A Method Of Classifying Abnormalities In Endoscopic Data, And A System Thereof

Abstract: ABSTRACT “A METHOD OF CLASSIFYING ABNORMALITIES IN ENDOSCOPIC DATA, AND A SYSTEM THEREOF” Present disclosure provides system and method for classifying abnormalities in endoscopic data. The method comprises receiving endoscopic data of subject. The endoscopic data comprises a plurality of frames. Thereafter, the method comprises identifying one or more abnormal frames, corresponding to an abnormality using an AI model. Thereafter, the method comprises determining a degree of similarity of features between each of the one or more abnormal frames and a plurality of pre-defined abnormality groups, at each of a plurality of hierarchical levels. Subsequently, the method comprises allocating the one or more abnormality types associated with one of the plurality of pre-defined abnormality groups, for each of the one or more abnormal frames, at each of the plurality of hierarchical levels, using hierarchical tree classifier, based on degree of similarity and identifying an abnormality type from the one or more abnormality types, associated with each of the one or more abnormal frames. Figure 3A

Patent Information

Applicants

Inventors

Specification

Documents

Application Documents

Search Strategy