Case search device, case search method, and program

Abstract

To provide a case search device, a case search method and a program that provide information that accurately matches past cases and is useful for creation of the next treatment plan.SOLUTION: Provided is a case search device 100 comprising: an acquisition function 101 for acquiring information regarding a target block of a present case and a change tendency during multiple instances of follow-up recording of feature parameters of the target block; a search function 102 for searching history data for a case that has appropriately the same block structure as the target block of the present case and has appropriately the same change tendency as the change tendency of feature parameters of the present case; and a display control function 103 for causing a display unit to display the result of having compared the searched cases with the present case and a treatment report on each of the searched cases.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The embodiments disclosed in the present specification and drawings relate to a case search device, a case search method, and a program. [Background technology]

[0002] In recent years, with the advancement of information technology in hospitals and other medical institutions, an increasing number of medical institutions have introduced electronic medical records (EMRs) to manage patients' treatment status. EMRs record patients' personal information, diagnostic history, images of affected areas, and other information, making it easy to share and reuse data. Furthermore, with the increasing capacity of storage devices, it is now possible to take large volumes of digital images from medical imaging devices such as CT (Computed Tomography) and MRI (Magnetic Resonance Imaging) devices and then store them in storage devices.

[0003] In the operation of electronic medical records, it is sometimes necessary to search for cases with similar follow-up descriptions and examination images from stored past history data to serve as reference for the diagnosis of a current case (hereinafter referred to as the current case). Conventional search methods often set search parameters to parameters related to affected area information or a specific treatment plan, and search for past cases with similar geometric features to the affected area image of the current case, for example. However, cases found in this way may not be consistent with the current case and may have little reference value for the treatment plan.

[0004] Therefore, conventional methods only support simple screening by medical professionals and are unable to provide practical information for selecting the next treatment plan. Therefore, there is a need for a case search device that can accurately match past cases according to the changing trends of the characteristic parameters of the affected area and provide information useful for creating the next treatment plan. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] Japanese Patent Application Laid-Open No. 2015-203920 Summary of the Invention

[0006] One of the problems that the embodiments disclosed in this specification and drawings aim to solve is to accurately match past cases and provide information useful for creating the next treatment plan. However, the problems solved by the embodiments disclosed in this specification and drawings are not limited to the above problem. Problems corresponding to the effects of each configuration shown in the embodiments described below can also be positioned as other problems. [Means for solving the problem]

[0007] A case search device according to an embodiment includes an acquisition unit, a search unit, and a display control unit. The acquisition unit acquires information about a target block of a current case and a change trend of feature parameters of the target block during multiple follow-up records. The search unit searches historical data for cases that have substantially the same block structure as the target block of the current case and substantially the same change trend as the change trend of the feature parameters of the current case. The display control unit displays, on a display unit, a comparison result between the searched case and the current case and treatment reports for each of the searched cases. [Brief explanation of the drawings]

[0008] [Figure 1] FIG. 1 is a diagram illustrating an example of the configuration of a case retrieval apparatus according to the first embodiment. [Figure 2A] FIG. 2A is a flowchart showing the processing procedure of the case retrieval apparatus according to the first embodiment. [Figure 2B] FIG. 2B is a diagram showing an example of the treatment report displayed in step S300 of FIG. 2A. [Figure 3A] FIG. 3A is a flowchart showing the detailed processing procedure in step S100 of FIG. 2A. [Figure 3B]FIG. 3B is a flowchart showing the detailed processing procedure in step S102 of FIG. 3A. [Figure 4] FIG. 4 is a schematic diagram showing the obtained PET image. [Figure 5] FIG. 5 is a schematic diagram showing how a block structure in a CT image is identified based on a PET image. [Figure 6] Figure 6 is a schematic diagram showing the registration of CT images during the follow-up recording. [Figure 7] FIG. 7 is a diagram showing the change tendency of the feature parameter when the SUV is used as the feature parameter. [Figure 8A] FIG. 8A is a diagram schematically showing patient information stored in a storage circuit. [Figure 8B] FIG. 8B is a diagram schematically showing patient information stored in the memory circuitry. [Figure 9] FIG. 9 is a flowchart showing the detailed processing procedure in step S200 of FIG. 2A. [Figure 10] FIG. 10 is a diagram showing a procedure for searching for similarity of block structures. [Figure 11] FIG. 11 is a schematic diagram showing that some cases were excluded from all searched cases. [Figure 12] FIG. 12 is a diagram showing an example of a search performed by the case search apparatus according to the first embodiment. [Figure 13] FIG. 13 is a diagram showing an example of a search result displayed on a display. [Figure 14] FIG. 14 is a diagram showing a method for comparing similarities when the trend line is a curve in the second embodiment. [Figure 15] FIG. 15 is a diagram showing search results when the trend line is a curve. [Figure 16] FIG. 16 is a diagram illustrating an example of the configuration of a case retrieval apparatus according to the third embodiment. [Figure 17]FIG. 17 is a diagram showing an example in which one candidate block is selected as a target block from among a plurality of candidate blocks by the selection function of the case retrieval apparatus according to the third embodiment. [Figure 18] FIG. 18 is a diagram illustrating a modified example of the selection function of the case retrieval apparatus according to the third embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0009] Hereinafter, a case retrieval device, a case retrieval method, and a program according to embodiments will be described with reference to the drawings. In the following embodiments, a PET-CT device will be used as an example for acquiring PET (Positron Emission Computed Tomography)-CT image data. However, the embodiments are not limited to this. Separate PET and CT devices may be used to collect PET image data and CT image data, respectively, and registration may be performed to determine the position of a block structure (e.g., a tumor). In this embodiment, data obtained by scan analysis of a subject is referred to as a case. Specifically, data obtained by analysis processing of a current subject is referred to as a current case, and data obtained by analysis processing of subjects other than the current subject are referred to as other cases.

