Image processing device, image processing method, and image processing program

The image processing device uses trained models to detect and highlight invisible lesions in medical images, improving diagnostic accuracy by reporting their presence when visible detection fails.

JP2024168570AActive Publication Date: 2024-12-05FUJIFILM CORP +1
View PDF 5 Cites 0 Cited by

Patent Information

Application Number
JP2023085371
Authority / Receiving Office
JP · JP
Patent Type
Applications
Current Assignee / Owner
Filing Date
2023-05-24
Publication Date
2024-12-05
Estimated Expiration
2043-05-24

AI Technical Summary

Technical Problem

In medical images where lesions are not visibly depicted, existing technologies fail to effectively inform interpreters about the potential presence of invisible lesions, relying on indirect findings which can be misleading.

Method used

An image processing device and method that uses trained models to derive the probability and visibility of lesions in medical images, reporting the possibility of invisible lesions when probability indicates presence but visibility is low.

Benefits of technology

Appropriately indicates the presence of invisible lesions, enhancing diagnostic support by clearly distinguishing visible and invisible lesion areas.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 2024168570000001_ABST
    Figure 2024168570000001_ABST
Patent Text Reader

Abstract

To provide an image processing device, an image processing method, and an image processing program that enable appropriate presentation of a possibility that there is a lesion that is invisible in a medical image.SOLUTION: An image processing device derives a possibility that there is a lesion in a medical image and a degree of visibility indicating a degree of easiness to view of the lesion in the medical image. When the possibility indicates that there is a lesion in the medical image and the degree of visibility indicates that the lesion is invisible in the medical image, the image processing device issues a notification of information indicating that there is a lesion invisible in the medical image.SELECTED DRAWING: Figure 7
Need to check novelty before this filing date? Find Prior Art

Description

[Technical field]

[0001] The present disclosure relates to an image processing device, an image processing method, and an image processing program. [Background technology]

[0002] Patent document 1 discloses a technology in which a priority is determined based on predetermined conditions for a lesion detection area in a medical image that has been subjected to detection processing for multiple types of lesions, and the display form of the detection area detected in the medical image is changed according to the determined priority. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Patent Publication No. 2021-029387 Summary of the Invention [Problem to be solved by the invention]

[0004] In medical images obtained by photographing a diagnostic target site of a patient having a lesion, the lesion may not be depicted in the medical image or may be depicted only slightly. In such cases, the reader of the medical image judges the presence or absence of a lesion based on changes in the shape and properties of the surrounding tissues that accompany the occurrence of the lesion, even if the reader cannot see the lesion itself. Therefore, in such cases, in order to effectively support the reader, it is preferable to appropriately notify the reader of the possibility that a lesion that cannot be seen in the medical image may exist.

[0005] The present disclosure has been made in consideration of the above circumstances, and aims to provide an image processing device, an image processing method, and an image processing program that can appropriately indicate the possibility of the presence of a lesion that is not visible in a medical image. [Means for solving the problem]

[0006] A first aspect of the image processing device is an image processing device having at least one processor, which derives a probability that a lesion exists in a medical image and a visibility indicating the ease of seeing the lesion in the medical image, and when the probability indicates that a lesion exists in the medical image and the visibility indicates that the lesion is not visible in the medical image, reports information indicating that a lesion that is not visible is present in the medical image.

[0007] In the image processing device of the second aspect, in the image processing device of the first aspect, the processor further controls the display to be different for each of the cases where the probability indicates that a lesion is present in the medical image and the visibility indicates that the lesion is not visible in the medical image, where the probability indicates that a lesion is present in the medical image and the visibility indicates that the lesion is visible in the medical image, and where the probability indicates that a lesion is not present in the medical image.

[0008] The image processing device of the third aspect is the image processing device of the first or second aspect, in which the processor is a trained model that takes a medical image as input and outputs a probability, and derives the probability by inputting a medical image of the diagnosis target to the first trained model that has been trained using multiple sets of medical images and probabilities as training data.

