Handling Supervision Scarcity in Chest X-ray Classification: Long-Tailed and Zero-Shot Learning

TL;DR

This paper introduces specialized methods for chest X-ray classification that address challenges of limited supervision, including long-tailed disease distributions and zero-shot recognition of unseen findings, achieving top performance in a dedicated challenge.

Contribution

It presents tailored solutions for long-tailed multi-label classification and zero-shot recognition in chest X-ray analysis, advancing handling of supervision scarcity in medical imaging.

Findings

Achieved top ranking on the CXR-LT challenge leaderboard.

Developed an imbalance-aware multi-label learning strategy.

Proposed a zero-shot prediction approach for unseen disease categories.

Abstract

Chest X-Ray (CXR) classification in clinical practice is often limited by imperfect supervision, arising from (i) extreme long-tailed multi-label disease distributions and (ii) missing annotations for rare or previously unseen findings. The CXR-LT 2026 challenge addresses these issues on a PadChest-based benchmark with a 36-class label space split into 30 in-distribution classes for training and 6 out-of-distribution (OOD) classes for zero-shot evaluation. We present task-specific solutions tailored to the distinct supervision regimes. For Task 1 (long-tailed multi-label classification), we adopt an imbalance-aware multi-label learning strategy to improve recognition of tail classes while maintaining stable performance on frequent findings. For Task 2 (zero-shot OOD recognition), we propose a prediction approach that produces scores for unseen disease categories without using any supervised labels or examples from the OOD classes during training. Evaluated with macro-averaged mean Average Precision (mAP), our method achieves strong performance on both tasks, ranking first on the public leaderboard of the development phase. Code and pre-trained models are available at https://github.com/hieuphamha19/CXR_LT.