Control device and control method

Abstract

Ensure that diagnostic reference levels are properly managed. [Solution] The imaging information data management device of the embodiment manages diagnostic reference levels. The storage unit stores multiple groups of imaging information data. The determination unit determines whether the imaging information data of the subject input by the input unit matches the imaging information data of one of the multiple groups stored by the storage unit. The grouping unit groups the input imaging information data into one of the multiple groups if the determination unit determines that the input imaging information data matches the imaging information data of one of the multiple groups. The group generation unit generates a new group for the input imaging information data if the determination unit determines that the input imaging information data does not match the imaging information data of any of the multiple groups.

Description

Technical Field

[0001] Embodiments of the present invention relate to a management device and a management method.

Background Art

[0002] When performing medical imaging, usually, the setting of the radiation dose of the subject is performed by referring to the Diagnostic Reference Levels (DRLs).

[0003] There is a management device for imaging information data that clusters and groups imaging information data including inspection information data and dose data of each of a plurality of subjects, performs interquartile range (IQR) statistics on a plurality of dose data for each group, sets the 75th percentile, which is the third quartile, as the diagnostic reference level in the group, and when new inspection information data is grouped into the group, updates the diagnostic reference level in the group by performing interquartile range statistics on the plurality of dose data in the group.

[0004] However, in the conventional management device for imaging information data, there are few groups for the imaging information data. For example, in the Japanese Diagnostic Reference Level Standard (Japan DRLs 2020), in setting the diagnostic reference level for adult CT, the weight range is set only for 50 to 70 kg, and for adults with a weight exceeding 80 kg, the diagnostic reference level is not set. Therefore, when taking a medical image of an adult with a weight exceeding 80 kg, the dose cannot be set by referring to the diagnostic reference level.

[0005] In addition, when performing medical imaging of a subject, due to differences in the weight, physique, imaging device used, and scanning technique of the subject, the dose used becomes too large or too small, that is, the dose used becomes an outlier with respect to the diagnostic reference level. The conventional management device for imaging information data could not analyze the cause of the occurrence of the above outlier. [Prior art documents] [Patent Documents]

[0006] [Patent Document 1] Korean Published Patent Publication No. 2016-0009848 [Overview of the Initiative] [Problems that the invention aims to solve]

[0007] The problem that this invention aims to solve is to provide a management device and management method that can reliably manage the diagnostic reference level. [Means for solving the problem]

[0008] The imaging information data management device of this embodiment manages diagnostic reference levels and comprises a storage unit, an input unit, a determination unit, a grouping unit, and a group generation unit. The storage unit stores multiple groups relating to imaging information data. The input unit is for inputting imaging information data of a subject. The determination unit determines whether the imaging information data input by the input unit matches the imaging information data of one of the multiple groups stored by the storage unit. If the determination unit determines that the input imaging information data matches the imaging information data of one of the multiple groups, the grouping unit groups the input imaging information data into one of the multiple groups. If the determination unit determines that the input imaging information data does not match the imaging information data of any of the multiple groups, the group generation unit generates a new group for the input imaging information data. [Brief explanation of the drawing]

[0009] [Figure 1] Figure 1 is a block diagram showing an example of the functional configuration of a data management device for captured image information according to an embodiment. [Figure 2]Figure 2 shows an example of six groups of image information data according to the embodiment. [Figure 3] Figure 3 is a flowchart illustrating the method for managing image information data according to this embodiment. [Figure 4] Figure 4 is a diagram illustrating an example of the contents in step S120 according to the embodiment. [Figure 5] Figure 5 shows how to generate subgroups for a group that has abnormal dose data according to the embodiment. [Figure 6] Figure 6 shows the grouping of input shooting information data using the shooting information data management method according to the embodiment. [Figure 7] Figure 7 illustrates how, in one embodiment, subgroups are generated from manually selected anomalous dose data. [Figure 8] Figure 8 illustrates how, in one embodiment, subgroups are generated from manually selected anomalous dose data. [Figure 9] Figure 9 provides a brief explanation of an example of steps S303 and S304 in Figure 7. [Figure 10] Figure 10 illustrates how, in one embodiment, subgroups are generated from manually selected anomalous dose data. [Modes for carrying out the invention]

[0010] The following describes the camera data management device and camera data management method according to the embodiment, with reference to the drawings.