[0009] An image processing device of a fourth aspect is an image processing device of any one of the first to third aspects, in which a processor derives visibility by inputting a medical image to be diagnosed into a second trained model that is a trained model that receives a medical image as input and outputs visibility, and that has been trained using multiple sets of medical images and visibility as training data.

[0010] In a fifth aspect of the image processing device, in the image processing device of the fourth aspect, the processor derives visibility by inputting the area portion in the medical image in which the lesion is determined to exist into a second trained model when the probability indicates that a lesion is present in the medical image.

[0011] An image processing device of a sixth aspect is the image processing device of the first or second aspect, in which the processor derives the probability based on one or more indirect findings in the medical image.

[0012] A seventh aspect of the image processing device is the image processing device of the sixth aspect, wherein the processor derives a first probability based on one or more indirect findings, derives a second probability by performing a lesion detection process on the medical image, derives a third probability based on the first probability and the second probability, and if the third probability indicates that a lesion is present in the medical image and the visibility indicates that the lesion is not visible in the medical image, reports information indicating that a lesion that is not visible is present in the medical image.

[0013] The image processing method of the eighth aspect derives the probability that a lesion exists in a medical image and a visibility indicating the ease of seeing the lesion in the medical image, and if the probability indicates that a lesion exists in the medical image and the visibility indicates that the lesion is not visible in the medical image, a processor provided in the image processing device executes a process to notify information indicating that a lesion that is not visible exists in the medical image.

[0014] The image processing program of the ninth aspect is intended to cause a processor provided in an image processing device to derive the probability that a lesion exists in a medical image and a visibility indicating the ease of seeing the lesion in the medical image, and if the probability indicates that a lesion exists in the medical image and the visibility indicates that the lesion is not visible in the medical image, to execute a process of notifying information indicating that a lesion that is not visible is present in the medical image. Effect of the Invention

[0015] According to the present disclosure, it is possible to appropriately present the possibility that a lesion that cannot be visually recognized is present in a medical image. [Brief description of the drawings]

[0016] [Figure 1] FIG. 1 is a block diagram showing a schematic configuration of a medical information system. [Diagram 2] FIG. 2 is a block diagram showing an example of a hardware configuration of the image processing device. [Diagram 3] FIG. 2 is a block diagram showing an example of a functional configuration of the image processing device. [Figure 4] FIG. 1 is a diagram for explaining a first trained model. [Diagram 5] FIG. 13 is a diagram showing an example of a region in which a lesion is determined to exist. [Figure 6] FIG. 13 is a diagram for explaining a second trained model. [Figure 7] FIG. 4 is a diagram showing an example of a display screen. [Figure 8] 13 is a flowchart illustrating an example of a diagnosis support process. [Figure 9] 13 is a diagram for explaining a process for deriving the probability that a lesion exists from an indirect finding according to a modified example. FIG. [Figure 10] 13A and 13B are diagrams for explaining a process for deriving the probability that a lesion exists based on indirect findings and lesion detection processing according to a modified example. [Figure 11] FIG. 13 is a diagram for explaining a second trained model according to a modified example. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0017] Hereinafter, examples of embodiments for carrying out the technology of the present disclosure will be described in detail with reference to the drawings.

[0018] First, the configuration of a medical information system 1 according to this embodiment will be described with reference to Fig. 1. As shown in Fig. 1, the medical information system 1 includes an image processing device 10, an imaging device 12, and an image storage server 14. The image processing device 10, the imaging device 12, and the image storage server 14 are connected in a state in which they can communicate with each other via a wired or wireless network 18. The image processing device 10 is, for example, a computer such as a personal computer or a server computer.

[0019] The imaging device 12 is a device that generates a medical image representing a diagnostic target part of a subject by imaging the diagnostic target part. Examples of the imaging device 12 include a plain X-ray imaging device, an endoscope device, a CT (Computed Tomography) device, an MRI (Magnetic Resonance Imaging) device, and a PET (Positron Emission Tomography) device. In this embodiment, an example will be described in which the imaging device 12 is a CT device and the diagnostic target part is the abdomen. That is, the imaging device 12 according to this embodiment generates a CT image of the abdomen of the subject as a three-dimensional medical image consisting of multiple tomographic images. The medical image generated by the imaging device 12 is transmitted to the image storage server 14 via the network 18 and is stored by the image storage server 14.

