Program, information processing device, information processing method, and information processing system

Abstract

This prevents overlooking analysis results obtained through computer processing of medical images. [Solution] The information processing device acquires a first analysis result obtained by computer processing of the patient's current image (first medical image) (step S13), acquires a second analysis result obtained by computer processing of a past image of the same patient as the current image (second medical image) (step S14), compares the first analysis result and the second analysis result (step S15), and if there is a difference (step S16; YES), outputs the result in a different manner than when there is no difference (step S17).

Description

Technical Field

[0001] The present invention relates to a program, an information processing apparatus, an information processing method, and an information processing system.

Background Art

[0002] In recent years, a diagnostic support (CAD: Computer Aided Detection / Diagnosis) system has been put into practical use, which analyzes medical images using a computer, presents lesion candidates detected by the image analysis to a doctor, and asks for the doctor's judgment.

[0003] For example, in a diagnostic support system that analyzes medical images using a diagnostic support program, a technique has been proposed to display the analysis result for a medical image after recognizing the presence of additional information (such as a doctor's signature, an annotation created by the doctor, etc.) indicating that the doctor's diagnosis has been completed (see Patent Document 1). In the technique described in Patent Document 1, since the analysis result by the diagnostic support program is displayed when the doctor's diagnosis is completed, the doctor can make a fair diagnosis (reading the image) without being affected by the CAD result.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] However, in the prior art such as Patent Document 1, when there are many lesion candidates detected by CAD in a medical image, the doctor may overlook the lesion candidates to be seen. Furthermore, when CAD presents potential lesions it has detected, it is possible to narrow down the list of lesions to be displayed based on their confidence level and importance. However, depending on the settings, the number of lesions presented may become too large.

[0006] Thus, for radiologists, reviewing all potential lesions included in computer-generated analysis results is a significant burden, and there is a risk of overlooking lesions. Therefore, radiologists are required to efficiently identify noteworthy analysis results from CAD analysis while interpreting medical images.

[0007] The present invention has been made in view of the problems in the prior art described above, and aims to prevent the oversight of analysis results obtained by computer processing of medical images. [Means for solving the problem]

[0008] To solve the above problems, the invention described in claim 1 is a program for a computer to implement: a first acquisition function for acquiring a first analysis result obtained by computer processing of a first medical image of a patient; a second acquisition function for acquiring a second analysis result obtained by computer processing of a second medical image of the patient; and an output function for comparing the first analysis result and the second analysis result and outputting them in a different manner than when there is no difference, if there is a difference.

[0009] The invention described in claim 2 is that, in the program described in claim 1, the output function outputs information indicating that there is a difference when there is a difference, thereby outputting a different output pattern than when there is no difference.

[0010] The invention described in claim 3 is that, in the program described in claim 2, the output function notifies the user of information indicating that there is a difference.

[0011] The invention described in claim 4 is a program according to any one of claims 1 to 3, wherein the output function, when there is a difference, changes the display mode of the mouse cursor to output a different output mode than when there is no difference.

[0012] The invention described in claim 5 is a program according to any one of claims 1 to 4, wherein the output function, when there is a difference, changes the display mode of the window to output a different output mode than when there is no difference.

[0013] The invention described in claim 6 is a program according to any one of claims 1 to 5, wherein the output function highlights the portion where there is a difference between the first analysis result and the second analysis result, thereby outputting a different output pattern than when there is no difference.

[0014] The invention described in claim 7 is that, in the program described in claim 6, the highlighting is displayed by changing or adding at least one of the colors, shapes, characters, and marks of the portion that differs between the first analysis result and the second analysis result.

[0015] The invention described in claim 8 is a program according to any one of claims 1 to 5, wherein the output function displays only the portion where there is a difference between the first analysis result and the second analysis result, thereby outputting in a different manner than when there is no difference.

[0016] The invention described in claim 9 is the program described in any one of claims 1 to 8, wherein the first medical image is an image taken at a different date and time than the second medical image.

[0017] The invention described in claim 10 is a program according to any one of claims 1 to 9, wherein the computer processing includes processing for detecting lesion candidates from the first medical image or the second medical image, and the difference between the first analysis result and the second analysis result includes that a lesion candidate detected from the first medical image is not detected from the second medical image, or that a lesion candidate not detected from the first medical image is detected from the second medical image.

[0018] The invention described in claim 11 is a program according to any one of claims 1 to 9, wherein the computer processing includes processing for detecting lesion candidates from the first medical image or the second medical image, and the difference between the first analysis result and the second analysis result includes a difference of a predetermined amount or more in size or range between the lesion candidate detected from the first medical image and the lesion candidate detected from the second medical image.

[0019] The invention described in claim 12 is an information processing device comprising: a first acquisition means for acquiring a first analysis result obtained by computer processing of a first medical image of a patient; a second acquisition means for acquiring a second analysis result obtained by computer processing of a second medical image of the patient; and an output means for comparing the first analysis result and the second analysis result and outputting a different output configuration if there is a difference compared to when there is no difference.

[0020] The invention described in claim 13 is an information processing method comprising: a first acquisition step of acquiring a first analysis result obtained by computer processing of a first medical image of a patient; a second acquisition step of acquiring a second analysis result obtained by computer processing of a second medical image of the patient; and an output step of comparing the first analysis result and the second analysis result and outputting them in a different manner than when there is no difference, if there is a difference.

[0021] The invention according to claim 14 is an information processing system comprising: a first generation means for performing computer processing on a first medical image of a patient to generate a first analysis result; a second generation means for performing computer processing on a second medical image of the patient to generate a second analysis result; a first acquisition means for acquiring the first analysis result; a second acquisition means for acquiring the second analysis result; and an output means for comparing the first analysis result and the second analysis result and outputting them in different output modes when there are differences and when there are no differences.

Advantages of the Invention

[0022] According to the present invention, it is possible to prevent overlooking the analysis results obtained by computer processing of medical images.

