Data to 7 Oct 2026

Transfusion: Understanding Transfer Learning for Medical Imaging

arXiv 1902.07208 ↗First posted 14 Feb 2019cs.CV cs.LG stat.MLcited by 366

Not on the map. Why?

Abstract (arXiv)

Transfer learning from natural image datasets, particularly ImageNet, using standard large models and corresponding pretrained weights has become a de-facto method for deep learning applications to medical imaging. However, there are fundamental differences in data sizes, features and task specifications between natural image classification and the target medical tasks, and there is little understanding of the effects of transfer. In this paper, we explore properties of transfer learning for medical imaging. A performance evaluation on two large scale medical imaging tasks shows that surprisingly, transfer offers little benefit to performance, and simple, lightweight models can perform comparably to ImageNet architectures. Investigating the learned representations and features, we find that some of the differences from transfer learning are due to the over-parametrization of standard models rather than sophisticated feature reuse. We isolate where useful feature reuse occurs, and outline the implications for more efficient model exploration. We also explore feature independent benefits of transfer arising from weight scalings.

Key facts. Cited by 366 papers in the map (data through 7 Oct 2026). First posted on arXiv 14 Feb 2019.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 84; 2022: 77; 2023: 75; 2024: 36; 2025: 49; 2026: 45.

366 papers.

Sort:

By area:

2021Do Vision Transformers See Like Convolutional Neural Networks?cited by 671
2023Medical SAM Adapter: Adapting Segment Anything Model for Medical Image Segmentationcited by 485
2021Self-Supervised Pre-Training of Swin Transformers for 3D Medical Image Analysiscited by 418
2021Domain Adaptation for Medical Image Analysis: A Surveycited by 351
2021Big Self-Supervised Models Advance Medical Image Classificationcited by 320
2023BiomedGPT: A Generalist Vision-Language Foundation Model for Diverse Biomedical Taskscited by 201
2021BENDR: using transformers and a contrastive self-supervised learning task to learn from massive amounts of EEG datacited by 194
2022Benchmarking Self-Supervised Learning on Diverse Pathology Datasetscited by 148
2021Natural Synthetic Anomalies for Self-Supervised Anomaly Detection and Localizationcited by 116
2021Recent advances and clinical applications of deep learning in medical image analysiscited by 104
2021Exploring the Limits of Large Scale Pre-trainingcited by 104
2021Multi-modal Understanding and Generation for Medical Images and Text via Vision-Language Pre-Trainingcited by 96

38 more

2022Neural Collapse: A Review on Modelling Principles and Generalizationcited by 95
2023UniverSeg: Universal Medical Image Segmentationcited by 90
2021Parameter Prediction for Unseen Deep Architecturescited by 89
2022Position: Considerations for Differentially Private Learning with Large-Scale Public Pretrainingcited by 86
2023Differentially Private Diffusion Models Generate Useful Synthetic Imagescited by 85
2022Federated Learning Enables Big Data for Rare Cancer Boundary Detectioncited by 83
2021Self-Supervised Pretraining Improves Self-Supervised Pretrainingcited by 81
2023A Foundation Language-Image Model of the Retina (FLAIR): Encoding Expert Knowledge in Text Supervisioncited by 78
2021Self-supervised learning methods and applications in medical imaging analysis: A surveycited by 76
2022Masked Image Modeling Advances 3D Medical Image Analysiscited by 71
2021FREE: Feature Refinement for Generalized Zero-Shot Learningcited by 67
2022Medical Image Understanding with Pretrained Vision Language Models: A Comprehensive Studycited by 66
2023Polyp-SAM: Transfer SAM for Polyp Segmentationcited by 64
2022What Makes Transfer Learning Work For Medical Images: Feature Reuse & Other Factorscited by 61
2022Expanding Small-Scale Datasets with Guided Imaginationcited by 61
2021Is it Time to Replace CNNs with Transformers for Medical Images?cited by 60
2022Post hoc Explanations may be Ineffective for Detecting Unknown Spurious Correlationcited by 59
2021Multiple Instance Captioning: Learning Representations from Histopathology Textbooks and Articlescited by 58
2022Transferability in Deep Learning: A Surveycited by 57
2023PRIOR: Prototype Representation Joint Learning from Medical Images and Reportscited by 55
2023ScribblePrompt: Fast and Flexible Interactive Segmentation for Any Biomedical Imagecited by 47
2023Rethinking Semi-Supervised Medical Image Segmentation: A Variance-Reduction Perspectivecited by 43
2021How Fine-Tuning Allows for Effective Meta-Learningcited by 41
2022Robust and Efficient Medical Imaging with Self-Supervisioncited by 40
2023Overcoming Data Scarcity in Biomedical Imaging with a Foundational Multi-Task Modelcited by 40
2021Self-supervised driven consistency training for annotation efficient histopathology image analysiscited by 39
2021Self-Tuning for Data-Efficient Deep Learningcited by 39
2024LiNeS: Post-training Layer Scaling Prevents Forgetting and Enhances Model Mergingcited by 39
2021Supervised Transfer Learning at Scale for Medical Imagingcited by 38
2021Correlated Input-Dependent Label Noise in Large-Scale Image Classificationcited by 37
2023Trainable Projected Gradient Method for Robust Fine-tuningcited by 37
2021Adversarial Training Helps Transfer Learning via Better Representationscited by 36
2021Revisiting the Transferability of Supervised Pretraining: an MLP Perspectivecited by 36
2021Zoo-Tuning: Adaptive Transfer from a Zoo of Modelscited by 35
2023Does progress on ImageNet transfer to real-world datasets?cited by 35
2021An Empirical Framework for Domain Generalization in Clinical Settingscited by 32
2021How Transferable Are Self-supervised Features in Medical Image Classification Tasks?cited by 32
2023Training Like a Medical Resident: Context-Prior Learning Toward Universal Medical Image Segmentationcited by 32

50 most cited of 366 All 366 (JSON)