US 12,254,673 B1
Systems and methods for classifying or selecting images based on image segmentation
Yossi Tsadok, Pardes Hanna-Karkur (IL); and Dori Peleg, Kiryat Bialik (IL)
Assigned to GIVEN IMAGING LTD, Yoqneam (IL)
Filed by Given Imaging LTD, Yoqneam (IL)
Filed on Dec. 7, 2021, as Appl. No. 17/543,985.
Claims priority of provisional application 63/138,905, filed on Jan. 19, 2021.
Int. Cl. G06V 10/764 (2022.01); A61B 1/04 (2006.01); G06T 7/00 (2017.01); G06V 10/26 (2022.01); G06V 10/46 (2022.01); G06V 10/762 (2022.01); G06V 10/82 (2022.01)
CPC G06V 10/765 (2022.01) [A61B 1/041 (2013.01); G06T 7/0012 (2013.01); G06V 10/267 (2022.01); G06V 10/46 (2022.01); G06V 10/762 (2022.01); G06V 10/82 (2022.01); G06T 2207/10068 (2013.01); G06T 2207/20084 (2013.01); G06T 2207/30028 (2013.01); G06T 2207/30092 (2013.01); G06V 2201/03 (2022.01)] 18 Claims
OG exemplary drawing
 
1. A classification system for classifying images, comprising:
one or more processors; and
at least one memory storing machine executable instructions which, when executed by the one or more processors, cause the classification system to:
input an image captured by a capsule endoscopy device during a CE procedure to a deep learning neural network to generate image segmentation scores for pixels of the image, each of the image segmentation scores corresponding to at least one pixel of the pixels of the image, and
classify the entire image based on the image segmentation scores for the pixels of the image, wherein each of the image segmentation scores indicates a probability of the corresponding at least one pixel being a pixel of interest or not being a pixel of interest.