Feature learning based on visual similarity triplets in medical image analysis: A case study of emphysema in chest CT scans

TL;DR

This paper introduces a novel CNN-based method that learns disease-relevant features from visual similarity triplets in chest CT scans, enabling better representation of emphysema without requiring detailed annotations.

Contribution

It is the first to use similarity triplets for feature learning in medical images, reducing annotation effort and variability while capturing disease-relevant features.

Findings

CNNs learned meaningful emphysema features from similarity triplets

The method performed well on 973 chest CT images

It enables embedding of unseen test images based on learned features

Abstract

Supervised feature learning using convolutional neural networks (CNNs) can provide concise and disease relevant representations of medical images. However, training CNNs requires annotated image data. Annotating medical images can be a time-consuming task and even expert annotations are subject to substantial inter- and intra-rater variability. Assessing visual similarity of images instead of indicating specific pathologies or estimating disease severity could allow non-experts to participate, help uncover new patterns, and possibly reduce rater variability. We consider the task of assessing emphysema extent in chest CT scans. We derive visual similarity triplets from visually assessed emphysema extent and learn a low dimensional embedding using CNNs. We evaluate the networks on 973 images, and show that the CNNs can learn disease relevant feature representations from derived similarity triplets. To our knowledge this is the first medical image application where similarity triplets has been used to learn a feature representation that can be used for embedding unseen test images