Brief Description of the Drawings

[0023] [Figure 1] [[ID=十五]]It is a system configuration diagram of a medical image display system. [Figure 2] It is a block diagram showing the functional configuration of a medical image management server. [Figure 3] It is a diagram showing the data configuration of a data management table. [Figure 4] It is a block diagram showing the functional configuration of an information processing device. [Figure 5] It is a flowchart showing the medical image analysis process executed by a medical image management server. [Figure 6] (a) is an example when the position of a lesion candidate for a medical image is specified by a point. (b) is an example when the position of a lesion candidate for a medical image is specified by a circle. (c) is an example when the position of a lesion candidate for a medical image is specified by an ellipse. <() [Figure 7] It is a flowchart showing the analysis result output process executed by an information processing device. [Figure 8] It is a flowchart showing the analysis result comparison process. [Figure 9] It is an image diagram showing the comparison between the analysis results of a past image and a current image. [Figure 10] This is an example of a popup window in output example A. [Figure 11] This is an example of the image interpretation screen in output example B. [Figure 12] This is an example of the image interpretation screen in output example C. [Figure 13] (a) is an example of the image interpretation screen when there is no difference between the analysis results of the current image and the analysis results of the past image. (b) is an example of the image interpretation screen when there is a difference between the analysis results of the current image and the analysis results of the past image in output example D. [Figure 14] This is an example of the image interpretation screen in output example E. [Figure 15] This is an example of the image interpretation screen in output example F. [Figure 16] This is an example of the image interpretation screen in output example G. [Figure 17] This is an example of the image interpretation screen in output example H. [Figure 18] This is an example of the image interpretation screen in Output Example I. [Modes for carrying out the invention]

[0024] The following describes an embodiment of the program, information processing apparatus, information processing method, and information processing system according to the present invention. However, the scope of the invention is not limited to the illustrated example.

[0025] [Configuration of the medical image display system] Figure 1 shows the system configuration of the medical image display system 100 as an information processing system. As shown in Figure 1, the medical image display system 100 consists of a medical image management server 10, modalities 20, 20, ..., and an information processing device 30, and each device is capable of data communication via a communication network N. The information processing device 30 only needs to be able to connect to the medical image management server 10 via the communication network N as needed. The medical image display system 100 can display medical images and analysis results stored in the medical image management server 10 on the information processing device 30. Each device constituting the medical image display system 100 conforms to the HL7 (Health Level Seven) and DICOM (Digital Image and Communications in Medicine) standards, and communication between each device is performed in accordance with HL7 and DICOM. The number of information processing devices 30 is not particularly limited.

[0026] Modality 20 involves imaging the patient (subject) and generating medical image data. Examples of Modality 20 include CR (Computed Radiography), DR (Digital Radiography), CT (Computed Tomography), MRI (Magnetic Resonance Imaging), US (UltraSonography), NM (Nuclear Medicine), and ES (Endoscopy). Modality 20 also adds supplementary information related to the medical image. This supplementary information includes patient ID, patient name, date of birth, gender, date and time of shooting, image ID, modality, body part, and orientation. The image data of the medical image generated by modality 20 is sent to the medical image management server 10.

[0027] The medical image management server 10 stores and manages image data of medical images generated in modality 20. Examples of medical image management servers 10 include PACS (Picture Archiving and Communication System).

[0028] The information processing device 30 is a computer device such as a PC (Personal Computer) or a tablet. The information processing device 30 is used by users such as doctors when interpreting medical images.

[0029] [Configuration of the medical image management server] Figure 2 shows the functional configuration of the medical image management server 10. The medical image management server 10 is composed of a control unit 11, a communication unit 12, an image analysis unit 13, a storage unit 14, etc., and each unit is connected by a bus 15.

[0030] The control unit 11 consists of a CPU (Central Processing Unit), RAM (Random Access Memory), etc., and comprehensively controls the processing operations of each part of the medical image management server 10. Specifically, the CPU reads various processing programs stored in the memory unit 14, loads them into RAM, and performs various processing operations in cooperation with those programs.

[0031] The communication unit 12 is configured with a network interface and other components, and transmits and receives data with external devices connected via the communication network N. For example, the communication unit 12 receives medical images obtained by photographing a patient using modality 20. The communication unit 12 also transmits medical images and analysis results requested by the information processing device 30 to the information processing device 30.

[0032] The image analysis unit 13 performs computer processing on medical images obtained by photographing patients and generates analysis result data. The computer processing is computer-aided detection / diagnosis (CAD) processing. The image analysis unit 13 detects candidate lesions (candidate abnormal shadows) from medical images, adds annotations, and calculates indices for quantitatively evaluating medical images. The analysis result data consists of, for example, DICOM GSPS (Grayscale Softcopy Presentation State) data, overlay data, etc. The image analysis unit 13 is implemented by software processing through the cooperation of a program stored in the memory unit 14 and the CPU of the control unit 11.

[0033] The storage unit 14 is composed of an HDD (Hard Disk Drive) or non-volatile semiconductor memory, and stores various processing programs, parameters and files necessary for the execution of those programs.

[0034] Furthermore, the storage unit 14 stores a user management table 141 and a data management table 142. The storage unit 14 also has a medical image storage area 143 and an analysis result storage area 144.

[0035] The user management table 141 is a table for managing users (medical professionals such as doctors) who use the medical image display system 100. The user management table 141 stores each user's information, including their user ID, password, name, affiliation, email address, and telephone number. A user ID is a user's identification information. The password is used for authentication when a user accesses the medical image management server 10 from the information processing device 30. The name is the user's name. The affiliation information refers to the medical facility, department, etc., to which the user belongs. The email address is the user's email address. The phone number is the user's phone number.

[0036] The data management table 142 is a table for managing data within the medical image management server 10. Figure 3 shows the data structure of the data management table 142. The data management table 142 stores the image ID, patient ID, date and time of acquisition, modality, body part, orientation, analysis result ID, etc., associated with each medical image stored in the medical image storage area 143. Image ID is identification information for medical images. The patient ID is the identifying information of the patient whose medical image was taken. The date and time of capture refers to the date and time the medical image was taken. A modality is a modality in which medical images are taken. The body part is the area that was the subject of the medical image. The direction is the direction in which the medical image is captured. The analysis result ID is identification information of the analysis result obtained by the image analysis unit 13 analyzing the medical image.

[0037] The medical image storage area 143 stores image data of medical images received from the modality 20.

[0038] The analysis result storage area 144 stores data of the analysis results for medical images. The analysis result data includes the location, size, and type of candidate lesions detected from the medical images, the content and location of annotations added to the medical images, the location of measurement lines drawn on the medical images, and numerical values ​​calculated from the medical images.

[0039] When the control unit 11 receives a medical image from the modality 20 via the communication unit 12, it stores the medical image in the storage unit 14 and also has the image analysis unit 13 analyze the medical image.

