JP7626626B2 - Image reading support device and operation method of image reading support device - Google Patents

Description

The embodiments disclosed in the present specification and drawings relate to a computer-aided image interpretation device and a method for operating the computer-aided image interpretation device .

In recent years, regular testing using tumor markers has become common as a means of checking the progression of cancer in cancer patients. For example, blood tests (such as breast cancer marker tests) using microRNA (ribonucleic acid) as a tumor marker are known to be able to detect cancer early. Tests using such tumor markers are a useful method for finding out whether cancer has recurred, the progression of the disease, and the presence of cancer that has not been completely removed by surgery or is too small to be seen by diagnostic imaging.

For example, after surgical removal of a cancer with a high tumor marker level, the tumor marker level often decreases and the condition improves. However, if the cancer recurs, the tumor marker level rises again, and tumor marker testing may be used for follow-up observation after surgery.

Here, in tests using tumor markers, it is difficult to identify the affected area using a single type of marker; the affected area is identified by using a combination of multiple types. In addition, since the tumor marker value itself does not accurately reflect the movement of cancer, it is difficult to understand the state of cancer using the tumor marker value alone. For example, in tests using tumor markers, there are cases where the tumor marker value increases even when there is no cancer (false positive), and cases where the tumor marker value shows a low value even when the cancer is progressing (false negative).

As such, the presence or absence of cancer and its progression cannot be determined by tumor marker values alone, so in addition to tests using tumor markers, tests using medical images are generally performed, and diagnoses are made based on a combination of these medical images. For this reason, there is a demand for collaboration between blood tests, medical information viewers, and medical image diagnostic devices to make diagnoses more efficiently.

JP 2017-188000 A

One of the problems that the embodiments disclosed in this specification and the drawings aim to solve is to improve the quality of testing. However, the problems that the embodiments disclosed in this specification and the drawings aim to solve are not limited to the above problem. Problems that correspond to the effects of each configuration shown in the embodiments described below can also be positioned as other problems.

The image interpretation support device according to the embodiment includes a judgment unit, a decision unit, and an acquisition unit. The judgment unit judges the state of the subject based on the change over time in the value of the tumor marker. The decision unit decides a tumor marker value to be compared from among a plurality of tumor marker values acquired over time according to the state of the subject. The acquisition unit acquires a medical image corresponding to the tumor marker value to be compared.

FIG. 1 is a block diagram showing an example of the configuration of an image reading apparatus according to the first embodiment. FIG. 2 is a diagram for explaining an example of a workflow of an examination using a tumor marker. FIG. 3 is a diagram for explaining an example of processing of the image reading apparatus according to the first embodiment. FIG. 4 is a diagram for explaining an example of a process related to image selection by the image reading apparatus according to the first embodiment. FIG. 5 is a diagram for explaining an example of a process performed by the decision function according to the first modification. FIG. 6 is a diagram illustrating an example of a process performed by a control function according to the second modification. FIG. 7 is a diagram for explaining an example of a process performed by a control function according to the third modification. FIG. 8 is a flowchart showing a procedure of processing by the image reading apparatus according to the first embodiment. FIG. 9 is a diagram for explaining an example of processing of the image reading apparatus according to the second embodiment.

Below, embodiments of the image reading support device and image reading support method are described in detail with reference to the drawings. Note that in the following embodiments, an example is described in which the image reading support device of the present application is applied to an image reading device, but the image reading support device and image reading support method according to the present application are not limited to the embodiments shown below. In addition, the embodiments can be combined with other embodiments or conventional technologies to the extent that no inconsistencies arise in the processing content.

(First embodiment)
An example of the configuration of an image reading apparatus according to the first embodiment will be described. Fig. 1 is a block diagram showing an example of the configuration of an image reading apparatus 5 according to the first embodiment. The image reading apparatus 5 is connected to an in-vitro diagnostic apparatus 1, an electronic medical record system 2, a PACS (Picture Archiving and Communication Systems) 3, and a medical image diagnostic apparatus 4 via a network. Note that other systems such as a HIS (Hospital Information System) and a RIS (Radiology Information System) may also be included in the network and connected to the image reading apparatus 5.

The in vitro diagnostic device 1 performs various tests on samples such as blood and tissue collected from a subject. For example, the in vitro diagnostic device 1 performs a blood test using microRNA as a tumor marker and obtains the tumor marker value of each tumor marker. The in vitro diagnostic device 1 then transmits the test results (e.g., tumor marker values, etc.) to each device on the network.

The electronic medical record system 2 stores various medical data related to the subject's medical records and patient information. Specifically, the electronic medical record system 2 generates medical data related to the subject or acquires the medical data from other devices via a network, and stores the medical data in a memory circuit within the system. For example, the electronic medical record system 2 is realized by computer equipment such as a server or a workstation.

The PACS 3 receives, stores, and manages medical image data collected by the medical image diagnostic device 4 via a network. For example, the PACS 3 stores and manages CT image data, ultrasound image data, MRI image data, and the like collected from the same subject.

The medical image diagnostic device 4 captures an image of a subject to generate medical image data. The medical image diagnostic device 4 then transmits the generated medical image data to various devices on the network. For example, the medical image diagnostic device 4 is an X-ray diagnostic device, an X-ray CT (Computed Tomography) device, an MRI (Magnetic Resonance Imaging) device, an ultrasound diagnostic device, a SPECT (Single Photon Emission Computed Tomography) device, a PET (Positron Emission computed Tomography) device, etc.

The image reading device 5 displays medical images acquired via the network. An operator, such as an image reading doctor, who operates the image reading device 5 performs image reading on the medical images displayed on the display of the image reading device 5. For example, as shown in FIG. 1, the image reading device 5 has a communication interface 51, an input interface 52, a display 53, a memory circuitry 54, and a processing circuitry 55. The image reading device 5 is connected via the network so as to be able to communicate with each device on the network.

The communication interface 51 is connected to the processing circuit 55 and controls communication between each device on the network. Specifically, the communication interface 51 receives various information from each device and outputs the received information to the processing circuit 55. For example, the communication interface 51 is realized by a network card, a network adapter, a NIC (Network Interface Controller), etc.