[0010] (First embodiment) Fig. 1 is a diagram showing an example of the configuration of a case retrieval device 100 according to the first embodiment. As shown in Fig. 1, the case retrieval device 100 according to the first embodiment includes a processing circuitry 110, a storage circuitry 104, an input interface 120, and a display 130. The case retrieval device 100 receives PET-CT image data from a PET-CT device 300 as input data and analyzes the received PET-CT image data.

[0011] The input interface 120 is realized by a trackball for performing various settings, a switch button, a mouse, a keyboard, a touchpad for performing input operations by touching the operation surface, a touchscreen in which the display screen and touchpad are integrated, a non-contact input circuit using an optical sensor, a voice input circuit, etc. The input interface 120 is connected to the processing circuitry 110 and converts input operations received from an operator into electrical signals and outputs the electrical signals to the processing circuitry 110. Note that although the input interface 120 is provided within the case retrieval device 100 in FIG. 1 , it may also be provided externally.

[0012] The display 130 is connected to the processing circuitry 110 and displays various types of information and image data output from the processing circuitry 110. For example, the display 130 is realized by a liquid crystal monitor, a CRT (Cathode Ray Tube) monitor, a touch panel, or the like. For example, the display 130 displays a GUI (Graphical User Interface) for receiving instructions from an operator, various display images, and various processing results by the processing circuitry 110. Note that although the display 130 is provided within the case retrieval device 100 in FIG. 1 , it may also be provided externally. Here, the display 130 is an example of a display unit.

[0013] The memory circuitry 104 is connected to the processing circuitry 110 and stores various data. For example, the memory circuitry 104 is realized by a semiconductor memory element such as a random access memory (RAM) or a flash memory, a hard disk, an optical disk, or the like. The memory circuitry 104 also stores programs corresponding to the processing functions executed by the processing circuitry 110. Note that although the memory circuitry 104 is provided within the case retrieval device 100 in FIG. 1 , it may also be provided externally.

[0014] The processing circuitry 110 controls each component of the case search device 100 in response to an input operation received from an operator via the input interface 120. For example, the processing circuitry 110 is realized by a processor. As shown in FIG. 1 , the processing circuitry 110 executes an acquisition function 101, a search function 102, and a display control function 103. Here, the acquisition function 101, the search function 102, and the display control function 103 are examples of an acquisition unit, a search unit, and a display control unit, respectively.

[0015] The term "processor" used in the above description refers to a circuit such as a central processing unit (CPU), a graphics processing unit (GPU), an application specific integrated circuit (ASIC), or a programmable logic device (e.g., a simple programmable logic device (SPLD), a complex programmable logic device (CPLD), or a field programmable gate array (FPGA)). If the processor is a CPU, for example, the processor realizes its function by reading and executing a program stored in the memory circuit 104. On the other hand, if the processor is an ASIC, for example, the program is directly embedded in the processor circuit instead of storing the program in the memory circuit 104. Note that each processor in this embodiment is not limited to being configured as a single circuit, but may be configured as a single processor by combining multiple independent circuits to realize its function. Furthermore, multiple components in FIG. 1 may be integrated into a single processor to realize its function.

[0016] Next, the processes of the acquisition function 101, search function 102, and display control function 103 executed by the processing circuit 110 will be described.

[0017] First, the PET-CT device 300 scans the subject and acquires PET-CT image data of the current subject, and then the acquisition function 101 receives the PET-CT image data from the PET-CT device 300 as input data, analyzes it, acquires information on the block structure of the current subject, and the change trends of the feature parameters of the block structure generated based on the information on the block structure, and stores these as part of the current case in the memory circuitry 104. The block structure is a pseudo-disease region inside the subject's body, and for convenience, a tumor may be used instead of a block structure in the following description, but the tumor is just an example and is not limited to a tumor.

[0018] Based on the block structure information acquired by the acquisition function 101, the search function 102 searches for past cases with block structures substantially identical to those of the current case, and excludes from the searched cases any cases with fewer than two follow-ups. The remaining cases are then searched for those with characteristic parameter change trends substantially identical to those of the current case. "Substantially identical" includes cases where the cases are identical, or similar but not completely identical. Hereinafter, for ease of explanation, the term "similar" will be used in place of the term "substantially identical." The change trend includes the initial value of the characteristic parameter at the start of the search period, the direction of change (increase or decrease) of the characteristic parameter value throughout the entire search period, and the magnitude of change. Note that if the number of follow-ups is more than two, there are at least three characteristic parameter values. In this case, the change trend also includes the frequency of change in the characteristic parameter value. When the trend line representing the change trend is a curve, the direction and magnitude of change are expressed in the form of the trend line. When the trend line representing the change trend is a straight line, the direction and magnitude of change are expressed by the slope and slope change of the trend line.

[0019] The display control function 103 controls the display 130 to display a GUI (Graphical User Interface) for the user to input various setting requests via the input interface 120, and to display search results including comparison results and treatment reports generated by the case search device 100. The comparison results include the results of comparison of image data between the current case and similar cases, and the results of comparison of change trends in feature parameters between the current case and similar cases. The treatment reports include, for example, a treatment report for the current stage of the current case and treatment reports for similar cases. The treatment reports may include only the treatment reports of the searched similar cases.

[0020] The memory circuitry 104 stores past history data, which includes, for each case, image data of a block structure that is a region of interest, values ​​of characteristic parameters at each follow-up, and treatment reports. Here, the memory circuitry 104 is an example of a memory unit.

[0021] FIG. 2A is a flowchart showing the processing procedure of the case retrieval apparatus 100 according to the first embodiment, and FIG. 2B is a diagram showing an example of a treatment report displayed in step S300.

[0022] 2A, first, in step S100, the acquisition function 101 acquires information about the target block of the current case and the change trends of the feature parameters, and then stores the information in the memory circuitry 104. The target block is selected in the current case by the case search device 100 and is the block to be the next search target.