[0040] When the control unit 11 receives a request from the information processing device 30 via the communication unit 12 to acquire a medical image, it reads the medical image from the storage unit 14 and provides the read medical image to the information processing device 30 via the communication unit 12. When the control unit 11 receives a request from the information processing device 30 via the communication unit 12 to acquire the analysis results, it reads the analysis results from the storage unit 14 and provides the read analysis results to the information processing device 30 via the communication unit 12.

[0041] [Configuration of the information processing device] Figure 4 shows the functional configuration of the information processing device 30. The information processing device 30 is composed of a control unit 31, an operation unit 32, a display unit 33, a communication unit 34, a storage unit 35, etc., and each unit is connected by a bus 36.

[0042] The control unit 31 consists of a CPU, RAM, etc., and comprehensively controls the processing operations of each part of the information processing device 30. Specifically, the CPU reads various processing programs stored in the memory unit 35, loads them into the RAM, and performs various processing operations in cooperation with those programs.

[0043] The operation unit 32 is configured with a keyboard equipped with cursor keys, character / number input keys, and various function keys, and a pointing device such as a mouse, and outputs operation signals input by key operations on the keyboard or mouse operations to the control unit 31. Furthermore, if the operation unit 32 is configured with a touch panel stacked on the display unit 33, it outputs operation signals to the control unit 31 according to the position of touch operations by the user's finger or the like.

[0044] The display unit 33 is equipped with a monitor such as an LCD (Liquid Crystal Display) and displays various screens according to the instructions of the display signals input from the control unit 31.

[0045] The communication unit 34 is composed of a network interface and the like, and performs data transmission and reception with external devices connected via the communication network N.

[0046] The storage unit 35 is composed of an HDD or non-volatile semiconductor memory, and stores various processing programs, parameters and files necessary for the execution of those programs. For example, the storage unit 35 stores an analysis result output processing program 351. The analysis result output processing program 351 is a program that performs processing to display the medical image to be interpreted and the analysis results when a medical image is interpreted in the information processing device 30.

[0047] The control unit 31 acquires the first analysis result obtained by computer processing of the patient's first medical image. In other words, the control unit 31 functions as the first acquisition means.

[0048] The control unit 31 acquires a second analysis result obtained by computer processing of a second medical image of the same patient as the first medical image. In other words, the control unit 31 functions as a second acquisition means.

[0049] The first medical image is an image taken at a different date and time than the second medical image. In this embodiment, we will describe the case where the first medical image is a current image and the second medical image is a past image. The current image is a newly taken medical image to be interpreted. The past image is a medical image taken in the past of the same patient as the current image (if there are multiple medical images taken earlier than the current image, the most recent one), and is the medical image to be used as a comparison for the analysis results.

[0050] The control unit 31 compares the first analysis result with the second analysis result and outputs a different output pattern if there is a difference between the two results, compared to when there is no difference. In other words, the control unit 31 functions as an output means.

[0051] The differences between the first and second analysis results include the fact that candidate lesions detected in the first medical image were not detected in the second medical image, and that candidate lesions not detected in the first medical image were detected in the second medical image. Furthermore, the difference between the first analysis result and the second analysis result includes a difference in size or extent of a predetermined amount or more between the candidate lesion detected from the first medical image and the candidate lesion detected from the second medical image.

[0052] The control unit 31 outputs information indicating that there is a difference between the first analysis result and the second analysis result, thereby differentiating the output from when there is no difference. "Outputting information indicating that there is a difference" includes, for example, displaying additional information indicating that there is a difference.

[0053] Specifically, the control unit 31 notifies the user of any discrepancies between the first and second analysis results. On the other hand, the control unit 31 does not notify the user if there are no discrepancies between the first and second analysis results. Notifications of discrepancies may include notifications via pop-up windows, email, telephone, or other communication methods.

[0054] For example, if there is a difference between the first analysis result and the second analysis result, the control unit 31 displays a pop-up window on the display unit 33 containing information indicating that there is a difference.

[0055] Furthermore, if there is a difference between the first analysis result and the second analysis result, the control unit 31 sends a request to the medical image management server 10 via the communication unit 34 to obtain the email address corresponding to the reader of the first medical image. The reader of the first medical image may be a user who selected the first medical image as the image to be read in the information processing device 30, or a user who is designated as the reader of the first medical image. The control unit 11 of the medical image management server 10 reads the email address corresponding to the reader of the first medical image from the user management table 141 of the storage unit 14 and sends this email address to the information processing device 30 via the communication unit 12. The control unit 31 of the information processing device 30 obtains the email address corresponding to the reader of the first medical image from the medical image management server 10 and sends an email containing information indicating the difference to this email address via the communication unit 34.

[0056] Furthermore, if there is a difference between the first analysis result and the second analysis result, the control unit 31 sends a request to the medical image management server 10 via the communication unit 34 to obtain the telephone number corresponding to the reader of the first medical image. The control unit 11 of the medical image management server 10 reads the telephone number corresponding to the reader of the first medical image from the user management table 141 of the storage unit 14 and sends this telephone number to the information processing device 30 via the communication unit 12. The control unit 31 of the information processing device 30 obtains the telephone number corresponding to the reader of the first medical image from the medical image management server 10 and performs call processing based on voice data containing information indicating that there is a difference, addressed to this telephone number.

[0057] Furthermore, notifications regarding information indicating discrepancies may be sent via email or telephone from the medical image management server 10.

[0058] The control unit 31 changes the display mode of the mouse cursor when there is a difference between the first analysis result and the second analysis result, thereby outputting a different output mode than when there is no difference. Specifically, when there is a difference between the first analysis result and the second analysis result, the control unit 31 changes the color, shape, etc. of the mouse cursor on the medical image display screen to what it would be in the normal case (when there is no difference). On the other hand, when there is no difference between the first analysis result and the second analysis result, the control unit 31 sets the display mode of the mouse cursor to the normal state (default color, shape, etc.).

[0059] The control unit 31, when there is a difference between the first analysis result and the second analysis result, changes the display mode of the window to output a different output mode than when there is no difference. The window is the display area of ​​the application. For example, the display area of ​​an application that displays medical images and analysis results can be used as the window whose display mode is changed. Changes to the display mode of the window include changing the window frame (color, type, etc.) and changing the color of the background area relative to the medical image display area. The background area includes the tool display area within the window. On the other hand, when there is no difference between the first analysis result and the second analysis result, the control unit 31 returns the display mode of the window to the normal state (default state).