The input interface 52 is connected to the processing circuit 55 and accepts various instructions and information input operations from the operator. Specifically, the input interface 52 converts the input operations accepted from the operator into electrical signals and outputs them to the processing circuit 55. For example, the input interface 52 is realized by a trackball, a switch button, a mouse, a keyboard, a touchpad that performs input operations by touching the operation surface, a touch screen in which the display screen and the touchpad are integrated, a non-contact input circuit using an optical sensor, and a voice input circuit. Note that in this specification, the input interface 52 is not limited to only those that have physical operating parts such as a mouse and a keyboard. For example, an electrical signal processing circuit that receives an electrical signal corresponding to an input operation from an external input device provided separately from the device and outputs this electrical signal to the processing circuit is also included as an example of the input interface 52.

The display 53 is connected to the processing circuit 55 and displays various information and images. Specifically, the display 53 converts the information and image data sent from the processing circuit 55 into an electrical signal for display and outputs it. For example, the display 53 can be realized by a liquid crystal display, a CRT (Cathode Ray Tube) display, an organic EL (Electro Luminescence) display, a plasma display, a touch panel, etc.

The memory circuitry 54 is connected to the processing circuitry 55 and stores various data. For example, the memory circuitry 54 stores various data received from each device arranged on a network (not shown), the processing results of the image reading device 5, and the like. The memory circuitry 54 also stores various programs that the processing circuitry 55 reads out and executes to realize various functions. As shown in FIG. 1, the memory circuitry 54 also stores a judgment condition 541 and corresponding information 542. The judgment condition 541 and the corresponding information 542 will be described in detail later. For example, the memory circuitry 54 is realized by a semiconductor memory element such as a RAM (Random Access Memory) or a flash memory, a hard disk, an optical disk, or the like.

The processing circuitry 55 controls the operation of the image reading device 5 in response to an input operation received from an operator via the input interface 52. Specifically, as shown in FIG. 1, the processing circuitry 55 controls the operation of the image reading device 5 by executing a control function 551, a judgment function 552, a decision function 553, an acquisition function 554, and a generation function 555. Here, the control function 551 is an example of a display control unit. The judgment function 552 is an example of a judgment unit. The decision function 553 is an example of a decision unit. The acquisition function 554 is an example of an acquisition unit. For example, the processing circuitry 55 is realized by a processor.

The control function 551 controls the execution of processes in response to various requests input via the input interface 52. For example, the control function 551 controls the sending and receiving of various information via the communication interface 51, the storage of information in the memory circuit 54, and the display of various information on the display 53. The control function 551 also controls the display of the display information generated by the generation function 555, which will be described in more detail later.

The assessment function 552 assesses the condition of the subject based on the change in the tumor marker over time. The determination function 553 determines a comparison tumor marker value from among multiple tumor marker values acquired over time, depending on the subject's condition. The acquisition function 554 acquires a medical image corresponding to the comparison tumor marker value. The processing of the assessment function 552, determination function 553, and acquisition function 554 will be described in detail later.

The generating function 555 generates various display information. Specifically, the generating function 555 generates display information to assist in interpretation. For example, the generating function 555 generates information showing changes in tumor markers over time, various medical images, and the like. To give one example, the generating function 555 generates a graph showing the tumor marker values for each date on which a blood test using the tumor marker was performed. The generating function 555 also generates medical images of various parts of the subject by performing various image processing on medical image data acquired via the network.

The overall configuration of the image reading device 5 has been described above. With this configuration, the image reading device 5 improves the quality of examinations. Specifically, the image reading device 5 automatically searches for and acquires medical images suitable for comparative image reading, thereby reducing the workflow of image selection work and improving the quality of examinations.

For example, when a tumor marker test is performed for follow-up observation, it is necessary to select medical images for comparative reading based on the tumor marker values of previous tests in relation to the current tumor marker value, and the workflow for selecting images suitable for comparative reading is extensive and time-consuming. Therefore, if there is not enough time to complete the work, sufficient image selection may not be performed, which may result in a decline in the quality of the test.

Figure 2 is a diagram for explaining an example of a workflow for testing using tumor markers. Here, Figure 2 shows an example of a workflow for testing using tumor markers in follow-up observation after cancer treatment.

For example, the flow of cancer treatment is as shown in Figure 2. First, a medical examination and diagnosis are performed on the examined subject. If a cancer diagnosis is made, treatment is then administered and the patient is monitored. If recurrence or metastasis occurs, treatment is administered again.

In this embodiment, the quality of the examination using tumor markers is improved in the follow-up observation described above. For example, in the examination using tumor markers in the follow-up observation after cancer treatment shown in FIG. 2, a tumor marker examination using various tumor markers is performed. In this examination, for example, the effects of chemotherapy or radiation therapy are confirmed, and the progression state of cancer is confirmed by periodically measuring tumor markers.

As mentioned above, when testing using tumor markers, a diagnosis cannot be made based on tumor marker values alone, so medical images are collected. For example, for follow-up observation, ultrasound images are collected using an ultrasound diagnostic device to examine the size of the cancer, the position of blood vessels, and the presence or absence of ascites or nodules in the liver. Also, for detailed examinations, such as when abnormalities are found in tumor markers, CT images and MR images are collected using an X-ray CT device or MRI device to examine the nature, distribution, and metastasis state of the cancer.

An operator, such as a radiologist, who operates the radiology device 5, compares the medical images collected by the various medical image diagnostic devices described above with past images and performs a diagnosis by combining the tumor marker value and the radiology results. In comparative interpretation of medical images, as shown in FIG. 2, the radiologist first (1) checks the current marker value, and then (2) checks the past marker values and considers their progress. From the results, the radiologist (3) determines a tumor marker value to be compared (the best time for comparison) from among the past tumor marker values, and (4) determines a medical image that corresponds to the determined tumor marker value to be compared. The radiologist then performs comparative interpretation of the determined medical image with the medical image collected this time.

Thus, when reading images associated with follow-up tests using tumor markers, the medical images to be compared must be selected based on the tumor marker values from previous tests, and the radiologist will follow steps (1) to (4) above. Therefore, it takes time to search for medical images that are appropriate for comparison. For this reason, when there is not enough time to search for images and it is difficult to consider which images are appropriate for comparison, the medical images collected during the previous test may be selected as the images to be compared.

The image reading device 5 according to the first embodiment automatically searches for and acquires medical images suitable for comparative image reading, thereby reducing the workflow of image selection work and improving the quality of examinations. The details of the processing performed by the image reading device 5 are described below.