[0020] The image storage server 14 is a computer that stores and manages various data, and is equipped with a large-capacity external storage device and database management software. The image storage server 14 receives medical images generated by the imaging device 12 via the network 18, and stores and manages the received medical images. The storage format of image data by the image storage server 14 and communication with other devices via the network 18 are based on protocols such as DICOM (Digital Imaging and Communication in Medicine).

[0021] Next, a hardware configuration of the image processing device 10 according to the present embodiment will be described with reference to Fig. 2. As shown in Fig. 2, the image processing device 10 includes a CPU (Central Processing Unit) 20, a memory 21 as a temporary storage area, and a non-volatile storage unit 22. The image processing device 10 also includes a display 23 such as a liquid crystal display, an input device 24 such as a keyboard and a mouse, and a network I / F (InterFace) 25 connected to a network 18. The CPU 20, the memory 21, the storage unit 22, the display 23, the input device 24, and the network I / F 25 are connected to a bus 27. The CPU 20 is an example of a processor according to the disclosed technology.

[0022] The storage unit 22 is realized by a hard disk drive (HDD), a solid state drive (SSD), a flash memory, or the like. The storage unit 22 as a storage medium stores an image processing program 30. The CPU 20 reads the image processing program 30 from the storage unit 22, expands it in the memory 21, and executes the expanded image processing program 30.

[0023] The storage unit 22 also stores a trained model 32 and a trained model 34. The trained model 32 and the trained model 34 will be described in detail later.

[0024] In a medical image obtained by photographing a diagnostic target site of a patient having a lesion, the lesion may not be depicted in the medical image or may be depicted only slightly. In this case, a radiologist or other image reader judges whether or not a lesion has occurred based on indirect findings associated with the occurrence of the lesion even if the lesion itself cannot be visually recognized. In this embodiment, the lesion means an abnormal region that is a target of direct treatment, such as cancer, cyst, and inflammation. In this embodiment, the indirect findings mean findings that represent at least one of the characteristics of the shape and properties of the tissue surrounding the lesion associated with the occurrence of the lesion. For example, even if the pancreatic cancer cannot be visually recognized in the pancreas region of an abdominal CT image, the image reader suspects the occurrence of pancreatic cancer due to the dilation of the pancreatic duct, which is an example of an indirect finding.

[0025] The image processing device 10 according to the present embodiment has a function of notifying the image reader of the possibility that a lesion that cannot be visually recognized may exist in the medical image, in order to effectively support the image reader.

[0026] Next, a functional configuration of the image processing device 10 according to the present embodiment will be described with reference to Fig. 3. As shown in Fig. 3, the image processing device 10 includes an acquisition unit 40, a first derivation unit 42, a second derivation unit 44, and a display control unit 46. The CPU 20 executes the image processing program 30 to function as the acquisition unit 40, the first derivation unit 42, the second derivation unit 44, and the display control unit 46.

[0027] The acquisition unit 40 acquires medical images of a diagnosis target (hereinafter, referred to as “diagnosis target images”) from the image storage server 14 via the network I / F 25.

[0028] The first derivation unit 42 derives the probability that a lesion exists in the diagnostic target image acquired by the acquisition unit 40 (hereinafter, referred to as "lesion existence probability") using the trained model 32. The trained model 32 is a trained model that receives a medical image as an input and outputs a lesion existence probability, and is trained using a plurality of sets of medical images and the lesion existence probabilities in the medical images as training data (sometimes referred to as teacher data). The trained model 32 is configured, for example, by a CNN (Convolutional Neural Network). The trained model 32 according to this embodiment outputs, as the lesion existence probability, "lesion present" which indicates the presence of a lesion or "lesion absent" which indicates the absence of a lesion. The trained model 32 is an example of a first trained model according to the disclosed technology.

