Unsupervised Representation Learning Meets Pseudo-Label Supervised Self-Distillation: A New Approach to Rare Disease Classification

TL;DR

This paper introduces a hybrid approach combining unsupervised representation learning with pseudo-label supervised self-distillation to improve rare disease classification, especially when labeled data is scarce.

Contribution

The paper proposes a novel hybrid method that leverages unsupervised URL and pseudo-label self-distillation, reducing labeling needs and enhancing rare disease classification performance.

Findings

Outperforms existing few-shot learning methods on rare skin lesion classification

Effectively integrates unsupervised and pseudo-supervised learning for better knowledge transfer

Establishes new state-of-the-art results in rare disease classification

Abstract

Rare diseases are characterized by low prevalence and are often chronically debilitating or life-threatening. Imaging-based classification of rare diseases is challenging due to the severe shortage in training examples. Few-shot learning (FSL) methods tackle this challenge by extracting generalizable prior knowledge from a large base dataset of common diseases and normal controls, and transferring the knowledge to rare diseases. Yet, most existing methods require the base dataset to be labeled and do not make full use of the precious examples of the rare diseases. To this end, we propose in this work a novel hybrid approach to rare disease classification, featuring two key novelties targeted at the above drawbacks. First, we adopt the unsupervised representation learning (URL) based on self-supervising contrastive loss, whereby to eliminate the overhead in labeling the base dataset. Second, we integrate the URL with pseudo-label supervised classification for effective self-distillation of the knowledge about the rare diseases, composing a hybrid approach taking advantages of both unsupervised and (pseudo-) supervised learning on the base dataset. Experimental results on classification of rare skin lesions show that our hybrid approach substantially outperforms existing FSL methods (including those using fully supervised base dataset) for rare disease classification via effective integration of the URL and pseudo-label driven self-distillation, thus establishing a new state of the art.