[0011] The image information data management device according to this embodiment is composed of multiple functional modules. These multiple functional modules can be installed as software on a device having a CPU (central process unit) and memory, such as an independent computer, or they can be installed in a distributed manner on multiple devices. This is achieved by executing each functional module of the image information data management device stored in memory using a processor. For example, such a functional module is a program.

[0012] The functions of the image data management device may be implemented in hardware form as circuits capable of executing each function. The circuits that implement the image data management device can transmit and receive data and collect data via a network such as the Internet.

[0013] Furthermore, the imaging data management device according to this embodiment can be installed at the medical imaging site and used to manage imaging data on-site. Alternatively, the imaging data management device may be directly implemented as part of a medical imaging diagnostic device (medical image acquisition device) such as a CT scanner or a magnetic resonance imaging scanner.

[0014] The following describes the camera data management device and camera data management method according to the embodiment, with reference to Figures 1 to 10.

[0015] Figure 1 is a block diagram showing an example of the functional configuration of a shooting information data management device according to an embodiment. As shown in Figure 1, the shooting information data management device 10 manages diagnostic reference levels and includes a storage means (storage unit) 11, an input means (input unit) 12, a determination means (determination unit) 13, a grouping means (grouping unit) 14, a group generation means (group generation unit) 15, a calculation means (calculation unit) 16, a display means (display unit) 17, and a subgroup generation means (subgroup generation unit) 18.

[0016] The storage means 11 stores a plurality of groups of imaging information data. For example, the storage means 11 is realized by various memories.

[0017] The input means 12 is for inputting imaging information data of a subject. For example, the input means 12 is realized by various input interfaces such as a mouse, a keyboard, etc.

[0018] The determination means 13 determines whether or not the imaging information data input by the input means 12 matches the imaging information data of any one of the plurality of groups stored by the storage means 11.

[0019] When the grouping means 14 is determined by the determination means 13 that the input imaging information data matches the imaging information data of one of the plurality of groups stored in the storage means 11, the input imaging information data is grouped into this one group.

[0020] When the group generation means 15 is determined by the determination means 13 that the input imaging information data does not match the imaging information data of any of the plurality of groups stored in the storage means 11, a new group is generated for the input imaging information data.

[0021] The calculation means 16 calculates the diagnostic reference level of each of the plurality of groups stored by the storage means 11.

[0022] The display means 17 displays the diagnostic reference levels of the plurality of groups stored in the storage means 11. The display means 17 may be constituted by a display (not shown) or the like.

[0023] The subgroup generation means 18 generates candidate groups with the largest number of samples based on the corresponding attribute data for imaging information data with abnormal dose data in the groups stored by the storage means 11, and generates candidate groups that satisfy predetermined conditions as subgroups of the stored group. The generated subgroups are stored in the storage means 11.

[0024] The imaging data includes the subject's examination information data and dose data.

[0025] The examination data includes subject information, scan information, scanner information, and hospital information. Subject information includes the subject's height, weight, age, sex, and the body area imaged. Scan information includes the scan protocol name, purpose of the scan, and type of scan sequence. Scanner information includes the scanner manufacturer, model number, and manufacturing date. Hospital information includes the hospital name, department, and radiologist.

[0026] The dose data is the dose applied when scanning a subject. The dose data for a group stored in the storage means 11 consists of multiple data points, including a diagnostic reference level. Typically, interquartile range statistics are performed on the multiple dose data points, and the 75th percentile may be set as the diagnostic reference level for this group, or the 25th percentile, 50th percentile, and 75th percentile may be set as the diagnostic reference levels for this group.

[0027] Figure 2 shows six groups of imaging information data. Here, the scan protocol name, the scanned body area of ​​the subject, the subject's age range, and the weight range are examination information data. CTDI (CT Dose Index) vol represents the absorbed radiation dose received per centimeter during a CT scan, and DLP (Dose-Length Product) is the absorbed radiation dose within the scan range. DLP = CTDI vol × scan range. CTDI vol and DLP are dose data, and in Figure 2, only the 75th percentile of CTDI vol and DLP are shown as diagnostic reference levels. The imaging information data for these six groups is stored by the storage means 11 and may be displayed by the display means 17.

[0028] The following explanation uses the first group as an example. In the first group, the scan protocol name used is "Chest (<1 year)", the body area scanned is "Chest", the age range of the subjects is "<1 year" (less than 1 year old), and the weight range is not statistically recorded. The 75th percentiles CTDIvol and DLP of the multiple dose data for the first group are "6 mGy" and "140 mGy*cm", respectively.