[0029] 4, the first derivation unit 42 inputs a diagnostic target image acquired by the acquisition unit 40 to the trained model 32. The trained model 32 outputs a lesion existence probability corresponding to the input diagnostic target image. In this way, the first derivation unit 42 derives the lesion existence probability.

[0030] The trained model 32 may output a numerical value between 0 and 1. In this case, the first derivation unit 42 may determine that "lesion is present" when the output of the trained model 32 is equal to or greater than a threshold value (e.g., 0.5), and may determine that "lesion is not present" when the output is less than the threshold value.

[0031] Also, as shown in FIG. 5, when the lesion existence probability output from the trained model 32 is "lesion present", the first derivation unit 42 specifies an area portion in the diagnostic target image in which it is determined that a lesion exists. Specifically, the first derivation unit 42 specifies an area in which a contribution rate is equal to or greater than a certain value in the case where the trained model 32 outputs "lesion present" as an area portion in the diagnostic target image in which it is determined that a lesion exists. The contribution rate in this case can be calculated by a known technique such as Gradient-weighted Class Activation Mapping (Grad-CAM). The example in FIG. 5 shows an example in which it is determined that a lesion exists in the area inside the dashed line, that is, the head of the pancreas and its surrounding area.

[0032] The second derivation unit 44 uses the trained model 34 to derive a visibility indicating the visibility of a lesion in the diagnostic target image acquired by the acquisition unit 40 (hereinafter referred to as "lesion visibility"). The trained model 34 is a trained model that receives a medical image as an input and outputs a lesion visibility, and is trained using a plurality of sets of medical images and the lesion visibility in the medical images as training data. The trained model 34 is configured by, for example, CNN. The trained model 34 according to this embodiment outputs, as the lesion visibility, "visible" indicating that the lesion is visible or "not visible" indicating that the lesion is not visible. The trained model 34 is an example of a second trained model according to the disclosed technology.

[0033] 6, the second derivation unit 44 inputs a region portion in the diagnostic target image identified by the first derivation unit 42, in which a lesion is determined to exist, to the trained model 34. The trained model 34 outputs a lesion visibility corresponding to the input region portion. In this way, the second derivation unit 44 derives the lesion visibility.

[0034] The second derivation unit 44 may output a numerical value between 0 and 1. In this case, the second derivation unit 44 may determine that the output of the trained model 34 is “visible” when the output is equal to or greater than a threshold value (e.g., 0.5), and may determine that the output is “not visible” when the output is less than the threshold value.

[0035] When the lesion presence probability derived by the first derivation unit 42 is "lesion present" and the lesion visibility derived by the second derivation unit 44 is "not visible," the display control unit 46 notifies information indicating the presence of an unvisible lesion in the diagnostic target image. Specifically, in this case, as shown in FIG. 7 as an example, the display control unit 46 controls the display 23 to display the diagnostic target image, information indicating a region in the diagnostic target image identified by the first derivation unit 42 in which a lesion is determined to exist, and a message indicating that an unvisible lesion may exist in the diagnostic target image. In the example of FIG. 7, the outline of the region in the diagnostic target image identified by the first derivation unit 42 in which a lesion is determined to exist is indicated by a dashed line.

[0036] The display control unit 46 may perform control to display the image in different manners in each of the following cases (1), (2), and (3). (1) When the lesion presence probability is "lesion present" and the lesion visibility is "not visible" (2) When the lesion presence probability is "lesion present" and the lesion visibility is "visible" (3) When the probability of lesion existence is "no lesion"

[0037] In this case, in (1), the display control unit 46 performs control to display on the display 23 the diagnostic target image, information representing a region in the diagnostic target image identified by the first derivation unit 42 in which a lesion is determined to exist, and a message indicating that a lesion that cannot be visually recognized may exist in the diagnostic target image, as described above. Also, in this case, in (2), the display control unit 46 performs control to display on the display 23 the diagnostic target image, and information representing a region in the diagnostic target image identified by the first derivation unit 42 in which a lesion is determined to exist. Also, in this case, in (3), the display control unit 46 performs control to display on the display 23 the diagnostic target image and a message indicating that a lesion is highly likely not present. In this way, the display control unit 46 performs control to distinguishably display each of the cases (1) to (3), so that the image reader can easily grasp the presence or absence of a lesion and whether the lesion is visually recognized or not.