[0023] Next, in step S200, the search function 102 searches for cases that are similar to the target block of the current case and have similar change trends in the feature parameters.

[0024] Then, in step S300, the display control function 103 controls the display 130 to display the comparison results between the searched similar cases and the current case, and the treatment reports for each similar case. As shown in FIG. 2B, the treatment report includes, for example, the patient's condition, treatment plan, treatment goals, patient survival time, side effects, etc. The treatment report shown in FIG. 2B is only an example, and the treatment report may include other information, but should include at least the following treatment plan:

[0025] Next, the process of step S100 in Fig. 2A will be described in detail. Fig. 3A shows the detailed process procedure of step S100 in Fig. 2A.

[0026] Step S100 in FIG. 2A includes steps S101 to S104 shown in FIG. 3A.

[0027] First, in step S101 of FIG. 3A, the acquisition function 101 acquires PET-CT image data generated by scanning with the PET-CT device 300. Then, in step S102, the acquisition function 101 analyzes the PET-CT image data to acquire image information of a target block structure (tumor) and feature parameters indicating certain characteristics of the block structure (tumor). The feature parameters include a standard uptake value (SUV), a metabolic tumor volume (MTV), a total lesional glycolysis (TLG), and a tumor / normal ratio (T / N ratio). The above feature parameters are merely examples, and the present embodiment is not limited to the above feature parameters.

[0028] Next, in step S103 of Fig. 3A, the acquisition function 101 generates a trend line showing the change tendency of the feature parameters using the values ​​of the feature parameters recorded at each follow-up of the current case stored in the memory circuitry 104. Then, in step S104 of Fig. 3A, the acquisition function 101 stores information about the target block structure and the trend line of the feature parameters in the memory circuitry 104 as the analysis results.

[0029] Here, the processing of step S102 in Fig. 3A will be described in detail. Fig. 3B shows the detailed processing procedure of step S102 in Fig. 3A. Step S102 in Fig. 3A includes steps S1021 to S1024 shown in Fig. 3B. Steps S1021 to S1024 will be described below with reference to Figs. 4 to 7.

[0030] FIG. 4 shows PET image data 400 acquired from a PET-CT device 300. In the PET image data 400, SUV is shown in different gradations, with darker colors indicating higher SUV. Excluding specific regions 411-413, such as the brain, heart, and bladder, regions 421-423 with high SUV may represent tumors. Therefore, in this embodiment, in step S1021 of FIG. 3B, the acquisition function 101 extracts block structures in which the SUV of regions other than the specific regions is higher than the SUV threshold based on a predetermined SUV threshold, and classifies (identifies) multiple block structures in the PET image data as target blocks according to the organs to which the block structures belong. Here, SUV is used as a feature parameter as an example, but other indices may be selected as the feature parameter.

[0031] FIG. 5 is a schematic diagram showing the division of block structures 511-513 in CT image data based on block structures 501-503 in PET image data 400. A subject may have block structures in different organs, or the same organ may have multiple block structures. FIG. 5 shows that the PET image data 400 has multiple block structures 501-503. In step S1022 of FIG. 3B, the acquisition function 101 classifies (identifies) the multiple block structures 511-513 in the CT image data according to the organs to which the multiple block structures 501-503 in the PET image data 400 belong, and performs registration. Similarly, even when the subject has only one block structure, the acquisition function 101 classifies (identifies) the block structure in the CT image data according to the organ to which the block structure in the PET image data 400 belongs, and performs registration.

[0032] Next, in step S1023 of FIG. 3B, the acquisition function 101 performs registration between the PET-CT image data acquired in the current follow-up and the PET-CT image data (past images) from previous follow-ups stored in the memory circuitry 104. FIG. 6 is a schematic diagram showing the registration with the PET-CT image data from the previous follow-up. As shown in FIG. 6, assuming that the current follow-up is the third follow-up, the acquisition function 101 registers the CT image data 603 "follow-up #3" acquired in the third follow-up with the CT image data 601 "follow-up #1" from the first follow-up and the CT image data 602 "follow-up #2" from the second follow-up, thereby obtaining the change status of the identical block structure 610.

[0033] Next, in step S1024 of FIG. 3B, the acquisition function 101 identifies the change trend for each target block according to the positional relationship where registration was performed. Specifically, the acquisition function 101 generates a graph showing the change trend of SUV based on the SUV obtained from the PET image data 400. FIG. 7 shows two different types of graphs 700 based on the feature parameter SUV. In one, a trend line 710 showing the change trend of the feature parameter is formed by connecting points indicating the value of the feature parameter during each follow-up recording in an XY coordinate system with SUV as the Y axis and the follow-up interval time as the X axis. In the other, a trend line 720 showing the change trend of the feature parameter is formed by connecting points indicating the value of the feature parameter during each follow-up recording in an XY coordinate system with SUV as the Y axis and the size of the target block per unit volume as the X axis. Although two types of trend line generation methods have been illustrated here, the present invention is not limited to these.

[0034] Here, when the above-mentioned SUV is SUV based on body weight (SUVbw, SUV calculated using body weight), the unit is g / ml. When the above-mentioned SUV is SUV based on body surface area (SUVbsa, SUV calculated using body surface area), the unit is cm. 2 / ml. If the SUV mentioned above is based on lean body mass (SUVlbm, SUV calculated using lean body mass), the unit is g / ml. For convenience of illustration, coordinates are omitted in some figures, but when searching, similar cases must be searched for using the same coordinate system as the current case.

