored in is stored.

  Then, the diagnostician selection unit 275 calculates the specificity, positive predictive value, negative predictive value, and correct diagnosis rate for the disease identified in S1110 for each of the interpreters 190 other than the interpreter 190 who made the initial diagnosis. (S1112). The diagnostician selection unit 275 determines the specificity, positive predictive value, negative predictive value, and positive value for each radiographer 190 based on the data stored in the diagnostic result database 145 described with reference to FIG. The diagnosis rate can be calculated.

  The diagnostician selection unit 275 selects another image interpreter 190 to be requested for diagnosis based on at least one of the specificity, positive predictive value, negative predictive value, and correct diagnosis rate calculated in S1112 (S1114). . At this time, the diagnostician selection unit 275 may calculate the reliability according to the magnitude of the specificity, the positive predicted value, the negative predicted value, and the correct diagnosis rate for each radiographer 190 in the same manner as in S1106. For example, the diagnostician selection unit 275 uses the specificities, positive prediction values, negative prediction values, and weighting factors determined in advance for each of the correct diagnosis rate, the specificity calculated by the index calculation unit 280 in S1104, the positive A total value of the predicted value, the negative predicted value, and the value obtained by multiplying the correct diagnosis rate may be calculated for each radiogram interpreter 190. Then, the diagnostician selection unit 275 may select the radiogram interpreter 190 having the higher total value as another radiogram interpreter 190 that should request diagnosis.

  Then, the output unit 210 transmits information specifying the diagnostician selected by the diagnostician selection unit 275 to the radiogram interpreter terminal 150 (S1116). In S1114, it is assumed that the diagnostician selection unit 275 calculates the weighting coefficient for the negative predictive value larger than the other index as an example. In this case, if another image interpreter 190 selected by the diagnostician selection unit 275 as a result diagnoses that the disease specified in S1110 is negative, the probability of being negative for the disease can be increased. Thus, according to the medical information processing system 100, it is possible to appropriately select a set of image interpreters 190 that can further improve the reliability as a whole.

  FIG. 12 shows an example of display contents displayed on the radiogram interpreter terminal 150 by the medical information processing apparatus 110. The medical information processing apparatus 110 displays a window 1200 including the reliability area 1210 and the recommended requester area 1220 on the radiogram interpreter terminal 150. The medical information processing apparatus 110 causes the reliability area 1210 to display the overall reliability indicating the overall reliability associated with each of a plurality of disease names. The total reliability referred to here may be the reliability calculated by the reliability calculation unit 265 in S1106. The medical information processing apparatus 110 uses the overall reliability calculated by the reliability calculation unit 265 for the other diagnosis results specified by the other diagnosis result specifying unit 260 in addition to the disease acquired by the diagnosis result acquisition unit 200. May be displayed in the reliability area 1210.

  In addition, the medical information processing apparatus 110 may display a plurality of other image interpreters 190 selected by the diagnostician selection unit 275 in S1114 in the recommended requester area 1220 as recommended requesters. The medical information processing apparatus 110 may display request buttons 1221, 1223, and 1224 in the recommended requester area 1220 in association with each recommended requester. When the request buttons 1221, 1223, and 1224 are clicked by a mouse operation or the like, the radiogram interpreter terminal 150 transmits to the medical information processing apparatus 110 a request for diagnosis from the associated recommendation requester. Then, the medical information processing apparatus 110 requests diagnosis from the recommendation requester received from the radiogram interpreter terminal 150.

  FIG. 13 shows an example of disease information displayed on the radiogram interpreter terminal 150 by the medical information processing apparatus 110. When the medical information processing apparatus 110 receives the link destination information received from the radiogram interpreter terminal 150 in step S922, the medical information processing apparatus 110 reads the data stored in the medical information database 130 as the link destination indicated by the link destination information. A disease information window 1300 including an area 1310, a case image area 1320, a differential case image area 1330, and a detailed explanation area 1340 is displayed on the radiogram interpreter terminal 150.

  The medical information database 130 stores important point data including important points to be noted, and the medical information processing apparatus 110 reads out the important point data stored in the medical information database 130 and displays the important point data in the overview area 1310. Power display data is generated and transmitted to the radiogram interpreter terminal 150. The medical information database 130 stores case images 1321-1324, and the medical information processing apparatus 110 reads the case images 1321-1324 stored in the medical information database 130 and displays them in the case image region 1320. Power display data is generated and transmitted to the radiogram interpreter terminal 150.