[0060] If there is a difference between the first analysis result and the second analysis result, the control unit 31 highlights the part where there is a difference between the first and second analysis results, thereby outputting a different output pattern than when there is no difference. The highlighting involves changing or adding at least one of the following: color, shape, characters, and marks, to the part where there is a difference between the first and second analysis results. Highlighting differences means, for example, displaying information that is included in the first analysis result but not in the second analysis result, or information that is not included in the first analysis result but is included in the second analysis result, in a different manner from the parts that do not differ. Specifically, if there is a difference between the first analysis result and the second analysis result, the control unit 31 highlights the portion of the first analysis result that differs from the second analysis result on the display screen of the first medical image. On the other hand, if there is no difference between the first analysis result and the second analysis result, the control unit 31 sets the display mode of the first medical image display screen to the normal state (e.g., with the first analysis result superimposed on the first medical image).

[0061] When there is a difference between the first and second analysis results, the control unit 31 displays only the portion of the first and second analysis results that differs, thereby outputting a different output pattern than when there is no difference. For example, when there is a difference between the first and second analysis results, the control unit 31 overlays only the portion of the first medical image that differs from the first medical image on the display screen of the first medical image, and hides the portion of the first medical image that does not differ from the second analysis result.

[0062] [Operation in medical image display systems] Next, the operation of the medical image display system 100 will be described. Figure 5 is a flowchart showing the medical image analysis process performed by the medical image management server 10. This process is realized through software processing involving the cooperation of the CPU of the control unit 11 and the program stored in the memory unit 14.

[0063] In the medical image management server 10, when it receives a medical image obtained by photographing a patient from the modality 20 via the communication unit 12 (step S1), the control unit 11 stores the received medical image in the medical image storage area 143 of the storage unit 14 (step S2). The control unit 11 also associates the image ID, patient ID, date and time of shooting, modality, body part, direction, etc., included in the supplementary information of the received medical image and stores them in the data management table 142 of the storage unit 14 (see Figure 3).

[0064] Next, the control unit 11 causes the image analysis unit 13 to analyze the medical image using its CAD function (step S3). The image analysis unit 13 analyzes the medical image using AI (Artificial Intelligence) and generates analysis result data. Specifically, the image analysis unit 13 detects potential lesions from the medical image and generates analysis results including the location, size, and type of the potential lesions.

[0065] The locations of potential lesions included in the analysis results are specified by points, circles, ellipses, etc. (1) When specified by point When the location of a potential lesion in a medical image is specified by a point, the position (vertical coordinates, horizontal coordinates) of point P1, which is the center of the potential lesion, is specified, as shown in Figure 6(a). In this case, the size and extent of the potential lesion are not specified, and the annotations (circles, squares, etc.) displayed at the location of the potential lesion on the medical image display screen are viewer-dependent.

[0066] (2) When specified as a circle (CIRCLE) When the location of a potential lesion in a medical image is specified by a circle, as shown in Figure 6(b), the position (vertical coordinates, horizontal coordinates) of point P2, which is the central pixel of circle C1, and the position (vertical coordinates, horizontal coordinates) of pixels on the circumference of circle C1 are specified.

[0067] (3) When specified as an ellipse When the location of a potential lesion in a medical image is specified by an ellipse, as shown in Figure 6(c), the positions of four points P3 to P6 (vertical coordinates, horizontal coordinates) are specified, where each side of the smallest rectangle E2 surrounding the ellipse E1 touches the ellipse E1.

[0068] Next, the control unit 11 stores the analysis results obtained by the image analysis unit 13 in the storage unit 14 in association with the medical image (step S4). Specifically, the control unit 11 stores the analysis results in the analysis result storage area 144 of the storage unit 14, and also stores the analysis result ID of the analysis result in the record corresponding to the medical image that was analyzed in the data management table 142 of the storage unit 14. This completes the medical image analysis process.

[0069] Thus, the image analysis unit 13 of the medical image management server 10 functions as a first generation means that performs computer processing on a first medical image of a patient and generates a first analysis result. Furthermore, the image analysis unit 13 can function as a second generation means that performs computer processing on a second medical image of the same patient as the first medical image and generates a second analysis result.

[0070] Figure 7 is a flowchart showing the analysis result output processing performed by the information processing device 30. This processing is realized through software processing in cooperation between the CPU of the control unit 31 and the analysis result output processing program 351 stored in the storage unit 35.

[0071] When a user (doctor) accesses the medical image management server 10 from the information processing device 30, they enter their user ID and password through an operation on the operation unit 32. The control unit 31 then transmits the entered user ID and password to the medical image management server 10 via the communication unit 34. When the medical image management server 10 receives a user ID and password via the communication unit 12, the control unit 11 determines whether the received user ID and password match any of the user ID and password combinations pre-registered in the user management table 141 of the storage unit 14. If a match is found, the control unit 11 grants permission for the user to use the information processing device 30.

[0072] The control unit 31 of the information processing device 30 acquires the current image from the medical image management server 10 via the communication unit 34 (step S11). For example, the control unit 31 sends a request to acquire a medical image, including an image ID specified by the user's operation unit 32, to the medical image management server 10 via the communication unit 34. The control unit 11 of the medical image management server 10 reads the medical image corresponding to the image ID included in the acquisition request from the medical image storage area 143 and sends the medical image to the information processing device 30 via the communication unit 12. The control unit 31 of the information processing device 30 may also acquire the current image from the modality 20.

[0073] Next, the control unit 31 of the information processing device 30 obtains past images of the same patient as the current image from the medical image management server 10 via the communication unit 34 (step S12). For example, the control unit 31 sends a request to the medical image management server 10 via the communication unit 34 to obtain past images, including the image ID of the current image. The control unit 11 of the medical image management server 10 refers to the data management table 142 of the storage unit 14 and extracts the most recent record in which the "patient ID," "modality," "location," and "direction" are the same as the current image, and the "shooting date and time" is earlier than the current image, and identifies the image ID (image ID of the past image) included in the extracted record. The control unit 11 reads the medical image corresponding to the image ID of the past image from the medical image storage area 143 and transmits the medical image to the information processing device 30 via the communication unit 12. The conditions used to identify past images are not limited to the examples above.