[0035] After the acquisition function 101 acquires information about the block structure and the trend line of the feature parameters, the information is stored as history data in the memory circuitry 104 and updated. FIGS. 8A and 8B schematically show the subject information stored in the memory circuitry 104. The memory circuitry 104 stores subject information, such as personal information, follow-up information, block structure (tumor), and treatment report, for each patient. In the example shown in FIG. 8A, subject information 800 about "Patient 1" includes "Follow-up Record #1" and "Follow-up Record #2" as follow-up record information, and each of "Follow-up Record #1" and "Follow-up Record #2" includes tumor information, such as "Tumor #1," "Tumor #2," and "Tumor #3." In the example shown in FIG. 8B, association information 810 associating "Follow-up Record #1" and "Tumor #1" in the subject information 800 about "Patient 1" includes, for example, personal information such as the patient's name "Zhang San," age "65," height "158 cm," weight "56 kg," and follow-up record information. The related information 810 also includes, as follow-up records, information such as the follow-up record number "#1," the follow-up time (date) "December 3, 2019," the scan mode "PET-CT," and the manufacturer of the scanner used. The related information 810 also includes, as tumor information, information such as the tumor number "#1," characteristic parameter values ​​(SUV, MTV, TLG, T / N ratio), the location, and the texture of the tumor, and as a treatment report, information such as the patient's condition, treatment plan, and treatment objectives. Although not shown in the table of FIG. 8B, the tumor information may also include image information of the tumor.

[0036] In this embodiment, the case where the values ​​of the feature parameters and the trend lines of the feature parameters are stored and then searched has been described. However, the values ​​of the feature parameters may be stored first, and the trend lines of the feature parameters may be generated during the search described below.

[0037] After the acquisition function 101 acquires information about the target block of the current case, the search function 102 executes step S200 in Fig. 2A. Step S200 includes steps S201 to S204 shown in Fig. 9. Steps S201 to S204 will be described below with reference to Figs. 9 to 13.

[0038] First, in step S201 of Fig. 9, the search function 102 acquires other cases having block structures similar to the target block. Each block structure is characterized by, for example, a texture feature ft based on thickness, density, etc., a gradation feature fg based on average value, variance, etc., and a shape feature fs based on position, size, etc. The texture feature ft, gradation feature fg, and shape feature fs are merely examples and are not limited to these.

[0039] For example, the search function 102 calculates the similarity C between the searched case and the current case using the following formula:

[0040]

number

[0041] where Xi represents the feature weights and C low and C up 10, when the current case is compared with other cases "Case #1," "Case #2," "Case #3," and "Case #4" stored in the memory circuitry 104, it is found that "Case #1" and "Case #4" have a block structure similar to that of the current case, and therefore the search function 102 acquires "Case #1" and "Case #4" as the search results 1000.

[0042] Next, in step S202 of FIG. 9 , the search function 102 excludes cases with insufficient follow-up records, in other words, cases with insufficient follow-up times or cases lacking information on characteristic parameters, from the search results. FIG. 11 is a schematic diagram showing the exclusion of some cases from all searched cases. FIG. 11 illustrates an example in which cases with fewer than two follow-up times have been excluded. Specifically, the memory circuitry 104 stores subject information 1101 to 1103 for "Patient 1," including personal information, follow-up information, block structure, and treatment report. In FIG. 11 , subject information 1101 includes two sets of follow-up information: PET and CT images from the first follow-up, and PET and CT images from the second follow-up. In FIG. 11 , subject information 1102 and 1103 include only one set of follow-up information: PET and CT images from the first follow-up. In this case, the subject information 1102 and 1103 are cases with less than two follow-ups, so the search function 102 excludes the subject information 1102 and 1103 from the search results as cases with insufficient follow-up records. Note that the search results may be configured to exclude cases with fewer follow-ups than the current case.

[0043] Here, in step S202 of FIG. 9, the search function 102 excludes cases with insufficient follow-up records from the search results for the following reasons: If the number of follow-ups is less than two, a trend line showing the change trend of the feature parameters cannot be generated, making it impossible to compare the change trends; if the number of follow-ups is less than the number of follow-ups of the current case, even if a trend line showing the change trend of the feature parameters can be generated, it cannot perfectly match the trend line of the current case; and if information on the feature parameters is missing, a trend line showing the change trend of the feature parameters accurately cannot be generated.

[0044] However, step S202 may be omitted as necessary. For example, in step S203 described below, when the search function 102 selects subject information of a case with two or more follow-ups, step S202 may be omitted.

[0045] Next, in step S203 of Fig. 9, the search function 102 compares trend lines, which represent the change trends of the feature parameters of the current case and other cases. Fig. 12 shows an example of a search performed by the case search device 100 of the first embodiment. The graph on the left side of Fig. 12 shows a trend line 1200 of the feature parameters of the current case, and the four graphs on the right side of Fig. 12 show trend lines 1201 to 1204 of the feature parameters of the searched similar cases.

[0046] In this embodiment, the trend line is a line formed by connecting points indicating the values ​​of feature parameters in each follow-up record. Depending on the number of follow-ups, the trend line may be composed of multiple line segments with different slopes. Hereinafter, "similar trend lines" is defined as the initial values ​​(starting values) of feature parameters (e.g., SUV) and the shape of the trend lines are all similar. Therefore, the search function 102 uses the initial values ​​of feature parameters (e.g., SUV) and the shape of the trend line as search conditions, searches for trend lines of other cases stored in the memory circuitry 104 based on these two search conditions, and searches whether the trend lines of the other cases contain line segments similar to the trend line of the current case. The retrieved trend line may be similar in its entirety to the trend line 1200 of the current case, or, as shown in FIG. 12, only some line segments may be similar to the trend line 1200 of the current case. For example, at follow-up intervals of 10 to 20 days, the line segments of trend lines 1201 to 1204 of similar cases are similar to the line segment of trend line 1200 of the current case. Furthermore, of the trend lines of the searched similar cases, it is preferable to highlight the line segments similar to trend line 1200 of the current case and display them on display 130, but this is not limitative.

[0047] This allows the search function 102 to search for trend lines of cases in which at least some of the line segments are similar to the trend line of the current case, thereby making it possible to search for other past cases more comprehensively without omissions.

