Systems and methods for identifying images of polyps

A method for identifying images including polyps, comprising:

• accessing a plurality of images of a gastrointestinal tract (GIT) captured by a capsule endoscopy device during a CE procedure, wherein:
• each image of the plurality of images is suspected to include a polyp and is associated with a probability of containing the polyp, and
• the plurality of images includes seed images, each seed image is associated with one or more images of the plurality of images, the one or more images associated with each seed image are identified as suspected to include the same polyp as the associated seed image;
• applying a polyp detection system on the seed images to identify seed images which include polyps, wherein the polyp detection system is applied to each seed image of the seed images based on the one or more images associated with the respective seed image and the probabilities associated with the respective seed image and with the one or more associated images associated with the respective seed image;
• displaying at least the seed images to a user;
• receiving user selections of images among the displayed images;
• determining at least one unselected image, which was not selected by the user and which is among the seed images identified to include polyps;
• and
• presenting the at least one unselected image to the user.

A computer-implemented method for recommending a colonoscopy, comprising:

• accessing a plurality of images of a gastrointestinal tract (GIT) captured by a capsule endoscopy device, the plurality of images having a likelihood of containing a polyp;
• for each image of the plurality of images:
• applying a classical machine learning system configured to provide an indication, based on input features corresponding to the image, of whether the image contains a polyp or does not contain a polyp, and
• accessing a soft margin of the classical machine learning system corresponding to the image;
• accessing a mapping of soft margins to probabilities of an image containing a polyp;
• and
• determining, without human intervention, whether to recommend a colonoscopy based on the soft margins of the plurality of images,
• wherein the determining of whether to recommend a colonoscopy is further based on the mapping of soft margins to probabilities of an image containing a polyp.

A system for recommending a colonoscopy, comprising:

• one or more processors;
• and
• at least one memory storing instructions which, when executed by the one or more processors, cause the system to:
• access a plurality of images of a gastrointestinal tract (GIT) captured by a capsule endoscopy device, the plurality of images having a likelihood of containing a polyp;
• for each image of the plurality of images:
• apply a classical machine learning system configured to provide an indication, based on input features corresponding to the image, of whether the image contains a polyp or does not contain a polyp, and
• access a soft margin of the classical machine learning system corresponding to the image;
• access a mapping of soft margins to probabilities of an image containing a polyp;
• and
• determine, without human intervention, whether to recommend a colonoscopy based on the soft margins of the plurality of images,
• wherein the determining of whether to recommend a colonoscopy is further based on the mapping of soft margins to probabilities of an image containing a polyp.