Learning More for Free - A Multi Task Learning Approach for Improved Pathology Classification in Capsule Endoscopy

TL;DR

This paper introduces a multi-task learning approach combining self-supervision and full supervision to improve pathology classification in Wireless Capsule Endoscopy, especially with limited data, inspired by the Human Visual System.

Contribution

It proposes a novel multi-task learning framework that leverages self-supervision signals inspired by human vision to enhance pathology classification in WCE with limited data.

Findings

Improved classification accuracy with limited data

Effective use of self-supervision signals inspired by HVS

Analysis of high-level features for robustness

Abstract

The progress in Computer Aided Diagnosis (CADx) of Wireless Capsule Endoscopy (WCE) is thwarted by the lack of data. The inadequacy in richly representative healthy and abnormal conditions results in isolated analyses of pathologies, that can not handle realistic multi-pathology scenarios. In this work, we explore how to learn more for free, from limited data through solving a WCE multicentric, multi-pathology classification problem. Learning more implies to learning more than full supervision would allow with the same data. This is done by combining self supervision with full supervision, under multi task learning. Additionally, we draw inspiration from the Human Visual System (HVS) in designing self supervision tasks and investigate if seemingly ineffectual signals within the data itself can be exploited to gain performance, if so, which signals would be better than others. Further, we present our analysis of the high level features as a stepping stone towards more robust multi-pathology CADx in WCE.