  The medical information database 130 stores case images 1331-1335 of other diseases that present similar shadows, and the medical information processing apparatus 110 stores the case images 1331-1335 stored in the medical information database 130. The display data to be read and displayed in the differentiated case image area 1330 is generated and transmitted to the radiogram interpreter terminal 150. Further, the medical information database 130 stores detailed explanation data including detailed explanation data including a concept of a disease, a pathological finding, a finding point in the case of finding from an image, and the medical information processing apparatus 110 includes a medical information database. The detailed explanation data stored in 130 is read out, display data to be displayed in the detailed explanation area 1340 is generated, and transmitted to the radiogram interpreter terminal 150.

  The content displayed in the disease information window 1300 can be electronically copied from the area selected by the mouse operation or the like of the image interpreter 190, and the content copied to the diagnosis report input window 730 is the mouse. It can be input by operation or the like. For example, when a display content such as “○ □ is visible” displayed under the image 1321-1324 is clicked or selected by a mouse operation, the content is copied to the memory of the reader terminal 150, and the copied content is displayed. Can be easily input to the diagnostic report input window 730 by operating the mouse or the like.

  As described above, according to the medical information processing system 100, since misdiagnosis information can be appropriately provided at the time of diagnosis by the image interpreter 190, it can be expected that the reliability of diagnosis is increased. Further, according to the medical information processing system 100, the information stored in the medical information database 130 and the diagnosis result stored in the case database 140 can be used, so that it is expected that the probability of an input error occurring is reduced. it can. Further, even after the image interpreter 190 makes a diagnosis, it is possible to recommend another appropriate image interpreter 190 according to the image interpretation result by the image interpreter 190, so that it is expected to improve the reliability of the diagnosis result as a whole. it can.

  FIG. 14 shows an example of a hardware configuration when the medical information processing apparatus 110 and the diagnosis support information database 120 are configured by an electronic information processing apparatus such as a computer. The medical information processing apparatus 110 and the diagnosis support information database 120 include a CPU peripheral unit, an input / output unit, and a legacy input / output unit. The CPU peripheral section includes a CPU 1505, a RAM 1520, a graphic controller 1575, and a display device 1580 that are connected to each other by a host controller 1582. The input / output unit includes a communication interface 1530, a hard disk drive 1540, and a CD-ROM drive 1560 that are connected to the host controller 1582 by the input / output controller 1584. The legacy input / output unit includes a ROM 1510, a flexible disk drive 1550, and an input / output chip 1570 connected to the input / output controller 1584.

  The host controller 1582 connects the RAM 1520, the CPU 1505 that accesses the RAM 1520 at a high transfer rate, and the graphic controller 1575. The CPU 1505 operates based on programs stored in the ROM 1510 and the RAM 1520 to control each unit. The graphic controller 1575 acquires image data generated by the CPU 1505 or the like on a frame buffer provided in the RAM 1520 and displays the image data on the display device 1580. Alternatively, the graphic controller 1575 may include a frame buffer that stores image data generated by the CPU 1505 or the like.

  The input / output controller 1584 connects the host controller 1582 to the hard disk drive 1540, the communication interface 1530, and the CD-ROM drive 1560, which are relatively high-speed input / output devices. The hard disk drive 1540 stores programs and data used by the CPU 1505. The communication interface 1530 is connected to the network communication device 1598 to transmit / receive programs or data. The CD-ROM drive 1560 reads a program or data from the CD-ROM 1595 and provides it to the hard disk drive 1540 and the communication interface 1530 via the RAM 1520.

  The input / output controller 1584 is connected to the ROM 1510, the flexible disk drive 1550, and the relatively low-speed input / output device of the input / output chip 1570. The ROM 1510 stores a boot program executed when the medical information processing apparatus 110 and the diagnosis support information database 120 are started up, or a program depending on the hardware of the medical information processing apparatus 110 and the diagnosis support information database 120. The flexible disk drive 1550 reads a program or data from the flexible disk 1590 and provides it to the hard disk drive 1540 and the communication interface 1530 via the RAM 1520. The input / output chip 1570 connects various input / output devices via the flexible disk drive 1550 or a parallel port, serial port, keyboard port, mouse port, and the like.

  A program executed by the CPU 1505 is stored in a recording medium such as the flexible disk 1590, the CD-ROM 1595, or an IC card and provided by the user. The program stored in the recording medium may be compressed or uncompressed. The program is installed in the hard disk drive 1540 from the recording medium, read into the RAM 1520, and executed by the CPU 1505. The program executed by the CPU 1505 includes the medical information processing apparatus 110, the diagnosis result acquisition unit 200, the diagnosis result input unit 205, the output unit 210, the case image selection unit 220, and the diagnostic image described with reference to FIGS. An acquisition unit 230, a reference case image identification unit 240, a misdiagnosis frequency update unit 245, a coincidence calculation unit 250, another diagnosis result identification unit 260, a reliability calculation unit 265, an other person diagnosis determination unit 270, a diagnosis person selection unit 275, and It functions as the index calculation unit 280 and causes the diagnosis support information database 120 to function as the diagnosis support information database 120 described with reference to FIGS.