The control function 551 acquires the tumor marker values of the subject in response to an operation via the input interface 52. Specifically, the control function 551 acquires the test results of the tumor marker tests previously performed on the subject to be interpreted in response to an operation via the input interface 52. For example, the control function 551 acquires the past tumor marker values and the tumor marker values at the time of interpretation for all tumor markers for which the subject has been tested. Note that the control function 551 may acquire the tumor marker values directly from the in vitro diagnostic device 1 via the network, or may acquire the tumor marker values from the electronic medical record system 2.

The determination function 552 determines the subject's condition based on the change over time in the tumor marker value. Specifically, the determination function 552 determines the subject's condition at the time of determination based on the change in the tumor marker value from a time point prior to the time of determination to the time of determination for which the subject's condition is to be determined. For example, the determination function 552 determines the subject's condition based on the change in the tumor marker value at the time of interpretation for past tumor marker values.

Here, the judgment function 552 judges the state of the subject based on the judgment conditions 541 stored in the memory circuitry 54. The judgment conditions 541 include information for judging the state of the subject for each type of tumor marker. Specifically, the judgment conditions 541 include a threshold value for each tumor marker for suggesting the presence of cancer.

In tests using tumor markers, various tumor markers are used according to the type of cancer. Here, examples of tumor markers used in tumor marker tests include "AFP", "PIVKA-II", "CA-125", "CA15-3", "CEA", and "NCC-ST-439". In tumor marker tests, various tumor markers are used according to the type of cancer to be tested, such as hepatocellular carcinoma, breast cancer, esophageal cancer, lung cancer, biliary tract cancer, prostate cancer, neuroblastoma, medullary thyroid cancer, gastric cancer, pancreatic cancer, colon cancer, cervical cancer, uterine cancer, and ovarian cancer.

The judgment condition 541 includes a threshold value for indicating the presence of cancer for each type of tumor marker described above. For example, the judgment function 552 judges the subject's condition to be "worsening" if the tumor marker value at the time of the image interpretation increases from the past tumor marker value and exceeds the above threshold value.

Furthermore, the judgment function 552 judges the degree of "worsening" based on the degree of increase in the tumor marker value. For example, if the increase from the past (previous) tumor marker value exceeds the threshold, the judgment function 552 judges the subject's condition to be "rapidly worsening." On the other hand, if the increase from the past (previous) tumor marker value does not exceed the threshold, the judgment function 552 judges the subject's condition to be "tending to worsening." The threshold value to be compared with the increased tumor marker value is set appropriately for each type of tumor marker, and is stored in the memory circuitry 54 as the judgment condition 541.

Similarly, the judgment function 552 judges the subject's condition to be "recovered" if the tumor marker value at the time of the image reading has dropped from the past tumor marker value and fallen below the threshold value. Furthermore, the judgment function 552 judges the degree of "recovery" based on the degree of drop in the tumor marker value. For example, the judgment function 552 judges the subject's condition to be "rapid recovery" if the drop in the value from the past (previous) tumor marker value exceeds the threshold value. On the other hand, if the drop in the value from the past (previous) tumor marker value does not exceed the threshold value, the judgment function 552 judges the subject's condition to be "tending to recovery." The threshold value to be compared with the drop in the tumor marker value is set appropriately for each type of tumor marker and stored in the memory circuitry 54 as the judgment condition 541.

The determination function 553 determines a tumor marker value to be compared from among multiple tumor marker values acquired over time, depending on the condition of the subject. Specifically, the determination function 553 determines a tumor marker value to be compared with the tumor marker at the time to be determined, from among multiple tumor marker values acquired at a time prior to the time to be determined, depending on the condition of the subject at the time to be determined. For example, the determination function 553 determines a past tumor marker value to be compared with the time to be the subject of interpretation, depending on the condition of the subject determined by the determination function 552.

Here, the determination function 553 determines the tumor marker value to be compared based on the correspondence information 542 stored in the memory circuitry 54. The correspondence information 542 includes information in which the state of the tumor marker value to be compared is associated with each state of the subject. For example, the correspondence information 542 includes information in which the state of the subject's condition "worsening" is associated with "stable," which indicates a low value in the transition of the past tumor marker value. When based on such information, when the state of the subject is determined to be "worsening," the determination function 553 determines the tumor marker value indicating the low value from among the past tumor marker values as the tumor marker value to be compared. In addition to the above information, the correspondence information 542 can also include information in which the state of the tumor marker value is associated with the state of the subject's condition "sudden deterioration" or "tending to worsening," for example.

The correspondence information 542 also includes information in which a "high value" indicating a high value after treatment in the transition of past tumor marker values is associated with the subject's condition of "recovery." When based on such information, when the subject's condition is determined to be "recovered," the determination function 553 determines the tumor marker value indicating the high value from among the past tumor marker values as the comparison target. In addition to the information described above, the correspondence information 542 can also include information in which, for example, the tumor marker value state is associated with the subject's condition of "rapid recovery" or "tending to recovery."

The determination function 553 determines the tumor marker value to be compared with the time point to be interpreted based on the above-mentioned correspondence information 542. In other words, the determination function 553 determines the tumor marker value to be compared with the time point to be interpreted by using the correspondence information 542, thereby determining the past time point to be compared with the time point to be interpreted.

The acquisition function 554 acquires a medical image corresponding to the tumor marker value of the comparison target determined by the determination function 553. Specifically, the acquisition function 554 acquires a medical image corresponding to the time point at which the tumor marker value of the comparison target was acquired. The acquisition function 554 also acquires a medical image corresponding to the time point that is the subject of interpretation. Here, the acquisition function 554 can acquire a medical image corresponding to the determined tumor marker value from the PACS 3 or the medical image diagnostic device 4.

For example, the acquisition function 554 acquires medical images acquired on the same day as the tumor marker value at the time of interpretation was acquired (or within a specified period from the day the tumor marker value was acquired). The acquisition function 554 also acquires medical images acquired on the same day as the tumor marker value determined by the determination function 553 was acquired (or within a specified period from the day the tumor marker value was acquired).

The generation function 555 generates various display information for displaying the medical image acquired by the acquisition function 554 on the display 53. Specifically, the generation function 555 generates various information for comparative interpretation of the medical image at the time point to be interpreted and the medical image corresponding to the time point at which the tumor marker value determined by the determination function 553 was acquired.

