Few-Shot Anomaly Detection for Polyp Frames from Colonoscopy

TL;DR

This paper introduces a few-shot anomaly detection method for colonoscopy frames that effectively identifies polyps by learning from normal images and a small set of abnormal images, achieving state-of-the-art results.

Contribution

The paper presents a novel few-shot anomaly detection approach that combines mutual information maximization with a score inference network, specifically tailored for polyp detection in colonoscopy videos.

Findings

Achieves state-of-the-art polyp detection accuracy.

Performance stabilizes with around 40 abnormal training images.

Effective in scenarios with limited abnormal samples.

Abstract

Anomaly detection methods generally target the learning of a normal image distribution (i.e., inliers showing healthy cases) and during testing, samples relatively far from the learned distribution are classified as anomalies (i.e., outliers showing disease cases). These approaches tend to be sensitive to outliers that lie relatively close to inliers (e.g., a colonoscopy image with a small polyp). In this paper, we address the inappropriate sensitivity to outliers by also learning from inliers. We propose a new few-shot anomaly detection method based on an encoder trained to maximise the mutual information between feature embeddings and normal images, followed by a few-shot score inference network, trained with a large set of inliers and a substantially smaller set of outliers. We evaluate our proposed method on the clinical problem of detecting frames containing polyps from colonoscopy video sequences, where the training set has 13350 normal images (i.e., without polyps) and less than 100 abnormal images (i.e., with polyps). The results of our proposed model on this data set reveal a state-of-the-art detection result, while the performance based on different number of anomaly samples is relatively stable after approximately 40 abnormal training images.