# US 2021/0279920 A1 · Similarity determination apparatus, similarity determination method, and similarity determination program > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 2021/0279920 A1 - Document ID: US20210279920A1 - Status: Published application - Kind: A1 - Application number: 17/327,772 - Filed: 2021-05-24 - Published: 2021-09-09 - Earliest claimed: 2018-11-27 - Assignees: FUJIFILM Corporation - Applicants: FUJIFILM Corporation - Inventors: Shoji KANADA - Web page: https://patentgrep.com/patent/US20210279920A1 ## CPC - G06V 10/82: using neural networks - A61B 6/03: Computed tomography [CT] - G06F 18/22: Matching criteria, e.g. proximity measures - G06T 7/00: Image analysis - G06T 12/10: Image preprocessing, e.g. calibration, positioning of sources or scatter correction - G06T 12/30: Image post-processing, e.g. metal artefact correction - G06V 10/764: using classification, e.g. of video objects - G06T 2207/10076: 4D tomography; Time-sequential 3D tomography - G06T 2210/41: Medical - G06V 10/759: {Region-based matching} - G06V 2201/03: Recognition of patterns in medical or anatomical images - G06V 2201/12: Acquisition of 3D measurements of objects - G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING - A61B: DIAGNOSIS; SURGERY; IDENTIFICATION (usefulness limited to only animals A61D) - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) - G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL ## Abstract A display control unit displays a tomographic image of a specific tomographic plane in a first medical image on a display unit. A finding classification unit classifies each pixel of a partial region of the first medical image into at least one finding. A feature amount calculation unit calculates a first feature amount for each finding in the partial region. A weighting coefficient setting unit sets a weighting coefficient indicating a degree of weighting, which varies depending on a size of each finding, for each finding. A similarity derivation unit performs a weighting operation for the first feature amount for each finding calculated in the partial region and a second feature amount for each finding calculated in a second medical image on the basis of the weighting coefficient to derive a similarity between the first medical image and the second medical image. ## Claims ### Claim 1 A similarity determination apparatus that determines a similarity between a first three-dimensional medical image and a second three-dimensional medical image, the similarity determination apparatus comprising at least one processor, wherein the processor is configured to: display a tomographic image of a specific tomographic plane in the first medical image on a display; classify each pixel of a partial region including at least the specific tomographic plane in the first medical image into at least one of a plurality of types of findings; calculate a first feature amount for each finding classified in the partial region; set a weighting coefficient indicating a degree of weighting, which varies depending on a size of each finding, for each finding classified in the partial region; and perform a weighting operation for the first feature amount for each finding calculated in the partial region and a second feature amount for each finding calculated in the second medical image on the basis of the weighting coefficient to derive the similarity between the first medical image and the second medical image. ### Claim 2 (depends on 1) The similarity determination apparatus according to claim 1, wherein the processor is configured to derive region similarities between the partial region and each of a plurality of small regions in the second medical image and determine a representative similarity among the plurality of region similarities as the similarity between the first medical image and the second medical image. ### Claim 3 (depends on 2) The similarity determination apparatus according to claim 2, wherein the processor is configured to increase a region similarity for a small region that is positionally closer to the partial region among the plurality of small regions and determine the representative similarity. ### Claim 4 (depends on 2) The similarity determination apparatus according to claim 2, wherein the second medical image is divided into the plurality of small regions having overlapping regions. ### Claim 5 (depends on 1) The similarity determination apparatus according to claim 1, wherein the first medical image is divided into a plurality of small regions, the first feature amount is calculated in each of the plurality of small regions, and the weighting coefficient is set in each of the plurality of small regions, and the processor is configured to use the small region including the specific tomographic plane as the partial region and perform the weighting operation for the first feature amount for each finding calculated in the partial region and the second feature amount for each finding calculated in the second medical image on the basis of the weighting coefficient to derive the similarity between the first medical image and the second medical image. ### Claim 6 (depends on 1) The similarity determination apparatus according to claim 1, wherein the processor is configured to include a discriminator that has been subjected to machine learning so as to classify the plurality of types of findings and classify each pixel of the partial region into the plurality of types of findings using the discriminator. ### Claim 7 (depends on 1) The similarity determination apparatus according to claim 1, wherein the processor is configured to: classify each pixel of the first medical image into at least one of the plurality of types of findings; calculate a third feature amount for each