[0038] In this embodiment, the display control unit 46 may make the lesion visibility distinguishable by varying the type, thickness, or color of the contour line of the area in the diagnostic image where it is determined that a lesion exists, depending on the lesion visibility. The display control unit 46 may make the lesion visibility distinguishable by controlling the display of a mark assigned in advance for each lesion visibility in the vicinity of the area where it is determined that a lesion exists. The display control unit 46 may also perform control to display the numerical value on the display 23 when the lesion visibility is a numerical value.

[0039] Furthermore, the display control unit 46 may fill in with a preset color an area portion in which a lesion is determined to exist in the diagnostic target image identified by the first derivation unit 42. Furthermore, the display control unit 46 may not display information representing an area portion in which the lesion visibility is "visible."

[0040] Next, the operation of the image processing device 10 according to this embodiment will be described with reference to Fig. 8. The CPU 20 executes the image processing program 30 to execute the diagnostic support process shown in Fig. 8. The diagnostic support process shown in Fig. 8 is executed, for example, when a command to start execution is input by a user.

[0041] In step S10 of Fig. 8, the acquisition unit 40 acquires a diagnostic target image from the image storage server 14 via the network I / F 25. In step S12, the first derivation unit 42 derives a lesion existence probability of the diagnostic target image acquired in step S10 using the trained model 32, as described above. In step S14, the first derivation unit 42 determines whether the lesion existence probability derived in step S12 is "lesion present." If this determination is positive, the process proceeds to step S16.

[0042] In step S16, the first derivation unit 42 identifies an area portion in the diagnostic target image acquired in step S10 in which a lesion is determined to exist, as described above. In step S18, the second derivation unit 44 derives a lesion visibility of the diagnostic target image acquired in step S10 using the trained model 34, as described above.

[0043] In step S20, the display control unit 46 judges whether or not the lesion visibility derived in step S18 is "not visible". If this judgment is a negative judgment, the process proceeds to step S24, and if this judgment is a positive judgment, the process proceeds to step S22. In step S22, the display control unit 46 controls the display 23 to display the diagnostic target image, information representing the area portion identified in step S16, and a message indicating that there may be an unvisible lesion in the diagnostic target image. When the process of step S22 ends, the diagnostic support process ends. In step S24, the display control unit 46 controls the display 23 to display the diagnostic target image and information representing the area portion identified in step S16. When the process of step S24 ends, the diagnostic support process ends.

[0044] If the determination in step S14 is a negative determination, the process proceeds to step S26. In step S26, the display control unit 46 performs control to display the diagnostic target image and a message indicating that there is a high possibility that a lesion does not exist on the display 23. When the process in step S26 ends, the diagnostic support process ends.

[0045] As described above, according to this embodiment, it is possible to appropriately present the possibility that a lesion that cannot be visually recognized is present in a medical image.

[0046] In the above embodiment, the first derivation unit 42 derives the lesion existence probability using the trained model 32, but the present invention is not limited to this. As an example, as shown in FIG. 9, the first derivation unit 42 may derive the lesion existence probability based on one or more indirect findings in the diagnostic target image. In the example of FIG. 9, the first derivation unit 42 derives "lesion present" as the lesion existence probability based on two indirect findings, namely, swelling occurring in the head of the pancreas and atrophy occurring in the tail of the pancreas. In this case, the presence of cancer in the pancreas is suspected. In this case, the first derivation unit 42 may derive the lesion existence probability using a lookup table in which one or more indirect findings are associated with the lesion name.

[0047] In this case, the first derivation unit 42 may derive the indirect findings using a trained model that takes medical images as input and outputs information representing indirect findings, and that has been trained using multiple sets of medical images and information representing indirect findings present in the medical images as training data.

