FoldIt: Haustral Folds Detection and Segmentation in Colonoscopy Videos

TL;DR

This paper introduces FoldIt, a novel GAN-based method for accurate detection and segmentation of haustral folds in colonoscopy videos, improving lesion localization and virtual colonoscopy registration.

Contribution

The paper presents a new generative adversarial network with a transitive loss for feature-consistent image translation and haustral fold segmentation in colonoscopy videos.

Findings

Effective segmentation of haustral folds in real videos

Improved virtual colonoscopy registration accuracy

Code and models made publicly available

Abstract

Haustral folds are colon wall protrusions implicated for high polyp miss rate during optical colonoscopy procedures. If segmented accurately, haustral folds can allow for better estimation of missed surface and can also serve as valuable landmarks for registering pre-treatment virtual (CT) and optical colonoscopies, to guide navigation towards the anomalies found in pre-treatment scans. We present a novel generative adversarial network, FoldIt, for feature-consistent image translation of optical colonoscopy videos to virtual colonoscopy renderings with haustral fold overlays. A new transitive loss is introduced in order to leverage ground truth information between haustral fold annotations and virtual colonoscopy renderings. We demonstrate the effectiveness of our model on real challenging optical colonoscopy videos as well as on textured virtual colonoscopy videos with clinician-verified haustral fold annotations. All code and scripts to reproduce the experiments of this paper will be made available via our Computational Endoscopy Platform at https://github.com/nadeemlab/CEP.