Reliable Mislabel Detection for Video Capsule Endoscopy Data

TL;DR

This paper presents a framework for detecting mislabels in medical video datasets, specifically for Video Capsule Endoscopy, improving data quality and classification accuracy by identifying and correcting annotation errors.

Contribution

The authors introduce a novel mislabel detection framework tailored for medical imaging datasets, validated on large Video Capsule Endoscopy datasets with expert re-annotation.

Findings

Successfully detects mislabeled samples in medical video data

Improves anomaly detection performance after dataset cleaning

Validated on large, publicly available datasets with expert review

Abstract

The classification performance of deep neural networks relies strongly on access to large, accurately annotated datasets. In medical imaging, however, obtaining such datasets is particularly challenging since annotations must be provided by specialized physicians, which severely limits the pool of annotators. Furthermore, class boundaries can often be ambiguous or difficult to define which further complicates machine learning-based classification. In this paper, we want to address this problem and introduce a framework for mislabel detection in medical datasets. This is validated on the two largest, publicly available datasets for Video Capsule Endoscopy, an important imaging procedure for examining the gastrointestinal tract based on a video stream of lowresolution images. In addition, potentially mislabeled samples identified by our pipeline were reviewed and re-annotated by three experienced gastroenterologists. Our results show that the proposed framework successfully detects incorrectly labeled data and results in an improved anomaly detection performance after cleaning the datasets compared to current baselines.