  The program shown above may be stored in an external storage medium. As the storage medium, in addition to the flexible disk 1590 and the CD-ROM 1595, an optical recording medium such as a DVD or PD, a magneto-optical recording medium such as an MD, a tape medium, a semiconductor memory such as an IC card, or the like can be used. In addition, a storage device such as a hard disk or a RAM provided in a server system connected to a dedicated communication network or the Internet is used as a recording medium, and is provided to the medical information processing apparatus 110 and the diagnosis support information database 120 as a program via the network. May be.

  As mentioned above, although this invention was demonstrated using embodiment, the technical scope of this invention is not limited to the range as described in the said embodiment. It will be apparent to those skilled in the art that various modifications or improvements can be added to the above-described embodiment. It is apparent from the scope of the claims that the embodiments added with such changes or improvements can be included in the technical scope of the present invention.

It is a figure which shows an example of the utilization environment of the medical information processing system 100 which concerns on one Embodiment. 2 is a diagram illustrating an example of a block configuration of a medical information processing apparatus 110. FIG. It is a figure which shows an example of the data which the medical information database 130 stores in a table format. It is a figure which shows an example of the data which the case database 140 stores in a table format. It is a figure which shows an example of the data which the diagnostic result database 145 has stored in the table format. It is a figure which shows an example of the other data which the diagnostic result database 145 stores in a table format. 6 is a diagram illustrating an example of a screen displayed on the radiogram interpreter terminal 150 by the medical information processing apparatus 110. FIG. 6 is a diagram illustrating an example of misdiagnosis information displayed on the radiogram interpreter terminal 150 by the medical information processing apparatus 110. FIG. It is a figure which shows an example of the operation | movement sequence of the radiogram interpreter terminal 150 and the medical information processing apparatus 110. FIG. It is a figure which shows an example of the processing flow of the medical information processing apparatus 110 in S908. It is a figure which shows an example of the processing flow after the medical information processing apparatus 110 acquires the initial diagnosis result. 6 is a diagram illustrating an example of display contents displayed on the radiogram interpreter terminal 150 by the medical information processing apparatus 110. FIG. It is a figure which shows an example of the disease information which the medical information processing apparatus 110 displays on the radiogram interpreter terminal 150. FIG. It is a figure which shows an example of a hardware configuration at the time of comprising the medical information processing apparatus 110 and the diagnostic assistance information database 120 by electronic information processing apparatuses, such as a computer.

Explanation of symbols

DESCRIPTION OF SYMBOLS 100 Medical information processing system 102 Diagnosis site 110 Medical information processing apparatus 120 Diagnosis support information database 130 Medical information database 140 Case database 145 Diagnosis result database 150 Interpreter terminal 160 Communication line 170 Medical imaging device 190 Interpreter 200 Diagnosis result acquisition part 205 Diagnosis result input unit 210 Output unit 212 Disease information presentation unit 220 Case image selection unit 230 Diagnostic image acquisition unit 240 Reference case image identification unit 245 Misdiagnosis frequency update unit 250 Concordance calculation unit 260 Other diagnosis result identification unit 265 Reliability calculation unit 270 Diagnosis determination unit 275 Diagnosis selection unit 280 Index calculation unit

Claims (12)