The control function 551 causes the display 53 to display the medical images acquired by the acquisition function 554 and the display information generated by the generation function 555. An operator, such as an image interpretation doctor, interprets the medical images displayed by the display 53. As described above, the image interpretation device 5 can automatically search for, acquire, and display medical images suitable for comparison with the medical image at the time of the image interpretation, thereby reducing the workflow of image selection work and enabling the quality of examinations to be improved.

An example of the processing by the image reading device 5 according to this embodiment will be described below. FIG. 3 is a diagram for explaining an example of the processing by the image reading device 5 according to the first embodiment. Here, FIG. 3 shows an example in which the tumor marker value increases during follow-up observation.

For example, the flow of cancer treatment is as shown in Figure 3. First, a medical examination and diagnosis are performed on the examined subject. If a cancer diagnosis is made, treatment is then given and the patient is monitored. If recurrence or metastasis occurs, treatment will be given again.

Here, suppose that during follow-up, a tumor marker test using a tumor marker shows an increase in tumor marker value from the previous test value and exceeds the threshold value.

In such a case, the tumor marker value results indicate a high urgency, so for example, a CT scan is performed to collect CT images, and interpretation is performed. The tumor marker value and the interpretation results are then integrated to make a diagnosis. In the first embodiment, medical images are automatically selected during interpretation in the above-mentioned process.

That is, as shown in FIG. 3, the image reading device 5 first (1) judges the current marker value as "the tumor marker value exceeds the threshold value". Then, the image reading device 5 (2) judges the subject's condition as "worsening" because a comparison with past marker values shows an "increase" from a "low value" to a "high value". Furthermore, the image reading device 5 determines the optimal marker value for comparison (optimal time for comparison) based on the judgment result and the correspondence information 542. That is, the image reading device 5 (3) determines the (post-treatment) tumor marker value that showed "stable" as a comparison target for the tumor marker value that showed "worsening".

After that, the image reading device 5 (4) determines and acquires a medical image corresponding to the tumor marker value indicating "stable." For example, the image reading device 5 narrows down the target examination (date and time) based on the tumor marker value result indicating "stable," and determines that the image is one in which the patient is stable based on the diagnosis result and examination information of the target examination, and selects the image. If there is no target image, the image reading device 5 performs image reconstruction, clinical analysis again, transition to another examination, etc. Then, the image reading device 5 acquires the medical image and (5) displays the acquired medical image and the medical image collected this time. This allows the image reading doctor to perform comparative interpretation and, for example, make a judgment such as "there is no recurrence, but metastasis to other tissues has been confirmed" based on the difference result of the displayed medical images.

In this way, the image reading device 5 automatically performs the above-mentioned steps (1) to (4). The following describes the detailed processing of the automatic procedures. FIG. 4 is a diagram for explaining an example of processing related to image selection by the image reading device 5 according to the first embodiment. Here, FIG. 4 shows the acquisition status of tumor markers and medical images, and a graph showing the transition of tumor marker values. Specifically, FIG. 4 shows the acquisition status of "tumor marker i" and "tumor marker ii" from "2019/1/1" to "2020/7/1", and the acquisition status of "ultrasound image (ultrasound in the figure)" and "CT image (CT in the figure)" from "2019/1/1" to "2020/7/1" in the upper part of the figure.

Figure 4 also shows a graph at the bottom showing the values of "tumor marker i" and "tumor marker ii" from "2019/1/1" to "2020/7/1." Here, curve L1 in Figure 4 shows the graph of "tumor marker i," and curve L2 shows the graph of "tumor marker ii."

For example, in reading after an examination using a tumor marker on "2020/7/1", the control function 551 obtains the tumor marker values of "tumor marker i" and "tumor marker ii" on "2020/7/1" and the previously obtained values of each tumor marker.

The determination function 552 determines the subject's condition on "2020/7/1" based on the tumor marker values of "tumor marker i" and "tumor marker ii" on "2020/7/1" and the past values of each tumor marker. That is, the determination function 552 determines the subject's cancer condition based on the transition state of "tumor marker i" and the transition state of "tumor marker ii."

For example, the determination function 552 determines, based on the transition of the value of "tumor marker i" shown in curve L1 of FIG. 4, that the value of "tumor marker i" has risen to the same level as the threshold value of "tumor marker i" as of "2020/4/1" and that this value has been maintained on "2020/7/1". In addition, based on the transition of the value of "tumor marker ii" shown in curve L2 of FIG. 4, the determination function 552 determines that the value of "tumor marker ii" on "2020/7/1" has risen from the value on "2020/4/1" and indicates a high value that exceeds the threshold value of "tumor marker ii".

Then, the judgment function 552 compares the degree of increase in each tumor marker value (for example, the amount of change in the value of tumor marker II from 4/1 to 7/1) with a threshold value and judges that the subject's cancer condition is "worsening."

Since the judgment result by the judgment function 552 is "worsening," the decision function 553 determines the tumor marker value indicating "stable" as the tumor marker value to be compared. For example, based on the transition of the value of "tumor marker i" (curve L1) and the transition of the value of "tumor marker ii" (curve L2) in FIG. 4, the decision function 553 determines the value of "2019/10/1," which indicates low values of "tumor marker i" and "tumor marker ii," as the value to be compared.

Here, in the transition of the values of "tumor marker i" and "tumor marker ii" shown in FIG. 4, both marker values are below the threshold even on "2020/1/1". As shown in FIG. 4, when there are tumor marker values that can be compared at multiple time points, the determination function 553 can determine the tumor marker value to be compared based on arbitrarily set conditions. For example, the determination function 553 can compare tumor marker values and determine the tumor marker value to be compared based on the comparison result. As an example, the determination function 553 determines the tumor marker value (time point) that shows the lowest or highest value as the comparison target.

In addition, for example, the determination function 553 can determine the tumor marker value to be compared based on the date on which the tumor marker value was acquired. As an example, the determination function 553 can determine, as the tumor marker value to be compared, the tumor marker value of the date closest to "2020/7/1" that is the subject of the image reading, or the tumor marker value of the date furthest from "2020/7/1" that is the subject of the image reading, among multiple time points.

For example, the determination function 553 can determine a tumor marker value to be compared based on the acquired medical image. As an example, the determination function 553 can determine, as a comparison target, the tumor marker value at the time when a medical image of the same type as the medical image acquired at the time to be interpreted was acquired among multiple time points. For example, as shown in FIG. 4, the determination function 553 determines, as a comparison target, the tumor marker value of "2019/10/1" when the same CT image as the CT image acquired on "2020/7/1" to be interpreted was acquired among "2019/10/1" and "2020/1/1" when the tumor marker value is stable.

