Systems and methods for identifying images containing indicators of a celiac-like disease

1 . A method for detecting indicators of a disease characterized by a presence of villous atrophy in images of a gastrointestinal tract (GIT), the method comprising:

accessing a consecutive set of images of at least a portion of the GIT comprising a small bowel comprising a proximal portion and a remaining portion, wherein each image of the consecutive set of images is associated with one or more classification scores and each classification score of the one or more classification scores is indicative of the respective associated image including a respective indicator of a disease characterized by the presence of villous atrophy;

selecting a subset of images, from the consecutive set of images, based on the one or more classification scores of each image of the consecutive set of images;

identifying, from the consecutive set of images, a segment of images which includes all of the images that show the proximal portion of the small bowel;

selecting a plurality of images, from the identified segment of images, that represent the proximal portion of the small bowel, wherein the selecting the plurality of images from the segment of images that represents the proximal portion of the small bowel includes selecting images spaced across the proximal portion of the small bowel;

selecting a second plurality of images, from the consecutive set of images, that represent the remaining portion of the small bowel;

displaying the plurality of images selected from the identified segment of images, the second plurality of images, and the subset of images on a display;

selecting a first subset of images from the images that show the proximal portion of the small bowel based on application of a first set of rules, at least one rule of the first set of rules based on the one or more classification scores associated with each image; and

selecting a second subset of images from the images that represent the remaining portion of the small bowel based on application of a second set of rules different from the first set of rules, at least one rule of the second set of rules based on the one or more classification scores associated with each image,

wherein the second set of rules is more selective than the first set of rules.

2 . The method of claim 1 , wherein the proximal portion of the small bowel includes a duodenum.

3 . The method of claim 1 , wherein the disease characterized by the presence of villous atrophy includes at least one of human immunodeficiency virus, common variable immune deficiency, Chron's disease, or celiac disease.

4 . The method of claim 1 , wherein selecting the plurality of images from the identified segment of images is based on the one or more classification scores.

5 . The method of claim 1 , wherein selecting images spaced across the proximal portion of the small bowel includes selecting images of the proximal portion of the small bowel which are not obscured.

6 . The method of claim 1 , further comprising detecting one or more indicators of a disease associated with the presence of villous atrophy in the consecutive set of images based on at least one of a deep learning classifier or a classical machine learning classifier.

7 . The method of claim 1 , wherein selecting the plurality of images from the identified segment of images includes uniformly sampling the proximal portion of the small bowel based on time or length of the small bowel, where length is based on at least one of a number of images in the consecutive set of images or an estimated advance of a capsule endoscopy device along the small bowel.

8 . The method of claim 1 , wherein selecting the plurality of images from the identified segment of images includes:

dividing the plurality of images from the identified segment of images into a predetermined number of sampling points; and

for each sampling point, selecting one or more images from a predetermined range of images surrounding the sampling point.

9 . The method of claim 1 , wherein the indicator includes at least one of scalloping of the mucosa of the small bowel, mosaic patterning of the mucosa of the small bowel, or atrophy of villi of the small bowel.

10 . A system for detecting indicators of a disease characterized by a presence of villous atrophy in images of a gastrointestinal tract (GIT), the system comprising:

one or more processors; and

at least one memory storing instructions which, when executed by the one or more processors, cause the system at least to:

access a consecutive set of images of at least a portion of the GIT comprising a small bowel comprising a proximal portion and a remaining portion, wherein each image of the consecutive set of images is associated with one or more classification scores and each classification score of the one or more classification scores is indicative of the respective associated image including a respective indicator of a disease characterized by the presence of villous atrophy;

select a subset of images, from the consecutive set of images, based on the one or more classification scores of each image of the consecutive set of images;

identify, from the consecutive set of images, a segment of images which includes all of the images that show the proximal portion of the small bowel;

select a plurality of images, from the identified segment of images, that represent the proximal portion of the small bowel, wherein the selecting the plurality of images from the segment of images that represents the proximal portion of the small bowel includes selecting images spaced across the proximal portion of the small bowel;

select a second plurality of images, from the consecutive set of images, that represent the remaining portion of the small bowel;

display the plurality of images selected from the identified segment of images, the second plurality of images, and the subset of images on a display;

select a first subset of images from the images that show the proximal portion of the small bowel based on application of a first set of rules, at least one rule of the first set of rules based on the one or more classification scores associated with each image; and

selecting a second subset of images from the images that represent the remaining portion of the small bowel based on application of a second set of rules different from the first set of rules, at least one rule of the second set of rules based on the one or more classification scores associated with each image,

wherein the second set of rules is more selective than the first set of rules.

11 . The system of claim 10 , wherein the proximal portion of the small bowel includes a duodenum.

12 . The system of claim 10 , wherein the disease characterized by the presence of villous atrophy includes at least one of human immunodeficiency virus, common variable immune deficiency, Chron's disease, or celiac disease.

13 . The system of claim 10 , wherein selecting the plurality of images from the identified segment of images is based on the one or more classification scores.

14 . The system of claim 10 , wherein selecting images spaced across the proximal portion of the small bowel includes selecting images of the proximal portion of the small bowel which are not obscured.

15 . The system of claim 10 , wherein the instructions, when executed by the one or more processors, further cause the system at least to:

detect one or more indicators of a disease associated with the presence of villous atrophy in the consecutive set of images based on at least one of a deep learning classifier or a classical machine learning classifier.

16 . The system of claim 10 , wherein selecting the plurality of images from the identified segment of images includes uniformly sampling the proximal portion of the small bowel based on time or length of the small bowel, where length is based on at least one of a number of images in the consecutive set of images or an estimated advance of a capsule endoscopy device along the small bowel.

17 . The system of claim 10 , wherein selecting the plurality of images from the identified segment of images includes:

dividing the plurality of images from the identified segment of images into a predetermined number of sampling points; and

for each sampling point, selecting one or more images from a predetermined range of images surrounding the sampling point.

18 . The system of claim 10 , wherein the indicator includes at least one of scalloping of the mucosa of the small bowel, mosaic patterning of the mucosa of the small bowel, or atrophy of villi of the small bowel.