[0048] 12 shows that the current case has only two follow-up records, this embodiment can also be applied to cases where the current case has two or more follow-up records. If the number of follow-up records exceeds two, the initial values ​​of the feature parameters and the slopes of multiple line segments must be used as search conditions.

[0049] As described above, in steps S201 to S203 of FIG. 9, the search function 102 searches for cases that have a block structure similar to that of the target block of the current case and that have a change trend similar to that of the feature parameters of the current case as similar cases.

[0050] 9, the search function 102 generates search results for similar cases similar to the current case. The search results may include a comparison result of the degree of similarity between the current case and the similar case and a treatment report of the similar case. The comparison result may show only trend lines of the characteristic parameters of the current case and the similar case, or may show trend lines of the characteristic parameters of the current case and the similar case and images of the affected areas of each case simultaneously. The treatment report of the similar case includes the following treatment plan.

[0051] Furthermore, if only the values ​​of the feature parameters are stored in the memory circuitry 104 and no trend lines of the feature parameters are stored, step S200 in Fig. 2A may further include a step of generating trend lines of the feature parameters between steps S202 and S203 in Fig. 9. This step of generating trend lines may be performed before step S203, which is the step of comparing the change trends, and the specific timing is not limited.

[0052] Then, in step S300 of FIG. 2A, the display control function 103 causes the display 130 to display the search results.

[0053] 13 shows an example of a search result displayed on the display 130. In FIG. 13, images of the affected areas of the patient's current case and similar cases (the "current" PET image and CT image, the "similar 1" PET image and CT image, and the "similar 2" PET image and CT image), trend lines of the characteristic parameters of the current case and similar cases (trend line 1200 of the current case and trend lines 1201 and 1202 of the similar cases), and treatment reports of the current case and similar cases are displayed. Regarding the treatment reports, the treatment reports of similar cases are treatment reports recorded in the last follow-up record for trend lines similar to the trend line of the current case.

[0054] Here, in the example shown in FIG. 13, the display control function 103 displays the trend line of the retrieved case on the display 130 without overlapping it with the trend line of the current case, but to make it easier for the user to see, the trend line of the retrieved case may be displayed on the display 130 overlapping the trend line of the current case.

[0055] As explained above, the case retrieval device 100 according to the first embodiment can accurately retrieve similar cases with similar disease progression by searching for past cases with similar block structures and similar change trends in feature parameters. Furthermore, the case retrieval device 100 according to the first embodiment can simultaneously display trend lines of feature parameters for the current case and similar cases, as well as the next treatment plan for the similar case, on the display 130, allowing the user to intuitively select a case with a similar disease progression and easily create the next treatment plan.

[0056] In the above embodiment, the trend graphs of the characteristic parameters of the current case and similar cases, and the next treatment plan for the similar case are simultaneously displayed, but the display format is not limited. Below, as Modifications 1 to 4, an example in which the comparison results are displayed according to a certain priority will be described.

[0057] (Variation 1) First, Variation 1 will be described. The trend lines of the retrieved similar cases have two patterns: one where a follow-up record similar to the trend line of the current case is followed by a subsequent follow-up record, and one where no subsequent follow-up record is present. When a subsequent follow-up record is present, the effectiveness of the previously created treatment plan can be verified based on the continuous changes in the trend line shape. Specifically, as shown in FIG. 12 , an example is given in which the line segments of trend lines 1201 to 1204 of the retrieved four similar cases are similar to the line segment of trend line 1200 of the current case at follow-up intervals of 10 to 20 days. Since each of the trend lines 1201 to 1204 includes a follow-up record similar to the line segment of trend line 1200 of the current case followed by a subsequent follow-up record, the user can obtain more useful information than when there is no subsequent follow-up record.

[0058] Therefore, in this modified example, the display control function 103 ranks the priority of the searched similar cases depending on whether or not there is a follow-up record after the follow-up record that is similar to the trend line of the current case, and displays similar cases with subsequent follow-up records preferentially on the display 130.

[0059] 12, among the trend lines 1201 to 1204 of the four similar cases with subsequent follow-up records, the trend line 1204 of the similar case in the lower right of FIG. 12 has the highest degree of decrease in SUV, which is the value of the characteristic parameter of the subsequent follow-up records, and therefore its treatment plan is the most effective. Therefore, if the priority of this similar case is set high, the user can preferentially refer to the treatment plan created in the second follow-up record of this similar case.

[0060] Therefore, in this modified example, the display control function 103 further ranks the priority of the searched similar cases according to the degree of decline in SUV of the follow-up records subsequent to the follow-up record similar to the trend line of the current case, and can preferentially display similar cases with more effective treatment plans on the display 130.

[0061] (Variation 2) Next, a description will be given of Modification 2. For example, the display control function 103 derives the similarity between the trend lines of the searched similar cases and the trend line of the current case, prioritizes the searched similar cases according to the derived similarity, and preferentially displays on the display 130 similar cases that have a higher similarity to the trend line of the current case. For example, the display control function 103 preferentially displays on the display 130 treatment reports of similar cases that have a higher similarity.

[0062] As a result, in the second modification, the user can more easily create the next treatment plan by referring to the treatment plan in the treatment report of a case with a high degree of similarity.

[0063] (Variation 3) Next, a third modification will be described. For example, the display control function 103 derives the similarity between the searched similar cases and the block structure (tumor) of the target block of the current case, prioritizes the searched similar cases according to the derived similarity, and preferentially displays on the display 130 similar cases that have a higher similarity to the block structure of the tumor of the current case. For example, the display control function 103 preferentially displays on the display 130 treatment reports of similar cases that have a higher similarity.

[0064] As a result, in the third modification, the user can more easily create the next treatment plan by referring to the treatment plan in the treatment report of a case with a high degree of similarity.

[0065] (Variation 4) Next, we will explain Modification 4. In Modification 4, a threshold value may be set for the similarity between the current case and the searched case, and after the similarity is derived in Modifications 2 and 3, the display control function 103 may display on the display 130 only search results for which the derived similarity exceeds the threshold value.