[0074] Next, the control unit 31 of the information processing device 30 obtains the analysis results of the current image from the medical image management server 10 via the communication unit 34 (step S13). For example, the control unit 31 sends a request to the medical image management server 10 via the communication unit 34 to obtain the analysis results of the current image. The control unit 11 of the medical image management server 10 refers to the data management table 142 of the storage unit 14 to identify the "analysis result ID" from the record corresponding to the "image ID" of the current image, reads the analysis result corresponding to this "analysis result ID" from the analysis result storage area 144, and sends the analysis result to the information processing device 30 via the communication unit 12.

[0075] Next, the control unit 31 of the information processing device 30 obtains the analysis results of past images from the medical image management server 10 via the communication unit 34 (step S14). For example, the control unit 31 sends a request to the medical image management server 10 via the communication unit 34 to obtain the analysis results of past images. The control unit 11 of the medical image management server 10 refers to the data management table 142 of the storage unit 14 to identify the "analysis result ID" from the record corresponding to the "image ID" of the past image, reads the analysis result corresponding to this "analysis result ID" from the analysis result storage area 144, and sends the analysis result to the information processing device 30 via the communication unit 12.

[0076] Next, the control unit 31 of the information processing device 30 performs an analysis result comparison process (step S15). The analysis result comparison process compares the analysis results of the current image with the analysis results of past images.

[0077] Now, with reference to Figure 8, we will explain the process of comparing the analysis results. The control unit 31 of the information processing device 30 performs alignment of the current image and the past image (step S21). Specifically, the control unit 31 applies scaling, rotation, etc., to both or either the current image and the past image to align the positions of bones, organs, etc., within the images between the current image and the past image.

[0078] Next, the control unit 31 adjusts the positions of the analysis results of the current image and the past image in conjunction with the alignment of the current image and the past image (step S22). Specifically, the control unit 31 converts the positions of lesion candidates, etc., included in the analysis results of the current image or the analysis results of the past image to the positions after the alignment of each image.

[0079] Next, the control unit 31 calculates the degree of overlap of lesion regions for a given lesion candidate based on the analysis results of the current image after position adjustment and the analysis results of the past image (step S23). For example, the degree of overlap of lesion regions is used as the ratio of "the overlapping region of the lesion region in the current image and the lesion region in the past image" to "the lesion region in the past image". Furthermore, for candidate lesions that exist in only one of the current image analysis results and the past image analysis results, the degree of overlap of the lesion area will be considered to be 0.

[0080] If the analysis results indicate that the location of a candidate lesion is specified as a single point (center position) (see Figure 6(a)), the control unit 31 calculates the percentage (degree of overlap) of overlap between pre-specified ranges (e.g., circles of a predetermined radius) based on the lesion center position in the analysis results of the current image and the analysis results of the past image. Here, the pre-specified range can be changed for each medical facility (medical facility where the imaging was performed, medical facility to which the user belongs, etc.), requesting department (clinical department to which the physician who requested imaging or interpretation belongs), and imaging area. Alternatively, the user (physician) may specify any value for the pre-specified range. Furthermore, the pre-specified range values ​​may be adjustable for the upper and lower parts of the medical image. For example, in the case of a chest image, the position of the diaphragm changes according to the patient's breathing, so the vertical position of the same lesion can vary significantly in the lower part of the chest image compared to the upper part, depending on when the image was taken.

[0081] If the location of a candidate lesion is indicated by a circle in the analysis results (see Figure 6(b)), the control unit 31 calculates the percentage (degree of overlap) of overlap between the circles indicating the location of the candidate lesion in the current image analysis results and the past image analysis results.

[0082] If the location of a candidate lesion is indicated by an ellipse in the analysis results (see Figure 6(c)), the control unit 31 calculates the percentage (degree of overlap) of overlap between the ellipses indicating the location of the candidate lesion in the current image analysis results and the past image analysis results.

[0083] Next, the control unit 31 determines whether the degree of overlap between the analysis results of the current image and the analysis results of the past image is above a predetermined threshold (step S24). Depending on this threshold, it is possible to detect not only the difference indicating the presence or absence of lesion candidates in both images, but also the difference indicating that the size or range of lesion candidates in both images has changed by a predetermined amount or more. The threshold used here may be changed for each medical facility, requesting department, imaging site, and user. In addition, different threshold values ​​may be used for the upper and lower parts of the medical image.

[0084] If the degree of overlap between the lesion area in the current image and the past image is greater than or equal to a threshold (step S24; YES), the control unit 31 determines that the lesion candidate detected from the current image matches the lesion candidate detected from the past image (step S25).

[0085] In step S24, if the degree of overlap between the lesion area in the current image and the past image is less than a threshold (step S24; NO), the control unit 31 determines that the lesion candidate detected from the current image and the lesion candidate detected from the past image are inconsistent (step S26).

[0086] After step S25 or step S26, the control unit 31 determines whether the comparison of all lesion candidates included in the analysis results of the current image and the analysis results of the past image has been completed (step S27). If there are any candidate lesions that have not yet been compared (step S27; NO), the process returns to step S23, and the control unit 31 repeats the process with a different candidate lesion as the target for processing.

[0087] In step S27, if the comparison of all candidate lesions has been completed (step S27; YES), the analysis result comparison process is terminated.

[0088] Alternatively, instead of using the degree of overlap in step S24, the control unit 31 may determine that the lesion candidate detected from the current image and the lesion candidate detected from the past image are inconsistent (a significant change has occurred) if the difference in the size (length, area, etc.) of the lesion candidate between the analysis results of the current image and the analysis results of the past image is greater than or equal to a predetermined amount. Furthermore, the control unit 31 may determine that the lesion candidate detected from the current image and the lesion candidate detected from the past image are inconsistent (a significant change has occurred) if the difference in the range of the lesion candidate (area of ​​non-overlapping regions, etc.) between the analysis results of the current image and the analysis results of the past image is greater than or equal to a predetermined amount.

[0089] Figure 9 is an illustrative diagram showing a comparison between the analysis results of past images 40 and the analysis results of current images 50. In the example in Figure 9, the analysis results of past images 40 included 5 lesion candidates 41-45, while the analysis results of current images 50 include 6 lesion candidates 51-56. The lesion candidates 51-55 included in the analysis results of current images 50 are thought to correspond to (be the same lesions as) the lesion candidates 41-45 included in the analysis results of past images 40. The analysis results of current images 50 also include a newly detected lesion candidate 56.