The acquisition function 554 acquires medical images at the time determined by the determination function 553. Specifically, the acquisition function 554 acquires medical images of the same type as the medical images acquired at the time of interpretation from among the medical images acquired at the time determined by the determination function 553. For example, the acquisition function 554 acquires CT images acquired at the time corresponding to "2019/10/1". The acquisition function 554 also acquires CT images corresponding to the time "2020/7/1" that is the subject of interpretation.

The control function 551 displays the medical image acquired by the acquisition function 554 on the display 53. For example, the control function 551 displays the CT image acquired at the time corresponding to "2019/10/1" and the CT image corresponding to "2020/7/1" on the display 53. This allows the radiologist to compare the CT image of "2020/7/1" suggesting that the subject's condition has deteriorated and the CT image of "2019/10/1" in which the subject's condition is stable, without having to take the time to select an image. As a result, the quality of the examination is improved. In the example shown in FIG. 4, the case where two types of tumor markers (tumor marker I and tumor marker II) are used is described, but the embodiment is not limited to this. For example, an image for comparative reading may be selected from the examination results using one type or three or more types of tumor markers.

Here, the judgment function 552 judges the condition of each part of the subject based on the type of tumor marker, the determination function 553 determines the value of the tumor marker to be compared for each type of tumor marker, and the acquisition function 554 can acquire medical images corresponding to the value of the tumor marker to be compared for each part of the subject based on the type of tumor marker.

In tests using tumor markers, tumor markers targeting various sites can be used in a single test. That is, tests can be performed for tumor markers other than the above-mentioned "tumor marker I" and "tumor marker II." Furthermore, the same type of tumor marker suggests the presence of cancer in different sites. In other words, the same type of tumor marker is used in tests for cancer in different sites.

The determination function 552 then extracts all target areas according to the type of tumor marker used. The acquisition function 554 acquires medical images of the areas extracted by the determination function 552 for the time points to be interpreted and the time points determined by the determination function 553.

Here, when there is no medical image of the extracted area, the acquisition function 554 can generate a new medical image by various image processing. When CT images or the like are acquired in conjunction with an examination using a tumor marker, medical image data is often collected over a wide area of the subject (e.g., the entire body). Thus, the acquisition function 554 reconstructs a medical image of the extracted area, for example, using medical image data collected over a wide area. Also, for example, the acquisition function 554 acquires medical images of each area by applying a clinical analysis app for each area to the medical image data collected from the subject.

The acquisition function 554 can also control the control function 551 to display information on the display 53 prompting the collection of medical image data of the extracted area when medical image data of a wide range has not been collected and medical image data including the extracted area has not been saved. In other words, the acquisition function 554 can also control the collection of medical images of the area from the subject.

As described above, the image reading device 5 according to this embodiment can select and display an appropriate medical image from past medical images according to the condition of the subject when reading medical images accompanying an examination using a tumor marker. Here, the image reading device 5 can execute various processes in addition to the processes described above. Below, modified examples of the image reading device 5 are described.

(Variation 1)
In the above-described embodiment, when there are tumor marker values that can be compared at multiple points in time, the tumor marker values to be compared are determined based on arbitrarily set conditions (conditions based on dates, conditions based on acquired medical images). The image interpretation device 5 can further determine the tumor marker values to be compared based on the characteristics of the medical images.

In such a case, the determination function 553 extracts multiple candidates for a tumor marker value to be compared from among the multiple tumor marker values acquired over time based on the tumor marker value, and determines a tumor marker value to be compared from among the multiple candidates based on the characteristics of each medical image corresponding to the multiple candidates. Specifically, the determination function 553 extracts multiple candidates based on the tumor marker value, and determines from among the extracted multiple candidates a candidate associated with a medical image having characteristics in line with the determination result by the determination function 552 as the tumor marker value to be compared.

Figure 5 is a diagram for explaining an example of processing by the decision function 553 according to the first modified example. Here, Figure 5 shows a graph in which the vertical axis indicates the tumor marker value and the horizontal axis indicates time. Figure 5 also shows the processing by the decision function 553 when performing an interpretation at time t4.

For example, as shown in FIG. 5, the tumor marker value at time t4 is in a state that greatly exceeds the threshold value. Therefore, the judgment function 552 judges that the subject's condition at time t4 is in a "worsening" state. In response to this judgment result, the decision function 553 first extracts times t1, t2, and t3, when the tumor marker value is low, as candidates for comparison in order to determine the point in time of the stable state.

Then, the determination function 553 acquires medical image I1 corresponding to time t1, medical image I2 corresponding to time t2, and medical image I3 corresponding to time t3, compares medical images I1 to I3, and determines the image that is considered to have been collected under more stable conditions as the comparison subject. For example, the determination function 553 extracts tumors in each of medical images I1 to I3 and measures the size of the extracted tumors. The determination function 553 then determines the medical image with the smallest measured tumor size as the one that was collected under more stable conditions as the comparison subject for time t4.

Also, for example, the determination function 553 extracts multiple candidates for a comparison tumor marker value from multiple tumor marker values acquired over time based on the tumor marker value, and determines a comparison tumor marker value from the multiple candidates based on the similarity between each medical image corresponding to the multiple candidates and the medical image at the time of determination when the subject's condition was determined.

As an example, the determination function 553 calculates the similarity between medical image I4 corresponding to time t4 in FIG. 5 and medical image I1, the similarity between medical image I4 and medical image I2, and the similarity between medical image I4 and medical image I3, and determines the time point of the medical image with the lowest calculated similarity as the comparison target. In other words, the determination function 553 determines that the more stable state is the time point at which the medical image with the lowest similarity to medical image I4, which was collected at time t4 when it was suggested that the subject's condition was "worsening."

In the above example, a case was described in which the subject's condition "worsens" and a time point at which the subject is in a stable state is determined as a point of comparison, but the determination function 553 can also perform similar processing when the subject's condition "recovers" and a time point at which the subject is in a worsening state is determined as a point of comparison.

(Variation 2)
In the above embodiment, the control function 551 has been described as displaying a medical image at a time point to be interpreted and a medical image at a time point determined as a comparison target on the display 53. However, the control function 551 can also display various other information on the display 53.