[0066] (Second embodiment) In the first embodiment and its modified example, the search function 102 directly connects the values ​​of the characteristic parameters in each follow-up record to derive the slope of each segment of the trend line, and the display control function 103 displays the derived results on the display 130 as search results. However, in the second embodiment, the values ​​of the characteristic parameters may be fitted to a curve, and similar curves may be searched for for the fitted curve.

[0067] 14 shows a method for comparing similarities when trend lines 1401 and 1402 in the second embodiment are curved lines. In FIG. 14, trend line 1401 indicates the trend line of the current case, and trend line 1402 indicates the trend line of another case stored in the memory circuitry 104.

[0068] First, for the current case and other cases stored in the memory circuit 104, the search function 102 uses the values ​​of the feature parameters recorded in each follow-up record to generate a curve showing the change trend of the feature parameters after fitting.

[0069] The retrieved other cases only need to be locally similar to the change trend of the current case, so it is necessary to calculate the similarity between each point on the curve of the other cases and each point on the curve of the current case.

[0070] Therefore, the search function 102 searches each feature point on the trend line 1401 of the current case as [b1, b2, ... b m ], and each feature point on the trend line 1402 of the other cases is [a1, a2, ... a n ] and then compares the similarity between each feature point of the other cases and each feature point of the current case to determine the similarity section with the current case.

[0071] The similarity d of each feature point of another case can be calculated using the following formula:

[0072]

number

[0073]

number

[0074]

number

[0075] Thereafter, the search function 102 compares the calculated similarity d of each feature point with a preset threshold value, and selects the curve with the highest similarity, thereby identifying a similar section to the current case.

[0076] As a result, in the second embodiment, the search method can be further optimized, and other cases can be searched for with higher accuracy.

[0077] Fig. 15 shows search results when the trend line is a curve. In the example shown in Fig. 15, the display control function 103 displays the results derived by the search function 102 as search results on the display 130, superimposing the fitted trend line of the current case on the trend line of the similar case, thereby allowing the user to more intuitively sense the similarity between the two. The display control function 103 may also display the fitted trend line of the similar case on the display 130 superimposed on the fitted trend line of the current case.

[0078] (Third embodiment) The third embodiment differs from the first and second embodiments in that it further includes a selection function that allows one or more block structures to be selected from a plurality of block structures (tumors).

[0079] FIG. 16 is a diagram showing an example of the configuration of a case retrieval device 100 according to the third embodiment. The case retrieval device 100 according to the third embodiment further includes a selection function 105 in addition to the components of the case retrieval device 100 according to the first and second embodiments. That is, in the case retrieval device 100 according to the third embodiment, the processing circuitry 110 executes an acquisition function 101, a selection function 105, a search function 102, and a display control function 103, as shown in FIG. 16. Here, the selection function 105 is an example of a selection unit. When there are multiple block structures (tumors) acquired by the acquisition function 101, the selection function 105 accepts a selection operation from the user and selects a specific block structure from the multiple block structures (tumors).

[0080] In this embodiment, the selection function 105 sets multiple block structures obtained by scanning with the PET-CT device 300 as candidate blocks, and selects one candidate block from the multiple candidate blocks as a target block. Then, the search function 102 searches the history data for a case that is similar to the target block selected by the selection function 105 and has a similar change trend in the feature parameters of the current case.

[0081] 17 shows an example in which the selection function 105 of the case retrieval device 100 according to the third embodiment selects one candidate block from multiple candidate blocks as a target block. In step S100 of FIG. 2A, the acquisition function 101 first acquires tumor information about the tumor of the current case based on the scan results of the PET-CT device 300. For example, as shown in FIG. 17, if multiple block structures (tumors) exist in the lung region of the patient as multiple candidate blocks, the selection function 105 displays information about the multiple candidate blocks (tumors) as tumor information on the display 130.

[0082] 17, the selection function 105 displays, as tumor information relating to a tumor in the patient's lungs, image information (PET images and CT images at the time of this follow-up) of the tumor in the patient's current case and change trend information for each of candidate blocks 1701 "lung (affected area #1)," 1702 "lung (affected area #2)," 1703 "lung (affected area #3)," and 1704 "lung (affected area #4)" on the display 130. For example, as change trend information, trend lines 1711 to 1714 of the characteristic parameters of the current case are displayed on the display 130 for each of candidate blocks 1701 "lung (affected area #1)," 1702 "lung (affected area #2)," 1703 "lung (affected area #3)," and 1704 "lung (affected area #4)."

[0083] Next, one block structure from among a plurality of candidate blocks is selected as a target block in response to a user's input operation on the input interface 120. In the example shown in Fig. 17, of candidate block 1701 "lung (affected area #1)," candidate block 1702 "lung (affected area #2)," candidate block 1703 "lung (affected area #3)," and candidate block 1704 "lung (affected area #4)," candidate block 1701 "lung (affected area #1)" is selected as the target block. Then, in step S200 of Fig. 2A, the search function 102 searches the history data for cases that are similar to the target block selected by the selection function 105 and have a similar change trend in the feature parameters of the current case.

[0084] In addition, while FIG. 17 shows that image information and change trend information of each tumor are simultaneously displayed on the display 130, the selection function 105 may first display only the image information of each tumor, select one block structure from multiple block structures (tumors), and then the search function 102 may generate a trend line for the selected block structure.

[0085] As explained above, in the third embodiment, when there are multiple block structures acquired by the acquisition function 101 as candidate blocks, the selection function 105 selects one candidate block from the multiple candidate blocks of the current case as a target block in response to a user's input operation (selection operation) on the input interface 120. This allows the search function 102 to search for cases in the history data that are similar to the selected target block and have similar changing trends in the feature parameters of the current case, making it possible to accurately search for other tumors that are similar to a specific tumor, as necessary.