[0090] After the analysis result comparison process, the process returns to Figure 7, and the control unit 31 of the information processing device 30 determines whether there is a difference between the analysis results of the current image and the analysis results of the past image (step S16). Specifically, the control unit 31 determines that there is a difference if there are lesion candidates that are determined to be inconsistent between the analysis results of the current image and the analysis results of the past image. On the other hand, the control unit 31 determines that there is no difference if all lesion candidates are determined to be the same between the analysis results of the current image and the analysis results of the past image.

[0091] If there is a difference between the analysis results of the current image and the analysis results of the past image (step S16; YES), the control unit 31 of the information processing device 30 outputs in a different manner than when there is no difference (step S17). For example, when the control unit 31 displays the current image and the analysis results of the current image on the display unit 33, it displays them according to the output example described later.

[0092] If, after step S17, or in step S16, there is no difference between the analysis results of the current image and the analysis results of the past image (step S16; NO), the analysis result output process is terminated.

[0093] The following describes specific output examples A to I in step S17 of the analysis result output processing (see Figure 7). In output examples A to I, the analysis results 50 of the current image and 40 of the past image shown in Figure 9 will be used. Output examples A to D (see Figures 10 to 13) are examples that show that there is a difference between the analysis results of the current image and the analysis results of the past image. Output examples E to H (see Figures 14 to 17) are examples that highlight the parts where there is a difference between the analysis results of the current image and the analysis results of the past image. Output example I (see Figure 18) is an example that displays only the parts where there is a difference between the analysis results of the current image and the analysis results of the past image.

[0094] <Output Example A> The control unit 31 of the information processing device 30 displays a message on the display unit 33 to notify the user that there is a difference between the analysis results of the current image and the analysis results of the past image.

[0095] Figure 10 shows an example of a pop-up window 331 displayed on the display unit 33. The pop-up window 331 is displayed, for example, on the current image interpretation screen (a screen in which the analysis results of the current image are superimposed on the current image). The pop-up window 331 includes a message display area 331A, an OK button 331B, and a cancel button 331C. The message display area 331A shows a message indicating that a candidate lesion that was not detected in the previous image (the image taken last time) has been detected in the current image (the image taken this time).

[0096] The OK button 331B is a button used by the user to indicate that they have confirmed a difference between the analysis results of the current image and the analysis results of the past image. When the user operates the operation unit 32 and presses the OK button 331B, the control unit 31 displays the current image with the analysis results (annotations) of the current image superimposed on it, and the past image with the analysis results of the past image superimposed on it, side by side on the display unit 33 (comparison image interpretation screen).

[0097] The cancel button 331C is a button used by the user to confirm that there is a difference between the analysis results of the current image and the analysis results of past images, and to instruct the user to continue interpreting the current image. When the user operates the control unit 32 and presses the cancel button 331C, the control unit 31 hides the pop-up window 331 and continues to display only the current image with the analysis results of the current image superimposed.

[0098] The control unit 31 does not notify a message like the one shown in Figure 10 if there is no difference between the analysis results of the current image and the analysis results of past images.

[0099] <Output Example B> The control unit 31 of the information processing device 30 changes the display mode of the mouse cursor from the normal display mode when there is a difference between the analysis results of the current image and the analysis results of past images. Figure 11 shows an example of the image interpretation screen 332 displayed on the display unit 33. On the image interpretation screen 332, the analysis results of the current image are superimposed on the current image. Also, the mouse cursor 332A on the image interpretation screen 332 has a different color than usual. The shape of the mouse cursor 332A may also be changed.

[0100] The control unit 31 does not change the display mode of the mouse cursor if there is no difference between the analysis results of the current image and the analysis results of past images.

[0101] <Output Example C> The control unit 31 of the information processing device 30 changes the display mode of the window from the normal display mode when there is a difference between the analysis results of the current image and the analysis results of past images. Figure 12 shows an example of the image interpretation screen 333 displayed on the display unit 33. The image interpretation screen 333 displays the analysis results of the current image superimposed on the current image. In addition, the window frame 333A of the image interpretation screen 333 has a different color compared to the normal case.

[0102] The control unit 31 does not change the window display mode if there is no difference between the analysis results of the current image and the analysis results of past images.

[0103] <Output Example D> Output Example D, like Output Example C, is an example of how to notify the user of a difference between the analysis results of the current image and the analysis results of past images by changing the display mode of the window. Figure 13(a) shows an example of the image interpretation screen 334 when there is no difference between the analysis results of the current image and the analysis results of the past image, and Figure 13(b) shows an example of the image interpretation screen 335 when there is a difference between the analysis results of the current image and the analysis results of the past image.

[0104] As shown in Figure 13(a), the image interpretation screen 334 when there are no differences includes a window frame 334A, a medical image display area 334B, out-of-image areas 334C and 334D, and a tool display area 334E. The medical image display area 334B displays the analysis results of the current image superimposed on the current image. The out-of-image areas 334C and 334D are background areas relative to the medical image display area 334B. The tool display area 334E displays tool selection buttons for selecting processing for the medical image. The tool display area 334E is part of the background area relative to the medical image display area 334B.

[0105] As shown in Figure 13(b), when there is a difference, the image interpretation screen 335 includes a window frame 335A, a medical image display area 335B, an area outside the image 335C, 335D, and a tool display area 335E. The window frame 335A, medical image display area 335B, area outside the image 335C, 335D, and tool display area 335E are the same as the window frame 334A, medical image display area 334B, area outside the image 334C, 334D, and tool display area 334E of the image interpretation screen 334, respectively.

[0106] In comparison with Figure 13(a), Figure 13(b) shows that the color of the window frame 335A has been changed, as has the color of the background area of ​​the medical image display area 335B (out-of-image areas 335C, 335D, and tool display area 335E). This allows the user to be prompted to pay attention to the analysis results of the current image when there is a difference between the analysis results of the current image and the analysis results of past images.

[0107] <Output Example E> The control unit 31 of the information processing device 30 highlights the parts that differ between the analysis results of the current image and the analysis results of the past image when there is a difference between them. In particular, it is desirable for the control unit 31 to highlight newly presented lesion candidates in the current analysis results by comparing them with the previous analysis results of the patient being read. The same applies to output examples F to H. On the other hand, if there is no difference between the analysis results of the current image and the analysis results of the past image, the control unit 31 will display the image in the normal manner.