[0029] According to the data presented in this first group, when performing a chest scan on subjects under one year of age, the radiation dose can be applied to the subject based on CTDIvol: 6 mGy and DLP: 140 mGy*cm.

[0030] Furthermore, it is assumed that a chest scan is performed on a subject under one year of age, and the applied radiation doses are CTDIvol: 5 mGy and DLP: 130 mGy*cm, respectively. At this time, this imaging information data is input to the imaging information data management device 10 via the input means 12, and the determination means 13 determines whether the imaging information data matches the imaging information data of one of the six groups of imaging information data. Since the input protocol name, scanned body area, and age range match the first group, the determination means 13 determines that the input imaging information data matches the imaging information data of the first group, and the grouping means 14 groups the input imaging information data into the first group. Because the number of dose data in the first group has changed, it is necessary to recalculate the dose data at the 75th percentile as a new diagnostic reference level by the calculation means 16.

[0031] Furthermore, the imaging information data management device 10 may include an extraction means (not shown) for extracting and storing examination information data and radiographic imaging data from the imaging information data input by the input means 12. When only one exposure is performed when scanning the examination area of ​​a subject, the radiation dose in this exposure is usually recognized as dose data for imaging by this scan. In this case, the radiographic imaging data included in the imaging information data becomes the dose data.

[0032] However, X-ray and CT scans typically involve multiple exposures, and this radiographic data includes dose data for each exposure. In this case, the imaging information data management device 10 may include an analysis means (not shown) that analyzes the radiographic data to obtain dose data from the input imaging information data. Specifically, the analysis means calculates at least one of the following as dose data from the input imaging information data: the average value of the dose data for each exposure, the sum of the dose data for each exposure, or the maximum value among the dose data for multiple exposures. For example, when performing a multi-phase CT scan of the liver, it is necessary to calculate the average value of the dose data for each exposure and the sum of the dose data for each exposure as dose data from the input imaging information data.

[0033] The input means 12 may have the above-described functions of the extraction means and analysis means, and when the subject's imaging information data is input, it may be configured to acquire examination information data and dose data from the imaging information data.

[0034] Next, the method for managing the image information data according to this embodiment will be explained using Figures 3 to 6.

[0035] Figure 3 is a flowchart illustrating the method for managing image information data according to this embodiment.

[0036] In step S100, imaging data is input. Specifically, imaging data generated by irradiating the examination area of ​​the subject with radiation using a medical imaging device (e.g., a CT scanner) is input.

[0037] In step S110, the inspection information data and radiography data are extracted and saved. Specifically, the extraction function of the extraction means or input means 12 extracts the inspection information data and radiography data from the radiography information data entered in step S100.

[0038] In step S120, the shooting information data entered in step S100 is matched with the groups of shooting information data stored in the storage means 11. If it is determined that the entered shooting information data matches with any of the groups of shooting information data, the entered shooting information data is grouped into that group. If the entered shooting information data does not match with any of the groups of stored shooting information data, a new group is generated for this entered shooting information data.

[0039] Next, we will explain the specific details of step S120 using Figure 4.

[0040] As shown in Figure 4, step S120 includes steps S201 to S204. In step S201, text information data is extracted from the inspection information data extracted in step S110, and the text information data is converted into a high-dimensional vector. The text information data includes the name of the radiological inspection site, the name of the radiological diagnostic reference level, the scan protocol name, and the image data name. The text information data may be converted into a high-dimensional vector using the GPT algorithm, or other algorithms such as ELMo or BERT may be used.

[0041] In step S202, the similarity between the inspection information data extracted in step S110 and the inspection information data of one of the multiple groups of photographic information data stored in the storage means 11 is calculated. The similarity of text information data in the inspection information data is calculated, for example, using a cosine similarity algorithm, while the similarity of numerical data such as age and weight, and data such as product model number and software version in the inspection information data is calculated using a strict matching method. Finally, the similarity between the inspection information data extracted in step S110 and the inspection information data of the groups stored in the storage means 11 is obtained by combining the similarity calculated for all the inspection information data. Here, the similarity between the inspection information data extracted in step S110 and the inspection information data of the groups stored in the storage means 11 is calculated by converting the text information data into a high-dimensional vector.

