Endoscopic image observation system, endosopic image observation device, and endoscopic image observation method
An endoscopic image observation system supports the observation of a plurality of images captured by a capsule endoscope. The endoscopic image observation system includes a distinguishing unit that outputs an accuracy score indicating the likelihood that each of the plurality of images represents an image of a region sought to be distinguished; a grouping unit that groups the plurality of images into a plurality of clusters in accordance with the accuracy score; and an identification unit that identifies a candidate image for a boundary of the region from among the plurality of images in accordance with the grouping into the plurality of clusters.
1. An endoscopic image observation system for supporting observation of a plurality of images captured by an endoscope comprising a processor comprising hardware, wherein the processor is configured to:
output an accuracy score indicating a likelihood that each of the plurality of images represents an image of a site sought to be distinguished;
group the plurality of images into a plurality of clusters in accordance with the accuracy score; and
identify a candidate image for a boundary of the site from among the plurality of images in accordance with the grouping into the plurality of clusters,
wherein, when there are a plurality of sites sought to be distinguished, the processor is configured to:
output as an accuracy score a vector including the likelihood that each of the plurality of images represents an image of each of the plurality of sites sought to be distinguished as a component;
identify a representative vector representing the accuracy score of one or more grouped images for each of the plurality of clusters; and
identify, from among the plurality of clusters, a predetermined number of clusters in ascending order of the sum of the respective inner products of the representative vector and respective standard basis vector for the plurality of sites sought to be distinguished, as a cluster into which the candidate image is grouped.
2. The endoscopic image observation system according to claim 1 ,
wherein given that the number of sites sought to be distinguished is L, the processor is configured to group the plurality of images into (L+1) clusters or more.
3. The endoscopic image observation system according to claim 1 ,
wherein the vector further includes information on relative time from an imaging start time of each of the plurality of images as a component.
4. The endoscopic image observation system according to claim 1 ,
wherein the processor is configured to output an accuracy score indicating the likelihood that each of the plurality of images represents at least any one of images of a stomach, a small intestine, and a large intestine.
5. The endoscopic image observation system according to claim 1 ,
wherein the processor is configured to display information regarding the candidate image on a display.
6. The endoscopic image observation system according to claim 5 ,
wherein the processor is configured to display the candidate image on a display in a list.
7. The endoscopic image observation system according to claim 5 ,
wherein the processor is configured to display a time bar with one end indicating the imaging start time and the other end indicating the imaging end time and also display an indicator for a candidate image in association with the time bar.
8. The endoscopic image observation system according to claim 5 ,
wherein the processor is configured to display a bar displaying that an image at each imaging time is an image of a predetermined site or the candidate image in association with the imaging time of the image.