Figure 6 is a diagram showing an example of processing by the control function 551 according to the second modified example. For example, as shown in Figure 6, the control function 551 can display a medical image together with a graph showing the tumor marker value on the vertical axis and the date on the horizontal axis. In such a case, the generation function 555 generates the graph L4 shown in Figure 6 based on the multiple tumor marker values acquired by the control function 551 (the tumor marker value at the time of the image interpretation and past tumor marker values).

Then, the control function 551 causes the display 53 to display display information in which the medical images are associated with the graph L4 generated by the generation function 555. For example, the control function 551 causes the graph to display, in addition to the medical image I7 at the time point to be interpreted, the medical image I6 at the time point to be compared (5/1) and the medical image I5 at the time point when the C treatment was administered (3/1).

Furthermore, the control function 551 can further display various information that is useful for diagnosing the subject. In such a case, the acquisition function 554 acquires information that is useful for diagnosing the subject. Specifically, the acquisition function 554 acquires various information that is useful for diagnosis from the electronic medical record system 2 via the network. For example, the acquisition function 554 acquires the subject's treatment history, the doctor's findings on the subject, and the like as information that is useful for diagnosis.

The control function 551 can display information showing the change in tumor marker over time, the treatment history of the subject, and the findings on the subject, in association with a medical image corresponding to the value of the tumor marker for comparison and a medical image at the time of the subject's condition assessment. For example, as shown in FIG. 6, the control function 551 can display display information that allows the relationship between each treatment and the tumor marker value to be grasped at a glance by displaying the time when treatment A was performed and the time when treatment B was performed in association with a graph. Furthermore, the control function 551 can display the doctor's findings at that time in association, as shown in "Recovery" on "5/1" in FIG. 6.

(Variation 3)
In the above embodiment, a case has been described in which a previous medical image of the same specimen is acquired as a medical image to be compared. However, the image reading device 5 can display information of a different specimen as a comparison object. Specifically, the image reading device 5 can display, as comparative information, information of a specimen whose tumor marker value changes over time similarly to that of the specimen to be read and whose diagnosis has been confirmed by a doctor.

In such a case, the control function 551 extracts subject information that is of a subject different from the subject and in which the change over time in the tumor marker is similar to the change over time in the tumor marker of the subject from the subject information for which a diagnosis has been confirmed, and controls the display 53 to display the extracted subject information.

Figure 7 is a diagram for explaining an example of processing by the control function 551 relating to variant example 3. For example, as shown in Figure 7, the control function 551 compares the transitions in the tumor marker value of patient A with the transitions in the tumor marker value stored in the medical network database 6. Then, the control function 551 obtains medical information of patient X, which shows a transition in the tumor marker value similar to the transition in the tumor marker value of patient A, from the medical network database 6.

Then, the control function 551 displays the acquired medical information of patient X on the display 53 as information to be compared with that of patient A. The medical network database 6 is connected to the network in FIG. 1, and the control function 551 accesses the medical network database 6 via the network to compare the transitions in the tumor marker values and acquire medical information.

Next, the procedure of processing by the image reading device 5 will be described. FIG. 8 is a flowchart showing the procedure of processing by the image reading device 5 according to the first embodiment. Here, steps S101 and S105 in FIG. 8 are realized by the processing circuitry 55 reading a program corresponding to the control function 551 from the storage circuitry 54 and executing it. Also, step S102 is realized by the processing circuitry 55 reading a program corresponding to the determination function 552 from the storage circuitry 54 and executing it. Also, step S103 is realized by the processing circuitry 55 reading a program corresponding to the decision function 553 from the storage circuitry 54 and executing it. Also, step S104 is realized by the processing circuitry 55 reading a program corresponding to the acquisition function 554 from the storage circuitry 54 and executing it.

As shown in FIG. 8, the processing circuitry 55 acquires the tumor marker value (step S101) and determines the state of the subject based on the change over time in the tumor marker value (step S102). Then, the processing circuitry 55 determines the tumor marker value to be compared (the time point to be compared) based on the state of the subject (step S103).

Then, the processing circuitry 55 extracts medical images corresponding to the tumor marker values to be compared (step S104) and displays the extracted medical images for comparison (step S105).

As described above, according to the first embodiment, the judgment function 552 judges the condition of the subject based on the change over time in the tumor marker value. The decision function 553 decides a tumor marker value to be compared from among a plurality of tumor marker values acquired over time, depending on the condition of the subject. The acquisition function 554 acquires a medical image corresponding to the tumor marker value to be compared. Therefore, the image reading device 5 according to the first embodiment can automatically select and display an appropriate medical image according to the condition of the subject, reducing the workload of an operator such as an image reading physician and enabling the quality of the examination to be improved.

Furthermore, according to the first embodiment, the judgment function 552 judges the state of the subject at the time of the judgment target based on the change in the tumor marker value from a time point prior to the time of the judgment target at which the state of the subject is judged to the time of the judgment target. The determination function 553 determines a comparison tumor marker value to be compared with the tumor marker at the time of the judgment target from among a plurality of tumor marker values acquired at a time point prior to the time of the judgment target, according to the state of the subject at the time of the judgment target. The acquisition function 554 acquires a medical image corresponding to the time point at which the comparison tumor marker value was acquired. Therefore, the image reading device 5 according to the first embodiment makes it possible to automatically acquire optimal past images when performing comparative image reading during follow-up observation.

Furthermore, according to the first embodiment, the judgment function 552 judges the condition of each part of the subject based on the type of tumor marker. The determination function 553 determines a tumor marker value to be compared for each type of tumor marker. The acquisition function 554 acquires a medical image corresponding to the tumor marker value to be compared for each part of the subject based on the type of tumor marker. Therefore, the image reading device 5 according to the first embodiment can automatically acquire appropriate medical images for each part targeted by the tumor marker used, making it possible to further improve the quality of the examination.

Furthermore, according to the first embodiment, the acquisition function 554 acquires a medical image of a region by performing a reconstruction process on medical image data collected from a subject. Therefore, the image reading device 5 according to the first embodiment makes it possible to generate and display an appropriate medical image.

Furthermore, according to the first embodiment, the acquisition function 554 acquires a medical image of the region by re-executing the clinical analysis app on the medical image data collected from the subject. Therefore, the image reading device 5 according to the first embodiment makes it possible to generate a medical image suitable for the region.