[0108] Figure 14 shows an example of the image interpretation screen 336 displayed on the display unit 33. On the image interpretation screen 336, the analysis results of the current image are superimposed on the current image. On the image interpretation screen 336, the color of annotation 66 corresponding to lesion candidates that were not detected in the analysis results of past images is represented in a different color from the other annotations 61-65, thereby highlighting the differences between the analysis results of the current image and the analysis results of past images.

[0109] Furthermore, not only lesion candidates newly identified in the current image analysis results, but also lesion candidates that have changed significantly from the analysis results of past images to the analysis results of current images (those that are considered to be the same lesion, but whose size or extent has changed by more than a certain amount) will be highlighted as areas of difference. For lesion candidates that were detected in past images but not in current images, the difference may be highlighted on the past image, or on a comparative image interpretation screen that includes both past and current images.

[0110] <Output Example F> Figure 15 shows an example of the image interpretation screen 337 displayed on the display unit 33. On the image interpretation screen 337, the analysis results of the current image are superimposed on the current image. On the image interpretation screen 337, the shape of annotation 76 (a rectangle in Figure 15) corresponding to a candidate lesion that was not detected in the analysis results of past images is represented by a different shape from the other annotations 71-75 (circles in Figure 15), thereby highlighting the differences between the analysis results of the current image and the analysis results of past images.

[0111] <Output Example G> Figure 16 shows an example of the image interpretation screen 338 displayed on the display unit 33. The image interpretation screen 338 displays the analysis results of the current image superimposed on the current image. On the image interpretation screen 338, the words "Newly Detected" 338A are added near annotations 86 that correspond to lesion candidates that were not detected in the analysis results of past images, thereby highlighting the differences between the analysis results of the current image and the analysis results of past images.

[0112] <Output Example H> Figure 17 shows an example of the image interpretation screen 339 displayed on the display unit 33. The image interpretation screen 339 displays the analysis results of the current image superimposed on the current image. Furthermore, the image interpretation screen 339 highlights the differences between the analysis results of the current image and the analysis results of the past image by adding marks 339A near annotations 96 that correspond to candidate lesions that were not detected in the analysis results of past images.

[0113] <Output Example I> The control unit 31 of the information processing device 30 displays only the portion of the current image analysis result that differs from the past image analysis result when there is a difference between the current image analysis result and the past image analysis result. On the other hand, if there is no difference between the analysis results of the current image and the analysis results of past images, the control unit 31 will superimpose the analysis results of the current image onto the current image (normal display). Alternatively, the control unit 31 may choose not to display the analysis results on the current image if there is no difference.

[0114] Figure 18 shows an example of the image interpretation screen 340 displayed on the display unit 33. The current image is displayed on the image interpretation screen 340. Furthermore, annotations 340A corresponding to lesion candidates that were not detected in past images but were detected only in the current image are superimposed on the current image on the image interpretation screen 340. Conversely, for lesion candidates that were detected in past images but not in the current image, annotations indicating the area with dashed lines or the like may be superimposed on the current image.

[0115] As described above, according to this embodiment, the control unit 31 of the information processing device 30 compares the analysis result of the current image (first analysis result) with the analysis result of the past image (second analysis result), and outputs a different output pattern when there is a difference between the two compared to when there is no difference, thereby preventing the oversight of analysis results obtained by computer processing of medical images.

[0116] For example, if the control unit 31 finds a difference between the analysis results of the current image and the analysis results of the past image, it outputs information indicating that there is a difference. This allows the user to be alerted when interpreting the current image while referring to the analysis results of the current image, or when comparing and interpreting the current image with the past image. Specifically, information indicating discrepancies can be sent to the user via pop-up windows, email, phone calls, etc.

[0117] Furthermore, if there is a difference between the analysis results of the current image and the analysis results of past images, the control unit 31 can prompt the user to pay attention to the analysis results by changing the display of the mouse cursor. Furthermore, if there is a difference between the analysis results of the current image and the analysis results of past images, the control unit 31 can change the display mode of the window to draw the user's attention to the analysis results.

[0118] Furthermore, the control unit 31 can draw the user's attention to areas where there have been changes from past images to the present image by highlighting the differences between the analysis results of the current image and the analysis results of past images. Specifically, it can clearly indicate the differences by changing the color, shape, etc., of the annotations in the differing areas, or by adding text, marks, etc., to the differing areas.

[0119] Furthermore, the control unit 31 displays only the parts where there are differences between the analysis results of the current image and the analysis results of the past image, thereby directing the user's attention only to the areas where changes have occurred from the past image to the current image.

[0120] In this way, by highlighting the differences between the analysis results of current images and past images, or by displaying only the parts with differences, it is possible to make areas that have not been checked before stand out. This allows users to efficiently recognize analysis results that deserve attention.

[0121] Furthermore, if a candidate lesion that was not detected in past images is detected in the current image, if a candidate lesion that was detected in past images is not detected in the current image, or if the size or extent of the candidate lesion detected in past images differs by a predetermined amount or more, then it can be determined that there is a difference between the analysis results of the current image and the analysis results of past images.

[0122] The above-described embodiments are examples of the program, information processing apparatus, information processing method, and information processing system according to the present invention, and are not limited thereto. The detailed configuration and detailed operation of each device constituting the system can also be modified as appropriate without departing from the spirit of the present invention.

[0123] For example, in step S17 of the analysis result output processing (see Figure 7), each output example A to I may be combined and implemented. Alternatively, each output example A to I may be pre-associated with each modality 20, and the output example associated with the modality 20 from which the medical image to be processed was generated may be applied. Alternatively, while a medical image is displayed on the display unit 33 of the information processing device 30, the user may switch between the applicable output examples A to I each time they press a predetermined button on the operation unit 32.

[0124] Furthermore, depending on the intended use of displaying medical images and analysis results, output examples A to I, which are applied when there is a difference between the first and second analysis results, may be modified. For example, for normal use, the shape of annotation 76 may be changed and displayed as in output example F (see Figure 15); for use in conferences, mark 339A may be added to the differing parts as in output example H (see Figure 17); and for use in reports, only annotation 340A of the differing parts may be displayed as in output example I (see Figure 18).

[0125] Furthermore, in each output example A to I, if the display of analysis results is unnecessary, the analysis results may be automatically hidden. Examples of situations where the display of analysis results is unnecessary include when annotations overlap with measurement lines such as the cardiothoracic ratio, or during CINE (video) playback.

