Progression prediction apparatus, progression prediction method, and progression prediction program
A processor is configured to derive, with reference to a database in which a plurality of reference images are saved such that an interpretation result about an abnormal shadow included in each reference image of the plurality of reference images is associated with the reference image, a degree of similarity between a target image and each of the plurality of reference images; the processor is configured to analyze, for a similar reference image, among the plurality of reference images, for which the degree of similarity is greater than or equal to a predetermined threshold value, an interpretation result regarding the similar reference image to thereby derive progression information about an abnormal shadow included in the similar reference image; and the processor is configured to statistically analyze the progression information to thereby derive prediction information for predicting future progression of an abnormal shadow included in the target image.
1. A progression prediction apparatus comprising:
at least one processor,
the processor being configured to
derive, with reference to a database in which a plurality of reference images are saved such that an interpretation result about an abnormal shadow included in each reference image of the plurality of reference images is associated with the reference image, a degree of similarity between a target image and each of the plurality of reference images,
analyze, for a similar reference image, among the plurality of reference images, for which the degree of similarity is greater than or equal to a predetermined threshold value, an interpretation result regarding the similar reference image to thereby derive progression information about an abnormal shadow included in the similar reference image, and
statistically analyze the progression information to thereby derive prediction information for predicting future progression of an abnormal shadow included in the target image.
2. The progression prediction apparatus according to claim 1 , wherein the prediction information includes at least one of a disease name, a change in size of the abnormal shadow, a change in shape of the abnormal shadow, presence or absence of an appearance of a new abnormal shadow, a location where the new abnormal shadow appears, a name of a complication that is likely to develop, or a period until a next diagnosis.
3. The progression prediction apparatus according to claim 1 , wherein the processor is configured to display the prediction information on a display.
4. The progression prediction apparatus according to claim 2 , wherein the processor is configured to display the prediction information on a display.
5. The progression prediction apparatus according to claim 4 , wherein the processor is configured to, in a case where the prediction information includes the location where the new abnormal shadow appears, display the target image so as to highlight, in the displayed target image, the location where the new abnormal shadow appears.
6. The progression prediction apparatus according to claim 1 , wherein the processor is configured to derive a feature value about the target image and derive the degree of similarity on the basis of the feature value derived for the target image and a feature value about each of the plurality of reference images.
7. The progression prediction apparatus according to claim 1 , wherein
the processor is configured to
extract at least one related similar reference image associated with the same subject as the similar reference image, and
analyze the interpretation result related to the similar reference image and the related similar reference image to thereby derive the progression information.
8. The progression prediction apparatus according to claim 7 , wherein
the similar reference image is associated with a first date, and
the related similar reference image is associated with a second date different from the first date.
9. The progression prediction apparatus according to claim 8 , wherein the second date is a later date relative to the first date.
10. The progression prediction apparatus according to claim 1 , wherein the progression information includes at least one of an imaging interval, a doubling time of an abnormal shadow, a changed shape, a place where a new abnormal shadow appears, and a name of a complication.
11. The progression prediction apparatus according to claim 1 , wherein
the processor is configured to
derive a degree of similarity between a target image group and each of reference image groups, the target image group including the target image and at least one past target image acquired on an imaging date earlier than an imaging date of the target image by imaging a photographic subject the same as a photographic subject for which the target image is acquired, each of the reference image groups including, among the plurality of reference images, a plurality of reference images acquired by imaging a same photographic subject and each having an imaging interval corresponding to imaging intervals of the target image and the at least one past target image, and
analyze, for a similar reference image group for which the degree of similarity is greater than or equal to a predetermined threshold value, interpretation results associated with a plurality of similar reference images included in the similar reference image group to thereby derive progression information about the similar reference image group.
12. The progression prediction apparatus according to claim 1 , wherein
the processor is configured to
obtain a plurality of the progression information, and
statistically analyze the plurality of progression information to thereby derive the prediction information.
13. The progression prediction apparatus according to claim 1 , wherein
the processor is configured to
derive a plurality of the progression information corresponding to the similar reference images respectively, and
derive the prediction information based on the plurality of the progression information.
14. A progression prediction method comprising:
deriving, with reference to a database in which a plurality of reference images are saved such that an interpretation result about an abnormal shadow included in each reference image of the plurality of reference images is associated with the reference image, a degree of similarity between a target image and each of the plurality of reference images;
analyzing, for a similar reference image, among the plurality of reference images, for which the degree of similarity is greater than or equal to a predetermined threshold value, an interpretation result regarding the similar reference image to thereby derive progression information about an abnormal shadow included in the similar reference image; and
statistically analyzing the progression information to thereby derive prediction information for predicting future progression of an abnormal shadow included in the target image.
15. A non-transitory computer recording medium storing a progression prediction program for causing a computer to perform:
deriving, with reference to a database in which a plurality of reference images are saved such that an interpretation result about an abnormal shadow included in each reference image of the plurality of reference images is associated with the reference image, a degree of similarity between a target image and each of the plurality of reference images;
analyzing, for a similar reference image, among the plurality of reference images, for which the degree of similarity is greater than or equal to a predetermined threshold value, an interpretation result regarding the similar reference image to thereby derive progression information about an abnormal shadow included in the similar reference image; and
statistically analyzing the progression information to thereby derive prediction information for predicting future progression of an abnormal shadow included in the target image.