[0086] (Variation) The selection function 105 may select one or more candidate blocks from among a plurality of candidate blocks as the target block, or may select a part of trend lines from among the overall trend lines as the target trend line.

[0087] FIG. 18 is a diagram for explaining a modified example of the selection function 105 of the case retrieval apparatus 100 according to the third embodiment.

[0088] In the first and second embodiments, we have described searching for trend lines similar to the overall trend line of the current case (the change trends of feature parameters during all follow-up records) based on the overall trend line. However, a user may focus on only a portion of these trend lines, and if a similarity search is performed on the overall trend line, the similarity may be calculated without distinguishing between the weights of the focused and unfocused parts, which may result in the similarity of the focused part of the searched case being low. Therefore, it is desirable for the user to select the part they are more interested in and perform a search on only the trend lines of this part.

[0089] As an example, it is assumed that the treatment time for the current case is long, there is a significant effect in the early stage, but a bottleneck occurs in the later stage, and the change in the characteristic parameters becomes gradual, in which case it is necessary to consider whether or not to adjust the treatment plan.

[0090] In this case, it is desirable to search for cases with similar bottleneck periods, and in particular, it is desirable to search for cases in which the characteristic parameters have decreased to some extent after the bottleneck period, as in variant 1 of the first embodiment, which is of greater significance for adjusting the treatment plan.

[0091] 18, the selection function 105 displays, as tumor information related to a tumor in the patient's lungs, image information (PET images and CT images at the time of this follow-up) of the tumor in the patient's current case and change trend information for each of the candidate blocks 1701 "lung (affected area #1)," 1702 "lung (affected area #2)," 1703 "lung (affected area #3)," and 1704 "lung (affected area #4)" on the display 130. For example, as change trend information, trend lines 1711 to 1714 of the characteristic parameters of the current case are displayed on the display 130 for each of the candidate blocks 1701 "lung (affected area #1)," 1702 "lung (affected area #2)," 1703 "lung (affected area #3)," and 1704 "lung (affected area #4)."

[0092] Next, some of the trend lines 1711 to 1714 for each of the plurality of candidate blocks are selected as target trend lines in response to a user's input operation on the input interface 120. In the example shown in Fig. 18, at least a portion of the trend line 1711 among the trend lines 1711 to 1714 is selected as the target trend line 1800. Then, in step S200 of Fig. 2A, the search function 102 searches the history data for cases similar to the target trend line 1800 selected by the selection function 105.

[0093] Thus, according to the modified example of the third embodiment, by selecting a part of the trend line of the current case, the user can flexibly select the stage of interest, which is more meaningful in creating a treatment plan for a specific case.

[0094] (Other embodiments) In addition to the above-described embodiments, different components in each embodiment can be added or removed, or they can be combined to obtain new embodiments. These new embodiments are also included in the gist of the present invention. For example, in the above-described embodiments, a PET-CT device is used as an example for acquiring PET-CT image data, but the embodiments are not limited to this, and MRI may be used instead of CT.

[0095] The case search device 100 may be an electronic device (processor) that realizes the above-mentioned case search function. The case search device 100 according to the first and second embodiments executes an acquisition function, a storage function, a search function, and a display control function. The case search device 100 according to the third embodiment executes an acquisition function, a storage function, a selection function, a search function, and a display control function. The above functions may be stored in a storage circuit inside or outside the computer in the form of a computer-executable program. The computer reads and executes each program to realize the function corresponding to each program (acquisition function, storage function, selection function, search function, and display control function).

[0096] Note that the components of each device illustrated in the above-described embodiments are functional concepts and do not necessarily have to be physically configured as illustrated. In other words, the specific form of distribution and integration of each device is not limited to that illustrated, and all or part of them can be functionally or physically distributed and integrated in any unit depending on the actual situation. Furthermore, the names and combinations of information shown in the above drawings are merely examples and can be changed as needed.

[0097] The case search method described in the above embodiment and modified examples can be realized by executing a prepared program on a computer such as a personal computer or a workstation. This program can be distributed via a network such as the Internet. The program for realizing the case search method according to this embodiment can also be recorded on a non-transitory computer-readable recording medium such as a hard disk, flexible disk (FD), CD-ROM, MO, or DVD, and can be read from the recording medium and executed by a computer.

[0098] According to at least one of the embodiments described above, accurate matching with past cases can be performed, and useful information can be provided for creating the next treatment plan.

[0099] Although several embodiments have been described, these embodiments are presented as examples and are not intended to limit the scope of the invention. These embodiments can be implemented in various other forms, and various omissions, substitutions, and modifications can be made without departing from the spirit of the invention. These embodiments and their modifications are included within the scope and spirit of the invention, as well as within the scope of the invention and its equivalents as defined in the claims. [Explanation of symbols]

[0100] 100 Case Search Device 101 Acquisition Function 102 Search function 103 Display control function

Claims

1. An acquisition unit that acquires information about a target block of a current case and a change trend of a feature parameter of the target block; a search unit that searches, in the history data, for a case that has substantially the same block structure as the target block of the current case and has substantially the same change trend as the change trend of the feature parameter of the current case; a display control unit that displays on a display unit a comparison result between the searched case and the current case and a treatment report including at least a next treatment plan for each of the searched cases; Equipped with The change tendency of the characteristic parameter includes an initial value of the characteristic parameter at the start of a search period, and a change direction and a change width of the value of the characteristic parameter throughout the entire search period. Case search device.

2. An acquisition unit that acquires information about a target block of a current case and a change trend of a feature parameter of the target block; a search unit that searches, in the history data, for a case that has substantially the same block structure as the target block of the current case and has substantially the same change trend as the change trend of the feature parameter of the current case; a display control unit that displays on a display unit a comparison result between the searched case and the current case and a treatment report including at least a next treatment plan for each of the searched cases; Equipped with The change trend of the characteristic parameter is a trend line formed by connecting points indicating the values ​​of the characteristic parameter during each follow-up record in a coordinate system; The trend line connecting the points in the coordinate system indicates the change tendency of the characteristic parameter by a line connecting the points indicating the value of the characteristic parameter during each follow-up record in an XY coordinate system having the value of the characteristic parameter on the Y axis and the follow-up interval time or the size of the target block of unit volume on the X axis. Case search device.