[0048] As an example, as shown in FIG. 10, the first derivation unit 42 may derive a first lesion existence probability based on one or more indirect findings in the diagnostic target image, and derive a second lesion existence probability by executing a lesion detection process on the diagnostic target image. The lesion detection process in this case may be a process using the trained model 32 as in the above embodiment, or a process using a known lesion detection algorithm. In this case, the first derivation unit 42 may derive a third lesion existence probability based on the first lesion existence probability and the second lesion existence probability. For example, when at least one of the first lesion existence probability and the second lesion existence probability is "lesion present", the first derivation unit 42 derives "lesion present" as the third lesion existence probability. In addition, when both the first lesion existence probability and the second lesion existence probability are "lesion absent", the first derivation unit 42 derives "lesion absent" as the third lesion existence probability. In this case, the display control unit 46 may notify information indicating that an unvisible lesion is present in the diagnostic target image when the third lesion presence probability is "lesion present" and the lesion visibility is "not visible." In the example of Fig. 10, the area inside the dashed line indicates the area in which it is determined that a lesion detected by the lesion detection process is present.

[0049] In this case, the display control unit 46 may further perform control to distinguish between the parts of the diagnostic image that have been determined to have a lesion based on indirect findings and the parts that have been determined to have a lesion based on the lesion detection process.

[0050] In the above embodiment, the second derivation unit 44 derives the lesion visibility by inputting a region portion in the diagnostic target image identified by the first derivation unit 42 in which a lesion is determined to exist, to the trained model 34. However, the present invention is not limited to this. As an example, as shown in FIG. 11, the second derivation unit 44 may derive the lesion visibility by inputting the same image as the input image to the trained model 32 (i.e., the diagnostic target image) to the trained model 34.

[0051] In the above embodiment, the second derivation unit 44 derives the lesion visibility in the diagnostic target image using the trained model 34, but the present invention is not limited to this. The second derivation unit 44 may derive the lesion visibility based on the variance of the CT value, the difference in the average value of the CT value from the surrounding area, etc.

[0052] In the above embodiment, the trained model 32 and the trained model 34 are configured by CNN, but the present invention is not limited to this. The trained model 32 and the trained model 34 may be configured by a machine learning method other than CNN.

[0053] In the above embodiment, the case where a CT image is applied as a diagnostic target image is described, but the present invention is not limited to this. As a diagnostic target image, a medical image other than a CT image, such as a radiological image taken by a plain X-ray imaging device and an MRI image taken by an MRI device, may be applied.

[0054] Furthermore, the processes of steps S10 to S12 of the diagnosis support process according to the above embodiment may be executed before the user inputs an instruction to start execution. In this case, when the user inputs an instruction to start execution, steps S14 and after are executed and the screen is displayed.

[0055] In the above embodiment, for example, the hardware structure of the processing unit that executes various processes, such as the acquisition unit 40, the first derivation unit 42, the second derivation unit 44, and the display control unit 46, may be the various processors shown below. As described above, the various processors include a CPU, which is a general-purpose processor that executes software (programs) and functions as various processing units, as well as a programmable logic device (PLD), which is a processor whose circuit configuration can be changed after manufacture, such as an FPGA (Field Programmable Gate Array), a dedicated electric circuit, which is a processor having a circuit configuration designed specifically for executing specific processes, such as an ASIC (Application Specific Integrated Circuit), etc.

[0056] A single processing unit may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, multiple processing units may be configured with a single processor.

[0057] As an example of configuring multiple processing units with one processor, first, there is a form in which one processor is configured with a combination of one or more CPUs and software, as typified by computers such as client and server, and this processor functions as multiple processing units. Second, there is a form in which a processor is used that realizes the functions of the entire system including multiple processing units with one IC (Integrated Circuit) chip, as typified by System on Chip (SoC), etc. In this way, the various processing units are configured using one or more of the above various processors as a hardware structure.

