Resolution scaling governs DINOv3 transfer performance in chest radiograph classification

Abstract

Self-supervised learning (SSL) has improved visual representation learning, but its value in chest radiography remains uncertain. DINOv3 extends earlier SSL models through Gram-anchored self-distillation and explicit high-resolution adaptation. Whether these changes improve transfer learning for chest radiograph classification has not been established. We benchmarked DINOv3 against DINOv2 and supervised ImageNet initialization across seven chest radiograph datasets comprising 816,183 radiographs from pediatric and adult cohorts. ViT-B/16 and ConvNeXt-B were evaluated under full fine-tuning at 224 and 512 pixels, with targeted 1024 experiments on three cohorts. Additional analyses examined parameter-efficient adaptation, synthetic label corruption, external validation, frozen 7B features, and computational efficiency. The primary outcome was mean AUROC across labels. In adult cohorts, DINOv3 did not consistently outperform DINOv2 at 224 x 224 pixels, but became the strongest initialization at 512 x 512, especially with ConvNeXt-B. Gains were greatest for small focal and boundary-dependent abnormalities, whereas large-structure findings changed little. The pediatric cohort showed no significant benefit from DINOv3, higher resolution, or backbone choice. Scaling to 1024 x 1024 rarely improved performance and markedly increased computational cost. ConvNeXt-B remained superior to ViT-B/16 under both full and parameter-efficient adaptation. External validation preserved the 512 x 512 DINOv3 advantage, whereas synthetic label corruption showed that this benefit should not be interpreted simply as superior noise robustness. For adult chest radiograph classification, DINOv3 provides its most reliable benefit at 512 x 512 pixels, particularly with ConvNeXt-B. Fully adapted mid-sized models at 512 x 512 pixels provided the best performance-cost trade-off in our benchmark.

References in corpus (13)

• ChestX-ray8: Hospital-scale Chest X-ray Database and Benchmarks on Weakly-Supervised Classification and Localization of Common Thorax Diseases
• Gaussian Error Linear Units (GELUs)
• PadChest: A large chest x-ray image dataset with multi-label annotated reports
• Large Language Models Streamline Automated Machine Learning for Clinical Studies
• CheXtransfer: Performance and Parameter Efficiency of ImageNet Models for Chest X-Ray Interpretation
• Private, fair and accurate: Training large-scale, privacy-preserving AI models in medical imaging
• Enhancing Network Initialization for Medical AI Models Using Large-Scale, Unlabeled Natural Images
• DINOv3
• Collaborative Training of Medical Artificial Intelligence Models with non-uniform Labels
• RadioRAG: Online Retrieval-augmented Generation for Radiology Question Answering
• Mind the Gap: Federated Learning Broadens Domain Generalization in Diagnostic AI Models
• Preserving privacy in domain transfer of medical AI models comes at no performance costs: The integral role of differential privacy
• Boosting multi-demographic federated learning for chest radiograph analysis using general-purpose self-supervised representations