A diagnostic information storage unit that stores the frequency at which the definitive diagnosis result matches the initial diagnosis result in association with the diagnostician who made the initial diagnosis result,
For the same being diagnosed, and the diagnostic result acquisition unit that acquires an initial diagnosis of each of the plurality of diagnostician beat,
A reliability calculation unit that calculates the reliability of the initial diagnosis result acquired by the diagnosis result acquisition unit based on the frequency stored in the diagnosis information storage unit in association with the diagnostician who made the initial diagnosis result; ,
An output unit that outputs the initial diagnosis result acquired by the diagnosis result acquisition unit according to the reliability calculated by the reliability calculation unit;
Other diagnosis determination unit for determining whether or not to diagnose other diagnosticians of the plurality of diagnosticians based on the reliability of the initial diagnosis result calculated by the reliability calculation unit;
When the other person diagnosis determination unit determines that another diagnosis person should diagnose the diagnosis object, the initial diagnosis result given by each of the plurality of diagnosis persons and the diagnosis result indicated by the initial diagnosis result Based on the information stored in the diagnostic information storage unit in association with the initial diagnosis result or the definitive diagnosis result that coincides with the priority, the other diagnosticist who can diagnose the diagnosis object with high validity is given priority. A diagnostician selection unit for selecting
The output unit, when the other person diagnosis determination unit determines that another diagnosis person should diagnose the diagnosis target, the other diagnosis person selected by the diagnosis person selection part and the selected person A medical information processing system that further outputs that other diagnosticians should make a diagnosis. More frequently, the reliability calculation unit stores the reliability of the initial diagnosis result acquired by the diagnosis result acquisition unit in the diagnosis information storage unit in association with the diagnostician who made the initial diagnosis result. The medical information processing system according to claim 1, which is calculated to be higher. The diagnostic information storage unit stores the frequency at which the definitive diagnosis result and the initial diagnosis result coincide with each other in association with the diagnostic result by the diagnostician and the diagnostician,
The reliability calculation unit stores the reliability of the initial diagnosis result acquired by the diagnosis result acquisition unit in the diagnosis information storage unit in association with the initial diagnosis result and a diagnostician who has made the initial diagnosis result. The medical information processing system according to claim 2, wherein the higher the frequency, the higher the calculation frequency. The diagnostic information storage unit stores an initial diagnosis result, a definitive diagnosis result, and a frequency at which the definitive diagnosis result is given to a diagnosis object for which the diagnostician has made the initial diagnosis result in association with the diagnostician. ,
The medical information processing system includes:
Based on the information stored in the diagnostic information storage unit in association with the diagnostician who made the initial diagnosis result, an index indicating the validity of the initial diagnostic result is calculated for each diagnostician who made the initial diagnostic result An index calculation unit;
The medical information processing system according to claim 3, wherein the reliability calculation unit calculates a reliability of an initial diagnosis result indicating that the index calculated by the index calculation unit is more appropriate. The diagnostic information storage unit stores an initial diagnosis result, a definitive diagnosis result, and the number of the definitive diagnosis results for the diagnosis target for which the diagnostician has made the initial diagnosis result in association with the diagnoser. The medical information processing system according to claim 4. The index calculation unit is based on information stored in the diagnostic information storage unit in association with a diagnostician who has made an initial diagnosis result, specificity for the initial diagnosis result, positive predictive value for the initial diagnosis result, initial diagnosis For each diagnoser who made the initial diagnosis result, at least one of the negative predictive value for the result and the probability that the diagnosis result indicated by the definitive diagnosis result when the diagnosis person gave the diagnosis result indicated by the initial diagnosis result Calculate
The medical information processing according to claim 5, wherein the reliability calculation unit calculates the reliability of the initial diagnosis result having a greater specificity, a positive prediction value, a negative prediction value, or a probability calculated by the index calculation unit. system. The index calculation unit, when the other person diagnosis determination unit determines that another diagnosis person should diagnose the diagnosis target, based on the information stored in the diagnosis information storage unit, the reliability Specificity, positive predictive value, negative predictive value for the diagnostic result whose reliability is calculated by the calculation unit, and the probability that the diagnostic result indicated by the definitive diagnostic result will be given when the diagnostician gives the diagnostic result indicated by the initial diagnostic result Is calculated for each diagnostician stored in the diagnostic information storage unit,
The diagnostician selection unit, when the other person diagnosis determination unit determines that the other diagnostician should diagnose the diagnosis target, the specificity calculated by the index calculation unit, a positive prediction value, a negative prediction value The medical information processing system according to claim 6, wherein a doctor with higher probability is selected with higher priority. Based on the information stored in the diagnostic information storage unit, the frequency that is given in the definitive diagnosis when the initial diagnosis result obtained by the diagnostic result acquisition unit is made in the initial diagnosis is higher than a predetermined frequency Further comprising another diagnostic result specifying unit for specifying other diagnostic results,
