Task-Agnostic Continual Learning for Chest Radiograph Classification

2026-02-17

AI Summary

This work addresses the challenge of continual learning for chest X-ray classification in clinical deployment, where models must adapt to new tasks without access to historical data and without prior knowledge of task identities, while preserving validated diagnostic performance. To this end, the authors propose CARL-XRay, a method that freezes the backbone network and incrementally introduces lightweight task-specific adapters and classification heads. By integrating feature replay with a compact prototype-based latent task selector, the approach enables continual learning without storing original images. CARL-XRay achieves, for the first time, effective continual learning under unknown task identities in chest X-ray classification, attaining an inference-time AUROC of 0.75 and a task routing accuracy of 75.0%—surpassing the joint training baseline (62.5%)—while substantially reducing the number of trainable parameters, thus balancing diagnostic efficacy with parameter efficiency.

Technology Category

• Machine Learning: Life-Long and Continual Learning
• Computer Vision: Learning & Optimization for CV
• Search and Optimization: Learning to Search

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Clinical deployment of chest radiograph classifiers requires models that can be updated as new datasets become available without retraining on previously ob- served data or degrading validated performance. We study, for the first time, a task-incremental continual learning setting for chest radiograph classification, in which heterogeneous chest X-ray datasets arrive sequentially and task identifiers are unavailable at inference. We propose a continual adapter-based routing learning strategy for Chest X-rays (CARL-XRay) that maintains a fixed high-capacity backbone and incrementally allocates lightweight task-specific adapters and classifier heads. A latent task selector operates on task-adapted features and leverages both current and historical context preserved through compact prototypes and feature-level experience replay. This design supports stable task identification and adaptation across sequential updates while avoiding raw-image storage. Experiments on large-scale public chest radiograph datasets demonstrate robust performance retention and reliable task-aware inference under continual dataset ingestion. CARL-XRay outperforms joint training under task-unknown deployment, achieving higher routing accuracy (75.0\% vs.\ 62.5\%), while maintaining competitive diagnostic performance with AUROC of 0.74 in the oracle setting with ground-truth task identity and 0.75 under task-unknown inference, using significantly fewer trainable parameters. Finally, the proposed framework provides a practical alternative to joint training and repeated full retraining in continual clinical deployment.

Problem

• continual learning
• chest radiograph classification
• task-agnostic
• clinical deployment
• task-incremental

Innovation

• continual learning
• adapter-based routing
• task-agnostic inference
• feature-level replay
• chest radiograph classification