[0126] Furthermore, in the analysis result comparison process (see Figure 8), in addition to determining whether the candidate lesions in the analysis results of the current image and the analysis results of the past image "match" or "mismatch" based on the degree of overlap of the lesion regions, it is also possible to distinguish between three or more levels, such as "match," "suspected mismatch," and "mismatch." Additionally, the display of analysis results may be differentiated by changing the color of the annotation corresponding to the candidate lesion (for example, "red" for "mismatch" and "blue" for "suspected mismatch").

[0127] Furthermore, in the above embodiment, the control unit 31 of the information processing device 30 functions as the first acquisition means, the second acquisition means, and the output means, but devices other than the information processing device 30 may also be equipped with each of these means. Furthermore, in the above embodiment, the medical image management server 10 is equipped with a CAD function (first generation means, second generation means) that performs computer processing on medical images and generates analysis results. However, a device other than the medical image management server 10 may also be equipped with the CAD function.

[0128] Furthermore, although the above embodiment illustrates an example where a chest image is used as the medical image, the imaging site is not limited to the chest. In addition, the medical image to be processed may be multiple slice images obtained by CT, multi-frame images (video), or images obtained by 3D conversion of images obtained by CT or MRI.

[0129] Furthermore, in the above embodiment, the medical image management server 10 stores the medical image data and the analysis result data in separate areas (medical image storage area 143, analysis result storage area 144). However, the analysis result data may also be incorporated into the medical image file containing the medical image data.

[0130] Furthermore, in the above embodiment, as an example of outputting a different output when there is a difference between the first and second analysis results compared to when there is no difference, the case of changing the display on the display screen was mainly described. However, it is also possible to output display data to display a different output when there is a difference between the first and second analysis results compared to when there is no difference. In addition, it is also possible to output print data to print a different output when there is a difference between the first and second analysis results compared to when there is no difference.

[0131] Furthermore, since both the first and second analysis results are generated uniformly according to the same rules by computer processing in the image analysis unit 13 of the medical image management server 10, it is easy to detect differences between the first and second analysis results. If a physician were to detect potential lesions from medical images and manually add annotations to obtain analysis results, it would be difficult to add annotations to both images at the same location and size, even if the lesions in the first and second medical images were identical. Therefore, it would be difficult to detect true differences between manually generated analysis results, or between manually generated analysis results and computer-processed analysis results. Therefore, when comparing the results of the first and second analyses, it is important to use the results obtained through computer processing.

[0132] Furthermore, the programs for executing each process in each device may be stored on a portable recording medium. Alternatively, a carrier wave may be used as the medium for providing the program data via a communication line. [Explanation of symbols]

[0133] 10 Medical Image Management Server 11 Control Unit 12 Communications Department 13 Image Analysis Department 14 Storage section 20 Modalities 30 Information Processing Devices 31 Control Unit 32 Operation section 33 Display section 34 Communications Department 35 Storage section 100 Medical Image Display Systems 141 User Management Table 142 Data Management Tables 143 Medical Image Storage Area 144 Analysis result storage area 351 Analysis Result Output Processing Program N Communication Network

Claims

1. On the computer, A first acquisition function that obtains a first analysis result obtained by computer processing of a patient's first medical image, A second acquisition function for acquiring a second analysis result obtained by computer processing of the second medical image of the patient, The first analysis result and the second analysis result are compared, and if there is a difference, the output is output in a different manner than when there is no difference. A program to achieve this.

2. The program according to claim 1, wherein the output function outputs information indicating that there is a difference when there is a difference, thereby differentiating the output from the case when there is no difference.

3. The program according to claim 2, wherein the output function notifies the user of information indicating that there is a difference.

4. The program according to any one of claims 1 to 3, wherein the output function changes the display mode of the mouse cursor when there is a difference, thereby outputting a different output mode than when there is no difference.

5. The program according to any one of claims 1 to 4, wherein the output function changes the display mode of the window when there is a difference, thereby outputting a different output mode than when there is no difference.

6. The program according to any one of claims 1 to 5, wherein the output function highlights the portion where there is a difference between the first analysis result and the second analysis result, thereby outputting a different output pattern than when there is no difference.

7. The program according to claim 6, wherein the highlighting is displayed by changing or adding at least one of the colors, shapes, characters, and marks of the portion that differs between the first analysis result and the second analysis result.

8. The program according to any one of claims 1 to 5, wherein the output function displays only the portion where there is a difference between the first analysis result and the second analysis result, thereby outputting in a different manner than when there is no difference.

9. The program according to any one of claims 1 to 8, wherein the first medical image is an image taken at a different date and time than the second medical image.

10. The computer processing includes a process for detecting candidate lesions from the first medical image or the second medical image. The program according to any one of claims 1 to 9, wherein the difference between the first analysis result and the second analysis result includes the fact that a candidate lesion detected from the first medical image is not detected from the second medical image, or that a candidate lesion not detected from the first medical image is detected from the second medical image.

11. The computer processing includes a process for detecting candidate lesions from the first medical image or the second medical image. The program according to any one of claims 1 to 9, wherein the difference between the first analysis result and the second analysis result includes a difference in size or extent of a predetermined amount or more between the lesion candidate detected from the first medical image and the lesion candidate detected from the second medical image.

12. A first acquisition means for acquiring a first analysis result obtained by computer processing of a first medical image of a patient, A second acquisition means for acquiring a second analysis result obtained by computer processing of the second medical image of the patient, An output means that compares the first analysis result and the second analysis result and outputs the results in a different manner than when there is no difference, An information processing device equipped with the following features.

13. A first acquisition step of obtaining a first analysis result obtained by computer processing of a first medical image of a patient, A second acquisition step of obtaining a second analysis result obtained by computer processing of the second medical image of the patient, An output process is performed which, if there is a difference between the first analysis result and the second analysis result, the output is output in a different manner than when there is no difference. Information processing methods including

14. A first generation means that performs computer processing on a first medical image of a patient and generates a first analysis result, A second generation means that performs computer processing on the second medical image of the patient and generates a second analysis result, A first acquisition means for acquiring the first analysis result, A second acquisition means for obtaining the second analysis result, An output means that compares the first analysis result and the second analysis result and outputs the results in a different manner than when there is no difference, An information processing system equipped with the following features.

Images