[0058] Furthermore, more specifically, the hardware structure of these various processors can be an electric circuit that combines circuit elements such as semiconductor elements.

[0059] In the above embodiment, the image processing program 30 is stored (installed) in advance in the storage unit 22, but the present invention is not limited to this. The image processing program 30 may be provided in a form recorded on a recording medium such as a CD-ROM (Compact Disc Read Only Memory), a DVD-ROM (Digital Versatile Disc Read Only Memory), or a USB (Universal Serial Bus) memory. The image processing program 30 may be downloaded from an external device via a network. [Explanation of symbols]

[0060] 1 Medical Information System 10 Image processing device 12 Imaging Device 14 Image storage server 18 Network 20 CPU 21 Memory 22 Memory section 23 Display 24 Input Devices 25 Network I / F 27 Bus 30 Image Processing Programs 32, 34 Pre-trained models 40 Acquisition Department 42 First derivation part 44 Second derivation part 46 Display control section

Claims

1. An image processing device comprising at least one processor, The processor, Deriving a probability that a lesion exists in a medical image and a visibility index that indicates the ease of seeing the lesion in the medical image; When the probability indicates that the lesion is present in the medical image and the visibility indicates that the lesion is not visible in the medical image, information indicating that a lesion that is not visible in the medical image is present is notified. Image processing device.

2. The processor further comprises: the probability represents the presence of the lesion in the medical image and the visibility represents the lesion is not visible in the medical image; the probability indicates that the lesion is present in the medical image and the visibility indicates that the lesion is visible in the medical image; and the probability represents that the lesion is not present in the medical image; and In each of the above cases, the display is controlled to be different. The image processing device according to claim 1 .

3. The processor, A first trained model is a trained model that uses the medical image as an input and outputs the probability, and the probability is derived by inputting a medical image of a diagnosis target to a first trained model that has been trained using a plurality of sets of the medical images and the probability as training data.

3. The image processing device according to claim 1 or 2.

4. The processor, A second trained model is a trained model in which the medical image is input and the visibility is output, the second trained model being trained using a plurality of sets of the medical images and the visibility as training data, and the visibility is derived by inputting a medical image of a diagnosis target.

3. The image processing device according to claim 1 or 2.

5. The processor, If the probability indicates that the lesion is present in the medical image, the visibility is derived by inputting the area portion of the medical image in which the lesion is determined to be present to the second trained model. The image processing device according to claim 4.

6. The processor, Deriving the probability based on one or more indirect findings in the medical image.

3. The image processing device according to claim 1 or 2.

7. The processor, deriving a first said probability based on the one or more indirect findings; deriving a second said probability by performing a lesion detection process on the medical image; deriving a third probability based on the first probability and the second probability; When the third probability indicates that the lesion is present in the medical image and the visibility indicates that the lesion is not visible in the medical image, information indicating that a lesion that is not visible in the medical image is present is notified. The image processing device according to claim 6.

8. Deriving a probability that a lesion exists in a medical image and a visibility index that indicates the ease of seeing the lesion in the medical image; When the probability indicates that the lesion is present in the medical image and the visibility indicates that the lesion is not visible in the medical image, information indicating that a lesion that is not visible in the medical image is present is notified. An image processing method in which processing is executed by a processor provided in the image processing device.

9. Deriving a probability that a lesion exists in a medical image and a visibility index that indicates the ease of seeing the lesion in the medical image; When the probability indicates that the lesion is present in the medical image and the visibility indicates that the lesion is not visible in the medical image, information indicating that a lesion that is not visible in the medical image is present is notified. An image processing program for causing a processor included in the image processing device to execute the processing.

Citation Information

Patent Citations

  • Medical image processing device, tomographic device, medical image processing method and program

    JP2021058272A

  • Medical image diagnostic predictor

    US20240215938A1

  • Image processing device, operation method for image processing device, and operation program for image processing device

    WO2017203701A1

  • Medical image processing device, method for operating medical image processing device, and operation program for medical image processing device

    WO2022209889A1

  • Medical information processing device and program

    JP2021029387A