The index calculation unit, when the other person diagnosis determination unit determines that the other diagnosis person should diagnose the diagnosis target, based on the information stored in the diagnosis information storage unit, the other diagnosis Specificity, positive predictive value, negative predictive value for the diagnosis result specified by the result specifying unit, and at least the probability that the diagnosis result indicated by the definitive diagnosis result is given in the definitive diagnosis when the diagnosis result is made in the initial diagnosis One is calculated for each diagnostician stored in the diagnostic information storage unit,
The diagnostician selection unit calculates the index for the diagnosis result specified by the other diagnosis result specifying unit when the other person diagnosis determination unit determines that the other diagnostician should diagnose the diagnosis target. The medical information processing system according to claim 7, wherein a diagnosis person with higher specificity, positive predictive value, negative predictive value, or higher probability is selected with higher priority. The frequency of the other diagnosis result specifying unit to be determined in the definitive diagnosis when the initial diagnosis result acquired by the diagnosis result acquiring unit is determined in the initial diagnosis based on the information stored in the diagnosis information storage unit The medical information processing system according to claim 8, wherein the other diagnosis result having a frequency higher than a predetermined frequency is specified for each diagnostician who made the initial diagnosis result. The diagnosis result and the frequency stored in the diagnostic information storage unit are extracted in association with the diagnosis person and the initial diagnosis result made by the diagnosis person, and a total value obtained by summing the extracted frequencies over a plurality of diagnosis persons is determined. It further includes an index calculation unit that calculates each diagnosis result,
The medical information processing system according to claim 3, wherein the reliability calculation unit calculates a reliability of a definite diagnosis result having a larger total value calculated by the index calculation unit. Computer
A diagnostic result acquisition unit that acquires initial diagnostic results made by each of a plurality of diagnosticians for the same diagnostic target,
The reliability of the initial diagnosis result acquired by the diagnosis result acquisition unit is associated with the diagnoser who has made the initial diagnosis result, and is associated with the diagnoser who has made the initial diagnosis result. A reliability calculation unit that calculates based on the frequency stored in the diagnostic information storage unit that stores the frequency of matching
Other diagnosis determination unit that determines whether or not other diagnosticians of the plurality of diagnosticians should diagnose the diagnosis target based on the reliability of the initial diagnosis result calculated by the reliability calculation unit;
When the other person diagnosis determination unit determines that another diagnosis person should diagnose the diagnosis object, the initial diagnosis result given by each of the plurality of diagnosis persons and the diagnosis result indicated by the initial diagnosis result Based on the information stored in the diagnostic information storage unit in association with the initial diagnosis result or the definitive diagnosis result that coincides with the priority, the other diagnosticist who can diagnose the diagnosis object with high validity is given priority. The diagnostician selection unit to select,
An output unit that outputs the initial diagnosis result acquired by the diagnosis result acquisition unit in accordance with the reliability calculated by the reliability calculation unit, wherein the other person diagnosis determination unit assigns the diagnosis target to another diagnosis person Functioning as an output unit for further outputting that the other diagnostician selected by the diagnostician selecting unit and the selected other diagnostician should be diagnosed when it is determined that the diagnostic should be performed Program to let you. A method of operating a medical information processing system,
Diagnostic information storage unit provided in the medical information processing system associates with a diagnostician who has made an initial diagnosis result, and stores a diagnosis information storage stage for storing the frequency at which the final diagnosis result and the initial diagnosis result coincide with each other,
A diagnostic result acquisition unit in which the diagnostic result acquisition unit included in the medical information processing system acquires an initial diagnostic result performed by each of a plurality of diagnosticians for the same diagnostic target;
The reliability calculation unit included in the medical information processing system gives the initial diagnosis result by associating the reliability of the initial diagnosis result acquired in the diagnosis result acquisition stage with the diagnostician who made the initial diagnosis result. in association with the diagnostician, the reliability calculation step of the diagnostic information storage unit is calculated based on the frequency that contains that stores the frequency at which the definitive diagnosis result and the initial diagnosis results match,
The other person diagnosis determination unit included in the medical information processing system causes the other diagnosticians to diagnose the diagnosis target based on the reliability of the initial diagnosis result calculated in the reliability calculation stage. The other person's diagnosis judgment stage to judge whether or not
Each of the plurality of diagnosticians has made a decision when the diagnostician selection unit included in the medical information processing system determines that the diagnostic target should be diagnosed by another diagnostician in the other diagnostic determination stage. Based on the initial diagnosis result and information stored in association with the initial diagnosis result or the definite diagnosis result that matches the diagnosis result indicated by the initial diagnosis result, the diagnosis object can be diagnosed with high validity. A diagnostician selection step of preferentially selecting the other diagnosticians;
The output unit included in the medical information processing system outputs the initial diagnosis result acquired in the diagnosis result acquisition step according to the reliability calculated in the reliability calculation step, and the other diagnosis determination step When it is determined that another diagnosis person should diagnose the diagnosis target, the other diagnosis person selected in the diagnosis person selection stage and the other diagnosis person selected are further output. An operation method of a medical information processing system comprising an output stage.

Images