[0042] In step S203, the similarity calculated in step S202 is compared with a preset threshold. If the similarity is greater than or equal to the threshold, it is determined that the inspection information data extracted in step S110 matches the inspection information data of a group stored in the storage means 11. The image information data entered in step S100 is then grouped into this group stored in the storage means 11. If the similarity is less than the threshold, it is determined that the inspection information data extracted in step S110 does not match the inspection information data of a group stored in the storage means 11, and the process proceeds to step S204.

[0043] In step S204, attribute data that does not match the inspection information data extracted in step S110 is identified from the group stored in the storage means 11 that has the highest similarity to the inspection information data extracted in step S110, and this attribute data is expanded to generate a new group for the imaging information data entered in step S100.

[0044] Step S204 will be explained using Figure 6. Figure 6 shows the grouping of the input shooting information data using the shooting information data management method according to the embodiment.

[0045] In Figure 6, when the imaging data in the lower left, namely "abdomen, 15 years old, CTDIvol:14, DLP:600," is entered, there is no group that matches this entered imaging data among the previously stored groups of imaging data. Therefore, from group 6, which has the highest similarity between the entered imaging data and the examination data, the system identifies the age data (>18 years old), which is attribute data that does not match the age (15 years old) in this examination data. By expanding the age data to "10-<18 years old" to include the age (15 years old), a new group 7 is generated for the entered imaging data. This expands the groups related to the diagnostic reference level. Note that "10-<18 years old" refers to, for example, the range of 10 years old and above but under 18 years old.

[0046] Next, we will explain step S130, which follows step S120 in Figure 3. In step S130, the radiation imaging data extracted in step S110 is analyzed to obtain dose data from the imaging information data input in step S100.

[0047] In step S140, the diagnostic reference level for each group is calculated. As described above, if it is determined that the examination information data extracted in step S110 matches the examination information data of a group stored in the storage means 11, the imaging information data entered in step S100 is grouped into this group stored in the storage means 11. At this time, the number of imaging information data in this group is changed, and the diagnostic reference level for this group needs to be recalculated by the calculation means 16. For example, the original diagnostic reference level for that group is updated by calculating a new 75th percentile as the diagnostic reference level for this group.

[0048] In step S150, when grouping the imaging information data entered in step S100 into groups stored in the storage means 11, the dose data of the entered imaging information data is compared with the diagnostic reference levels within this group. If the entered dose data is determined to be abnormal, a subgroup is generated for that group.

[0049] Specifically, the details of step S150 will be explained with reference to Figure 5. Figure 5 shows the generation of subgroups for a group that has abnormal dose data according to the embodiment.

[0050] In step S205, when grouping the input imaging information data into groups stored in the storage means 11, the dose data of the input imaging information data is compared with the matching diagnostic reference level of this group, which was recalculated in step S140.

[0051] In step S206, it is determined whether the dose data from the input imaging information data is an abnormal value relative to the recalculated diagnostic reference level. When the diagnostic reference level is the 75th percentile in the interquartile range statistics, if the dose data is greater than the diagnostic reference level, the dose data is determined to be an abnormal value. When the diagnostic reference levels are the 75th percentile and the 25th percentile in the interquartile range statistics, if the dose data is greater than the diagnostic reference level for the 75th percentile or less than the diagnostic reference level for the 25th percentile, the dose data is determined to be an abnormal value. In other words, if the dose data is outside the range consisting of the diagnostic reference level for the 25th percentile and the diagnostic reference level for the 75th percentile, the dose data is determined to be an abnormal value.

[0052] In step S207, if the dose data of the imaging information data entered in step S206 is determined to be an abnormal value relative to the recalculated diagnostic reference level, cluster analysis is performed on all the abnormal dose data within this group. That is, all the abnormal dose data within this group are extracted and grouped into one group. Then, attribute data that can further subdivide this group, such as subject information, scan information, scanner information, and hospital information, is added to this group. This added attribute data may be a single piece of information or a combination of several pieces of information. Then, the group with the largest number of samples among the multiple groups subdivided by the addition of attribute data is generated as a candidate group. This determines the grouping rule, which is the attribute data for generating the candidate group with the largest number of samples.

[0053] In step S208, it is determined whether the grouping rule, i.e., the added attribute data, is valid. In this embodiment, for example, interquartile range statistics are performed on the data of each group. Interquartile range statistics are performed using at least three samples. Therefore, the determination of whether the grouping rule is valid is made based on whether the number of samples in the candidate group is greater than or equal to a predetermined threshold (e.g., 3). If the number of samples in the candidate group is greater than or equal to the predetermined threshold, the grouping rule is determined to be valid, and if the number of samples in the candidate group is less than the predetermined threshold, the grouping rule is determined to be invalid.