Furthermore, according to the first embodiment, the acquisition function 554 acquires medical images of the area by controlling the acquisition of images of the area from the subject. Therefore, the image reading device 5 according to the first embodiment makes it possible to prompt the user to acquire new medical images of the target area even if there are no medical images available.

Furthermore, according to the first embodiment, the acquisition function 554 further acquires information that is useful for diagnosing the subject. Therefore, the image reading device 5 according to the first embodiment makes it possible to obtain information that contributes to diagnosis in addition to medical images.

Furthermore, according to the first embodiment, the determination function 553 determines a tumor marker value to be compared from among a plurality of tumor marker values acquired over time based on the tumor marker value. Therefore, the image reading device 5 according to the first embodiment makes it possible to determine a comparison target based on the tumor marker value.

Furthermore, according to the first embodiment, the determination function 553 extracts multiple candidates for a tumor marker value to be compared from multiple tumor marker values acquired over time based on the tumor marker value, and determines a tumor marker value to be compared from the multiple candidates based on the characteristics of each medical image corresponding to the multiple candidates. Therefore, the image reading device 5 according to the first embodiment makes it possible to determine the most optimal time point from multiple time points showing similar tumor marker values.

Furthermore, according to the first embodiment, the determination function 553 determines a tumor marker value to be compared from among multiple tumor marker values acquired over time, based on the tumor marker value and the type of the corresponding medical image. Therefore, the image reading device 5 according to the first embodiment makes it possible to acquire medical images of the same type as a comparison target.

Furthermore, according to the first embodiment, the determination function 553 extracts multiple candidates for a comparison tumor marker value from multiple tumor marker values acquired over time based on the tumor marker value, and determines a comparison tumor marker value from the multiple candidates based on the similarity between each medical image corresponding to the multiple candidates and the medical image at the time of the subject's condition being determined. Therefore, the image reading device 5 according to the first embodiment makes it possible to acquire a comparison medical image based on the similarity of the images.

Furthermore, according to the first embodiment, the control function 551 controls the display 53 to display a medical image corresponding to the tumor marker value to be compared and a medical image at the time of the subject's condition to be determined. Therefore, the image reading device 5 according to the first embodiment makes it possible to automatically display the medical image to be compared.

Furthermore, according to the first embodiment, the control function 551 displays information indicating the change in tumor marker over time and information useful for diagnosing the subject, in association with a medical image corresponding to the tumor marker value to be compared and a medical image at the time of judgment when the subject's condition was judged. Therefore, the image reading device 5 according to the first embodiment can display various information in addition to medical images, making it possible to further improve the quality of examinations.

Furthermore, according to the first embodiment, the control function 551 extracts subject information that is of a subject different from the subject and in which the change over time of the tumor marker is similar to the change over time of the subject's tumor marker from the subject information in which a diagnosis has been confirmed, and controls the display 53 to display the extracted subject information. Therefore, the image reading device 5 according to the first embodiment can display information on a subject in which a diagnosis has been confirmed in a similar state to the subject to be read, making it possible to improve the accuracy of the diagnosis.

Second Embodiment
In the above-described first embodiment, an example in which the tumor marker value increases during follow-up has been described. In the second embodiment, an example in which the tumor marker value decreases during follow-up will be described.

Figure 9 is a diagram for explaining an example of the processing of the image reading device 5 according to the second embodiment. Here, Figure 9 shows an example in which the tumor marker value decreases and stabilizes during follow-up observation.

For example, the flow of cancer treatment is as shown in Figure 9. First, a medical examination and diagnosis are performed on the examined subject. If a cancer diagnosis is made, treatment is then given and the patient is monitored. If recurrence or metastasis occurs, treatment is given again.

Now, suppose that in follow-up observation, in tumor marker tests using tumor markers, the values have declined over the past few tests and fallen below the threshold.

In such cases, based on the results of the tumor marker values, for example, X-ray images are collected by mammography as follow-up observation, and interpretation is performed. The tumor marker values and interpretation results are then integrated to make a diagnosis. In the second embodiment, medical images are automatically selected during interpretation in the above-mentioned process.

That is, as shown in FIG. 9, the judgment function 552 first (1) judges that the current marker value is "below the threshold value". Then, the judgment function 552 judges the subject's condition to be "recovered" because (2) a comparison with past marker values shows a "stable decline" from a "high value" to a "low value (reference value)". Furthermore, the decision function 553 determines the optimal marker value for comparison (optimal time for comparison) based on the judgment result and the correspondence information 542. That is, the decision function 553 (3) determines the (post-treatment) tumor marker value that showed a "high value" as a comparison target for the tumor marker value that showed "recovery".

Then, the acquisition function 554 (4) determines and acquires a medical image corresponding to the tumor marker value that indicates a "high value." That is, the acquisition function 554 narrows down the target examination (date and time) based on the result of the tumor marker value that indicates a "high value," and determines that the image is one in which the patient's condition is deteriorating based on the diagnosis result and examination information of the target examination, and selects the image. If there is no target image, the acquisition function 554 performs image reconstruction, clinical analysis again, transition to another examination, etc. Then, the control function 551 acquires the medical image and (5) displays the acquired medical image and the medical image collected this time. This allows the radiologist to perform comparative interpretation and, for example, make a judgment such as "there is no recurrence and no metastasis to other tissues" based on the difference result of the displayed medical images.

In this way, the image reading device 5 according to the second embodiment also automatically performs the above steps (1) to (4). This enables the image reading device 5 to improve the quality of tests using tumor markers for various cancers.

Other Embodiments
Although the first and second embodiments have been described above, the present invention may be embodied in various different forms other than the first and second embodiments described above.

In the above-described embodiment, the case where the image reading support device according to the present application is applied to the image reading device 5 has been described. However, the embodiment is not limited to this, and may be applied to, for example, any of the devices in the network shown in FIG. 1. In such a case, the processing circuit of the applied device executes processing similar to the processing of the control function 551, judgment function 552, decision function 553, acquisition function 554, and generation function 555 described above.

In the image reading device 5 described in each embodiment, each processing function is stored in the storage circuitry 54 in the form of a program executable by a computer. The processing circuitry 55 is a processor that realizes the function corresponding to each program by reading the program from the storage circuitry 54 and executing it. In other words, the processing circuitry 55 in a state in which each program has been read has the function corresponding to the read program. In each of the above-mentioned embodiments, the processing functions are realized by a single processing circuitry 55, but the embodiment is not limited to this. For example, the processing circuitry 55 may be configured by combining multiple independent processors, and each processor may realize each processing function by executing each program. In addition, each processing function possessed by the processing circuitry 55 may be realized by being appropriately distributed or integrated into a single or multiple processing circuits.

