How many samples to label for an application given a foundation model? Chest X-ray classification study

2025/10/13 by Nechaev, Nikolay, Przhezdzetskaia, Evgeniia, Gombolevskiy, Viktor +2
#62H30 #62P10 (Secondary) #68T07 (Primary) 68T45 #Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences

paper · doi:10.48550/arxiv.2510.11553

Abstract

Chest X-ray classification is vital yet resource-intensive, typically demanding extensive annotated data for accurate diagnosis. Foundation models mitigate this reliance, but how many labeled samples are required remains unclear. We systematically evaluate the use of power-law fits to predict the training size necessary for specific ROC-AUC thresholds. Testing multiple pathologies and foundation models, we find XrayCLIP and XraySigLIP achieve strong performance with significantly fewer labeled examples than a ResNet-50 baseline. Importantly, learning curve slopes from just 50 labeled cases accurately forecast final performance plateaus. Our results enable practitioners to minimize annotation costs by labeling only the essential samples for targeted performance.

Citations

• DINOv2: Learning Robust Visual Features without Supervision
• Sigmoid Loss for Language Image Pre-Training
• RadGraph: Extracting Clinical Entities and Relations from Radiology Reports
• Learning Transferable Visual Models From Natural Language Supervision
• Learning Curves for Analysis of Deep Networks
• Optimal Regularization Can Mitigate Double Descent
• Deep Learning Scaling is Predictable, Empirically
• Deep Residual Learning for Image Recognition
• A Vision-Language Foundation Model to Enhance Efficiency of Chest X-ray Interpretation

Related

• Foundation X: Integrating Classification, Localization, and Segmentation through Lock-Release Pretraining Strategy for Chest X-ray Analysis
• Feature Quality and Adaptability of Medical Foundation Models: A Comparative Evaluation for Radiographic Classification and Segmentation
• Fixed-Budget Parameter-Efficient Training with Frozen Encoders Improves Multimodal Chest X-Ray Classification
• Benchmarking Chest X-ray Diagnosis Models Across Multinational Datasets
• Did You Get What You Paid For? Rethinking Annotation Cost of Deep Learning Based Computer Aided Detection in Chest Radiographs