[0054] If the grouping rule is determined to be invalid, the process proceeds to step S209 to determine if there are other rules. If there are other rules, steps S207 and S208 are executed again.

[0055] If the grouping rule is determined to be valid, the process proceeds to step S210, where the candidate group is generated as a subgroup of the corresponding group, the diagnostic reference level of the subgroup is calculated, and the subgroup and its diagnostic reference level are stored in the storage means 11.

[0056] Furthermore, in step S211, the diagnostic reference level for the previous group, i.e., the group corresponding to the subgroup, is recalculated.

[0057] In step S206, if the dose data of the input imaging information data is determined not to be an abnormal value relative to the diagnostic reference level of the matched group, the process proceeds to step S212. This is because the input imaging information data may be able to match with multiple groups stored in the storage means 11 simultaneously. In this case, it is determined that the matching of the input imaging information data with all groups is not yet complete, and the processes from step S205 onwards are continued.

[0058] If it is determined in step S212 that the input imaging information data has been matched with all groups, the process proceeds to step S160, where the display means 17 displays the diagnostic reference levels of multiple groups, including subgroups, that have been stored in the storage means 11 (see Figure 6).

[0059] Step S150 will be briefly explained using Figure 6.

[0060] In Figure 6, when the imaging data in the lower right, namely "chest, 36 years old, weight: 89 kg, CTDIvol: 26, DLP: 700," is input, this imaging data is grouped into group 5. However, the dose data in this imaging data, "CTDIvol: 26, DLP: 700," is much larger than the diagnostic reference level for group 5, "CTDIvol: 15, DLP: 500," and is recognized as abnormal dose data in group 5. Therefore, in step S150, attribute data for the weight range "80-100 kg" is added, and subgroup 5.1 is generated for group 5. By adding subgroup 5.1, it can be analyzed that the high dose data in the input imaging data in the original group is due to the subject's high weight. In other words, adding subgroups can be useful in analyzing the cause of abnormal dose data. It can also be used to broaden the groups related to diagnostic reference levels.

[0061] In the above embodiment, the method for managing the image information data according to this embodiment was explained using Figure 3. However, the method for managing the image information data according to this embodiment may have only steps S100 to S140, thereby generating a new group for input image information data that cannot be grouped. Alternatively, the method for managing the image information data according to this embodiment may have only steps S150 to S160, or only step S150, thereby generating a subgroup when the dose data of the input image information data becomes an abnormal value in a grouped group. Of course, even if no image information data is input, a subgroup of a group can be generated by step S150 for dose data in a group stored in the storage means 11 that has been determined to be abnormal.

[0062] Step S150 described the case where abnormal dose data in a group is automatically determined, but it is also possible to manually select abnormal dose data in a group. Next, Figures 7 to 10 will be used to describe the case where abnormal dose data in a group is manually selected.

[0063] Figure 7 illustrates how subgroups are generated from manually selected anomalous dose data.

[0064] In step S301, dose data corresponding to the user-specified group is displayed. In Figure 8, all dose data corresponding to the user-specified group "chest, 1-5 years old" is displayed at the bottom.

[0065] In step S302, the user selects data by drawing a curve on the graph or by using a predefined filter. As shown in Figure 8, the user selects dose data that has been identified as abnormal by circling it with a stylus on the graph displayed on the display means 17. As shown in Figure 10, dose data that has been identified as abnormal may be selected (filtered) using a filter that has been set in advance for ranges of attribute data including age range, examination site, weight range, scan mode, and manufacturer.

[0066] In step S303, cluster analysis is performed on the data selected in step S302 to determine a grouping rule that generates the candidate group with the largest number of samples. This step S303 is similar to step S207, where the data selected in step S302 is grouped into one group, and then attribute data that can further subdivide it, such as subject information, scan information, scanner information, and hospital information, is added to this group. This added attribute data may be a single piece of information or a combination of several pieces of information. Then, from the multiple groups subdivided by the addition of attribute data, the group with the largest number of samples is generated as the candidate group. This determines the grouping rule, which is the attribute data for generating the candidate group with the largest number of samples.

[0067] In step S304, it is determined whether the candidate group meets the user's expectations. If it does not meet the user's expectations, the process returns to step S303. If it does meet the user's expectations, the process proceeds to step S305.