3. An acquisition unit that acquires information about a target block of a current case and a change trend of a feature parameter of the target block; a search unit that searches, in the history data, for a case that has substantially the same block structure as the target block of the current case and has substantially the same change trend as the change trend of the feature parameter of the current case; a display control unit that displays on a display unit a comparison result between the searched case and the current case and a treatment report including at least a next treatment plan for each of the searched cases; Equipped with The change trend of the characteristic parameter is a trend line formed by connecting points indicating the values ​​of the characteristic parameter during each follow-up record in a coordinate system; at least some of the trend lines of the retrieved cases are substantially identical to the trend line of the current case; the substantially identical trendlines have substantially identical starting values; Case search device.

4. An acquisition unit that acquires information about a target block of a current case and a change trend of a feature parameter of the target block; a search unit that searches, in the history data, for a case that has substantially the same block structure as the target block of the current case and has substantially the same change trend as the change trend of the feature parameter of the current case; a display control unit that displays on a display unit a comparison result between the searched case and the current case and a treatment report including at least a next treatment plan for each of the searched cases; Equipped with The change trend of the characteristic parameter is a trend line formed by connecting points indicating the values ​​of the characteristic parameter during each follow-up record in a coordinate system; the display control unit ranks the retrieved cases according to whether or not there is a subsequent follow-up record after a follow-up record similar to the trend line of the current case in the retrieved cases; Case search device.

5. the display control unit ranks the retrieved cases according to a degree of decrease in the value of the characteristic parameter of the subsequent follow-up record. The case search device according to claim 4 .

6. An acquisition unit that acquires information about a target block of a current case and a change trend of a feature parameter of the target block; a search unit that searches, in the history data, for a case that has substantially the same block structure as the target block of the current case and has substantially the same change trend as the change trend of the feature parameter of the current case; a display control unit that displays on a display unit a comparison result between the searched case and the current case and a treatment report including at least a next treatment plan for each of the searched cases; Equipped with The change trend of the characteristic parameter is a trend line formed by connecting points indicating the values ​​of the characteristic parameter during each follow-up record in a coordinate system; the display control unit causes the trend line of the searched case to be superimposed on the trend line of the current case and displayed on the display unit; Case search device.

7. An acquisition unit that acquires information about a target block of a current case and a change trend of a feature parameter of the target block; a search unit that searches, in the history data, for a case that has substantially the same block structure as the target block of the current case and has substantially the same change trend as the change trend of the feature parameter of the current case; a display control unit that displays on a display unit a comparison result between the searched case and the current case and a treatment report including at least a next treatment plan for each of the searched cases; Equipped with The change trend of the characteristic parameter is a trend line formed by connecting points indicating the values ​​of the characteristic parameter during each follow-up record in a coordinate system; the display control unit causes the trend line of the searched case to be displayed on the display unit without being superimposed on the trend line of the current case. Case search device.

8. At least some of the trend lines of the retrieved cases are substantially identical to the trend line of the current case. The case search device according to claim 6 or 7.

9. a selection unit that selects some trend lines from the trend lines of the plurality of candidate blocks of the current case as target trend lines in response to an input operation; Furthermore, the search unit searches the history data for cases including a trend line substantially identical to the target trend line. The case search device according to claim 8 .

10. the display control unit ranks the retrieved cases according to the degree of similarity to the trend line of the current case. The case search device according to claim 6 or 7.

11. the display control unit ranks the retrieved cases according to a similarity between a block structure of the current case and the block structure of the target block; The case search device according to claim 6 or 7.

12. The acquisition unit analyzes the PET image data and the CT image data to acquire information about the target block and a change trend of the feature parameter. The case search device according to any one of claims 1 to 11.

13. a selection unit that selects one candidate block as the target block from among a plurality of candidate blocks of the current case in response to an input operation; The case retrieval device according to any one of claims 1 to 11, further comprising:

14. The characteristic parameters include at least one of a standardized absorption value, a metabolic tumor volume, a total tumor metabolic amount, and a tumor / normal ratio; The case search device according to any one of claims 1 to 11.

15. further comprising a storage unit that stores the history data; The history data includes, for each case, the block-structured image data, the characteristic parameter values ​​at each follow-up, and a treatment report. The case search device according to any one of claims 1 to 11.

16. The history data further records a change trend of the characteristic parameter. The case search device according to claim 15.

17. the search unit generates a change trend of the characteristic parameter based on the value of the characteristic parameter at each follow-up recorded in the history data, every time a search is performed. The case search device according to claim 15.

18. Obtain information about the target block of the current case and the change trend of the feature parameters of the target block; Searching the history data for cases that have substantially the same block structure as the target block of the current case and have substantially the same change trend as the change trend of the feature parameter of the current case; displaying on a display unit a treatment report including a comparison result between the retrieved cases and the current case and at least a next treatment plan for each of the retrieved cases; Each function of the computer executes the processing. The change tendency of the characteristic parameter includes an initial value of the characteristic parameter at the start of a search period, and a change direction and a change width of the value of the characteristic parameter throughout the entire search period. Case search methods.

19. Obtain information about the target block of the current case and the change trend of the feature parameters of the target block; Searching the history data for cases that have substantially the same block structure as the target block of the current case and have substantially the same change trend as the change trend of the feature parameter of the current case; displaying on a display unit a treatment report including a comparison result between the retrieved cases and the current case and at least a next treatment plan for each of the retrieved cases; Have the computer execute the process, The change tendency of the characteristic parameter includes an initial value of the characteristic parameter at the start of a search period, and a change direction and a change width of the value of the characteristic parameter throughout the entire search period. program.

Images