finding classified in the first medical image; set, as a first weighting coefficient, the weighting coefficient indicating the degree of weighting, which varies depending on the size of each finding, for each finding classified in the partial region and set, as a second weighting coefficient, a weighting coefficient indicating a degree of weighting, which varies depending on the size of each finding, for each finding classified in the first medical image, and perform the weighting operation for the first feature amount for each finding calculated in the partial region and the second feature amount for each finding calculated in the second medical image on the basis of the first weighting coefficient to derive the similarity between the first medical image and the second medical image as a first similarity, perform a weighting operation for the third feature amount for each finding calculated in the first medical image and a fourth feature amount for each finding calculated in the second medical image on the basis of the second weighting coefficient to derive a similarity between the first medical image and the second medical image as a second similarity, and derive a final similarity between the first medical image and the second medical image on the basis of the first similarity and the second similarity. ### Claim 8 (depends on 7) The similarity determination apparatus according to claim 7, wherein the processor is configured to weight and add the first similarity and the second similarity to derive the final similarity. ### Claim 9 (depends on 8) The similarity determination apparatus according to claim 8, wherein the processor is capable of changing a weighting coefficient in a case in which the first similarity and the second similarity are weighted and added. ### Claim 10 (depends on 1) The similarity determination apparatus according to claim 1, wherein the processor is configured to search for the second medical image similar to the first medical image as a similar medical image on the basis of similarities between the first medical image and a plurality of second medical images with reference to a case database in which the plurality of second medical images are registered and the second feature amount for each of the plurality of second medical images is registered so as to be associated with each of the plurality of second medical images. ### Claim 11 (depends on 7) The similarity determination apparatus according to claim 7, wherein the processor is further configured to search for the second medical image similar to the first medical image as a similar medical image on the basis of the final similarities between the first medical image and a plurality of second medical images with reference to a case database in which the plurality of second medical images are registered and the second and fourth feature amounts for each of the plurality of second medical images are registered so as to be associated with each of the plurality of second medical images. ### Claim 12 (depends on 10) The similarity determination apparatus according to claim 10, wherein, in a case in which the tomographic image of the specific tomographic plane is displayed for a predetermined time, the processor performs the search. ### Claim 13 (depends on 10) The similarity determination apparatus according to claim 10, wherein the processor is configured to display a search result of the similar medical image on the display. ### Claim 14 A similarity determination method that determines a similarity between a first three-dimensional medical image and a second three-dimensional medical image, the similarity determination method comprising: displaying a tomographic image of a specific tomographic plane in the first medical image on a display; classifying each pixel of a partial region including at least the specific tomographic plane in the first medical image into at least one of a plurality of types of findings; calculating a first feature amount for each finding classified in the partial region; setting a weighting coefficient indicating a degree of weighting, which varies depending on a size of each finding, for each finding classified in the partial region; and performing a weighting operation for the first feature amount for each finding calculated in the partial region and a second feature amount for each finding calculated in the second medical image on the basis of the weighting coefficient to derive the similarity between the first medical image and the second medical image. ### Claim 15 A non-transitory computer-readable storage medium that stores a similarity determination program that causes a computer to perform a process of determining a similarity between a first three-dimensional medical image and a second three-dimensional medical image, the similarity determination program causing the computer to perform: a step of displaying a tomographic image of a specific tomographic plane in the first medical image on a display; a step of classifying each pixel of a partial region including at least the specific tomographic plane in the first medical image into at least one of a plurality of types of findings; a step of calculating a first feature amount for each finding classified in the partial region; a step of setting a weighting coefficient indicating a degree of weighting, which varies depending on a size of each finding, for each finding classified in the partial region; and a step of performing a weighting operation for the first feature amount for each finding calculated in the partial region and a second feature amount for each finding calculated in the second medical image on the basis of the weighting coefficient to derive the similarity between the first medical image and the second medical image.