[0068] Step S305 is similar to step S210, in which candidate groups are generated as subgroups of the corresponding group, a diagnostic reference level for the subgroup is calculated, and the subgroup and its diagnostic reference level are stored in the storage means 11.

[0069] Step S306 is similar to step S211, and recalculates the diagnostic reference level for the group corresponding to the previous group, i.e., the subgroup.

[0070] In Figure 7, step S304 may be replaced with step S208, where it is determined whether the grouping rule, i.e., the added attribute data, is valid. If it is determined that the grouping rule is valid, the process proceeds to step S305; if it is determined that the grouping rule is invalid, the process returns to step S303.

[0071] Steps S303 and S304 in Figure 7 will be briefly explained using Figure 9.

[0072] In Figure 9, for the abnormal dose data of the "chest 1-5 years" group manually selected by the user, the system determines the grouping rule (attribute data) of weight (15-30 kg) to generate candidate groups, displays data such as the diagnostic reference level for the candidate groups, and if the candidate groups meet the user's expectations, the user selects "Confirm" in Figure 9; if the candidate groups do not meet the user's expectations, the user selects "Cancel" in Figure 9.

[0073] As described above, by generating subgroups for abnormal dose data in the user-selected group, it is possible to analyze the cause of the abnormality in the user-selected dose data and to expand the group related to diagnostic reference levels.

[0074] Each component of the apparatus according to the above-described embodiment is a functional concept and does not necessarily have to be physically configured as shown in the figures. In other words, the specific form of distribution and integration of each apparatus is not limited to what is shown in the figures, and all or part of it can be functionally or physically distributed and integrated in any unit according to various loads and usage conditions. Furthermore, all or any part of each processing function performed in each apparatus may be realized by a CPU and a program analyzed and executed by the CPU, or it may be realized as hardware based on wiring logic.

[0075] Furthermore, the management device described in the above-mentioned embodiment can be realized by a computer, such as a personal computer or workstation, executing a pre-prepared program. This program can be distributed via a network such as the Internet. In addition, the program can be recorded on a computer-readable non-volatile recording medium such as a hard disk, flexible disk (FD), CD-ROM, MO, or DVD, and then read from the recording medium and executed by a computer.

[0076] According to at least one embodiment described above, the diagnostic reference level can be reliably controlled.

[0077] While several embodiments of the present invention have been described, these embodiments are presented as examples only and are not intended to limit the scope of the invention. These embodiments can be carried out in a variety of other forms, and various omissions, substitutions, and modifications can be made without departing from the spirit of the invention. These embodiments and their variations are included in the scope and spirit of the invention, as well as in the claims and their equivalents. [Explanation of Symbols]

[0078] 10 Management device 11 Memory means 12 Input methods 13 Judgment means 14 Grouping methods 15 Group generation means 16 Calculation method 17 Display means 18 Subgroup generation means

Claims

1. A device for managing imaging information data that manages diagnostic reference levels, A memory unit that stores multiple groups of shooting information data, An input unit for inputting the subject's imaging information data, A determination unit determines whether or not the shooting information data input by the input unit matches the shooting information data of one of the multiple groups stored by the storage unit. If the determination unit determines that the input shooting information data matches the shooting information data of one of the multiple groups, the grouping unit groups the input shooting information data into one of the multiple groups. If the determination unit determines that the input shooting information data does not match any of the shooting information data of any of the multiple groups, the group generation unit generates a new group for the input shooting information data. A control device equipped with the following features.

2. The aforementioned imaging data includes the subject's examination information data and dose data. The determination unit determines whether the first inspection information data, which is the inspection information data of the input imaging information data, matches the second inspection information data, which is the inspection information data of the imaging information data of one of the plurality of groups. The control device according to claim 1.

3. The determination unit calculates the similarity between the first inspection information data and the second inspection information data, determines that the first inspection information data and the second inspection information data are a match if the similarity is equal to or greater than a predetermined threshold, and determines that the first inspection information data and the second inspection information data are not a match if the similarity is less than the predetermined threshold. The control device according to claim 2.

4. The aforementioned inspection information data includes text information data, The similarity between the first inspection information data and the second inspection information data is calculated so as to convert the aforementioned text information data into a high-dimensional vector. The control device according to claim 3.

