Back to results
US 2020/0265275 A1Published application
Classification of polyps using learned image analysis
AssigneeBaylor Research Institute
InventorsThemistocles DASSOPOULOS, Nikolaos BOURBAKIS
Life of this patent
Estimating the term…
Abstract
Computational techniques are applied to video images of polyps to extract features and patterns from different perspectives of a polyp. The extracted features and patterns are synthesized using registration techniques to remove artifacts and noise, thereby generating improved images for the polyp. The generated images of each polyp can be used for training and testing purposes, where a machine learning system separates two types of polyps.
Claims14
A system for classifying polyps, the system comprising:
- an polyp image database comprising, for a plurality of polyps, images of the plurality of polyps taken from different perspectives;
- a polyp imaging engine, the polyp imaging engine configured to:
- compile, for at least one of the plurality of polyps, images of at least one of the plurality of polyps taken from different perspectives,
- generate, from the compiled images of the at least one polyp, a new polyp image, the new polyp image having fewer reflection artifacts and occlusions than the compiled images, and
- compute, based on the generated new polyp image, a polyp surface model;
- and
- a learning engine, the learning engine configured to:
- apply linear subspace learning techniques and nonlinear subspace learning techniques to identify discriminate features in the polyp surface model, and
- based on the identified discriminate features, classify the polyp as adenomatous or hyperplastic.
6 more elements
A method for classifying polyps, the method comprising:
- storing, in an image database, for a plurality of polyps, images of the plurality of polyps taken from different perspectives;
- compiling, for at least one of the plurality of polyps, images of at least one of the plurality of polyps taken from different perspectives,
- generating, from the compiled images of the at least one polyp, a new polyp image, the new polyp image having fewer reflection artifacts and occlusions than the compiled images, and
- computing, based on the generated new polyp image, a polyp surface model;
- and
- applying, by a learning engine, a linear subspace learning techniques and nonlinear subspace learning techniques to identify discriminate features in the polyp surface model, and
- based on the identified discriminate features, classifying the polyp as adenomatous or hyperplastic.
4 more elements
Dependencies are inferred from the claim text.