The term "processor" used in the above description refers to circuits such as a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), an Application Specific Integrated Circuit (ASIC), a programmable logic device (e.g., a Simple Programmable Logic Device (SPLD), a Complex Programmable Logic Device (CPLD), and a Field Programmable Gate Array (FPGA)). The processor realizes its functions by reading and executing programs stored in the memory circuits 12 and 12a.

In the above-described embodiments, the memory circuitry 54 has been described as storing a program corresponding to each processing function. However, a configuration may be adopted in which multiple memory circuits are distributed and the processing circuitry 55 reads out the corresponding program from each individual memory circuit. Also, instead of storing the program in the memory circuitry, the program may be directly embedded in the processor circuitry. In this case, the processor realizes the function by reading and executing the program embedded in the circuitry.

The components of each device according to the above-described embodiments are conceptual and functional, and do not necessarily need to be physically configured as shown in the figures. In other words, the specific form of distribution and integration of each device is not limited to that shown in the figures, and all or part of the devices can be functionally or physically distributed and integrated in any unit depending on various loads and usage conditions. Furthermore, all or any part of the processing functions performed by each device can be realized by a CPU and a program analyzed and executed by the CPU, or can be realized as hardware using wired logic.

The image interpretation support method described in the above embodiment can be realized by executing a prepared processing program on a computer such as a personal computer or a workstation. This medical information processing program can be distributed via a network such as the Internet. The medical information processing program can also be recorded on a non-transitory recording medium that can be read by a computer, such as a hard disk, flexible disk (FD), CD-ROM, MO, or DVD, and executed by being read from the recording medium by a computer.

At least one of the embodiments described above can improve the quality of the test.

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, modifications, and combinations of embodiments can be made without departing from the spirit of the invention. These embodiments and their modifications are within the scope of the invention and its equivalents as set forth in the claims, as well as the scope and spirit of the invention.

5 Image reading device 551 Control function 552 Judgment function 553 Decision function 554 Acquisition function 555 Generation function

Claims (15)

a determination unit for determining a state of the subject based on a change in the value of the tumor marker over time;
A determination unit that determines a tumor marker value to be compared from among a plurality of tumor marker values acquired over time in accordance with the condition of the subject;
an acquisition unit for acquiring a medical image corresponding to the value of the tumor marker to be compared;
An image reading support device comprising: The determination unit determines a state of the subject at a time point to be determined based on a change in the value of the tumor marker from a time point prior to the time point to be determined,
the determination unit determines, in accordance with the state of the subject at the time to be determined, a comparative tumor marker value to be compared with the tumor marker at the time to be determined from among a plurality of tumor marker values acquired at a time prior to the time to be determined;
The image interpretation support device according to claim 1 , wherein the acquisition unit acquires a medical image corresponding to a time point at which the value of the tumor marker to be compared was acquired. The determination unit determines a condition of each site of the subject based on the type of the tumor marker,
The determination unit determines the value of the tumor marker to be compared for each type of tumor marker,
The image interpretation support device according to claim 1 , wherein the acquisition unit acquires a medical image corresponding to a value of the tumor marker to be compared for each part of the subject based on the type of the tumor marker. The image interpretation support device according to claim 3, wherein the acquisition unit acquires a medical image of the region by performing a reconstruction process on medical image data collected from the subject. The image interpretation support device according to claim 3, wherein the acquisition unit acquires a medical image of the region by re-executing a clinical analysis on the medical image data collected from the subject. The image interpretation support device according to claim 3, wherein the acquisition unit acquires medical images of the region by controlling the acquisition unit to collect images of the region from the subject. The image interpretation support device according to any one of claims 1 to 6, wherein the acquisition unit further acquires information useful for diagnosing the subject. The image interpretation support device according to any one of claims 1 to 7, wherein the determination unit determines the tumor marker value to be compared from among the multiple tumor marker values acquired over time based on the tumor marker value. The image interpretation support device according to any one of claims 1 to 7, wherein the determination unit extracts multiple candidates for the value of the tumor marker to be compared from the multiple tumor marker values acquired over time based on the tumor marker value, and determines the value of the tumor marker to be compared from the multiple candidates based on the characteristics of each medical image corresponding to the multiple candidates. The image interpretation support device according to any one of claims 1 to 7, wherein the determination unit determines the tumor marker value to be compared from among the multiple tumor marker values acquired over time based on the tumor marker value and the type of the corresponding medical image. The image interpretation support device according to any one of claims 1 to 7, wherein the determination unit extracts multiple candidates for the value of the tumor marker to be compared from the multiple tumor marker values acquired over time based on the tumor marker value, and determines the value of the tumor marker to be compared from the multiple candidates based on the similarity between each medical image corresponding to the multiple candidates and the medical image at the time of the determination target at which the state of the subject was determined. The image interpretation support device according to any one of claims 1 to 11, further comprising a display control unit that controls the display unit to display a medical image corresponding to the value of the tumor marker to be compared and a medical image at the time of the subject's condition being determined. The image interpretation support device according to claim 12, wherein the display control unit displays information indicating the change over time of the tumor marker and information useful for diagnosing the subject, in association with a medical image corresponding to the value of the tumor marker to be compared and a medical image at the time of judgment when the condition of the subject was judged. The image interpretation support device according to claim 12 or 13, wherein the display control unit extracts subject information that is of a subject different from the subject and in which a change over time in a tumor marker is similar to the change over time in the tumor marker of the subject from subject information for which a diagnosis has been confirmed, and controls the display unit to display the extracted subject information. A method for operating a diagnostic reading support device for supporting diagnostic reading of a medical image, comprising the steps of:
The image interpretation support apparatus includes a determination unit, a decision unit, and an acquisition unit,
a determination step in which the determination unit determines a state of the subject based on a change over time in the value of the tumor marker;
a determination step in which the determination unit determines a tumor marker value to be compared from among a plurality of tumor marker values acquired over time in accordance with a state of the subject;
an acquisition step in which the acquisition unit acquires a medical image corresponding to the value of the tumor marker to be compared;
A method for operating an image interpretation support device , comprising :

Images