5. The aforementioned text information data includes the name of the area to be examined by radiation, the name of the reference level for radiation diagnosis, the name of the scan protocol, and the name of the image data. The control device according to claim 4.

6. If the determination unit determines that the first inspection information data and the second inspection information data do not match, the group generation unit identifies attribute data that does not match the first inspection information data from the group of second inspection information data with the highest similarity to the first inspection information data, expands this attribute data, and generates a new group for the input imaging information data. The control device according to claim 3.

7. An extraction unit extracts and stores inspection information data and radiography data from the imaging information data input by the aforementioned input unit, An analysis unit that analyzes the aforementioned radiation imaging data and acquires dose data from the input imaging information data, The control device according to claim 2, further comprising:

8. A calculation unit that calculates the diagnostic reference level for each of the multiple groups stored in the memory unit, A display unit that displays the diagnostic reference levels of multiple groups stored in the aforementioned storage unit, The control device according to claim 1, further comprising:

9. The memory unit further comprises a subgroup generation unit that, for imaging information data with abnormal dose data in a group stored by the memory unit, generates a candidate group with the largest number of samples based on the corresponding attribute data, and generates candidate groups that satisfy predetermined conditions as subgroups of the stored group. The subgroup is stored in the memory unit. The control device according to claim 2.

10. The abnormal dose data is higher than the diagnostic reference level. The control device according to claim 9.

11. There are multiple diagnostic reference levels, The abnormal dose data is data that falls outside the range of the multiple diagnostic reference levels. The control device according to claim 9.

12. The management device according to claim 10, wherein a candidate group that satisfies predetermined conditions is a group whose sample size is equal to or greater than a predetermined threshold.

13. The abnormal dose data is data manually selected by the user from the stored group. The control device according to claim 9.

14. The abnormal dose data is data selected by the user by drawing a curve on a graph from the stored data of the group. The control device according to claim 13.

15. The abnormal dose data is data that has been filtered by the user using a filter that sets a range of attribute data in advance from the stored data of the group. The control device according to claim 13.

16. Candidate groups that meet the specified conditions are groups that meet the user's expectations, or groups whose sample size is above a specified threshold. The control device according to claim 13.

17. A device for managing imaging information data that manages diagnostic reference levels, A memory unit that stores multiple groups of shooting information data, A subgroup generation unit generates candidate groups with the largest number of samples based on the corresponding attribute data for imaging information data with abnormal dose data in the group stored by the storage unit, and generates candidate groups that satisfy predetermined conditions as subgroups of the stored group. Equipped with, The subgroup is stored in the memory unit. Management device.

18. The abnormal dose data is higher than the diagnostic reference level. The control device according to claim 17.

19. There are multiple diagnostic reference levels, The abnormal dose data is data that falls outside the range of the multiple diagnostic reference levels. The control device according to claim 17.

20. Candidate groups that meet the specified conditions are those with a sample size equal to or greater than a specified threshold. The control device according to claim 18 or 19.

21. The abnormal dose data is data manually selected by the user from the stored group. The control device according to claim 17.

22. The abnormal dose data is data selected by the user by drawing a curve on a graph from the stored data of the group. The control device according to claim 21.

23. The abnormal dose data is data that has been filtered by the user using a filter that sets a range of attribute data in advance from the stored data of the group. The control device according to claim 21.

24. Candidate groups that meet the specified conditions are groups that meet the user's expectations, or groups whose sample size is above a specified threshold. The control device according to claim 21.

25. A method for managing imaging information data to manage diagnostic reference levels, A determination step to determine whether the input subject imaging information data matches the imaging information data of one of several groups of imaging information data stored in the memory unit, If, in the determination step, it is determined that the input shooting information data matches the shooting information data of one of the multiple groups, a grouping step is performed to group the input shooting information data into this one of the multiple groups. If, in the determination step, it is determined that the input shooting information data does not match the shooting information data of any of the multiple groups, a group generation step is performed to generate a new group for the input shooting information data. Management methods, including those mentioned above.

26. A method for managing imaging information data to manage diagnostic reference levels, A first storage step involves storing multiple groups of shooting information data in a memory unit, A subgroup generation step is performed in which, for imaging information data with abnormal dose data in the group stored in the storage unit in the first storage step, a candidate group with the largest number of samples is generated based on the corresponding attribute data, and a candidate group that satisfies predetermined conditions is generated as a subgroup of the stored group. A second storage step of storing the subgroup in the storage unit, Management methods, including those mentioned above.

Images