What Makes Transfer Learning Work for Medical Images: Feature Reuse & Other Factors
Open accessThe DOI matches its Crossref record; 22 of the 22 references it cites were found in a registry or an index.
Abstract
Transfer learning is a standard technique to transfer knowledge from one domain to another. For applications in medical imaging, transfer from ImageNet has become the de-facto approach, despite differences in the tasks and im-age characteristics between the domains. However, it is un-clear what factors determine whether - and to what extent- transfer learning to the medical domain is useful. The long- standing assumption that features from the source domain get reused has recently been called into question. Through a series of experiments on several medical image bench-mark datasets, we explore the relationship between transfer learning, data size, the capacity and inductive bias of the model, as well as the distance between the source and tar-get domain. Our findings suggest that transfer learning is beneficial in most cases, and we characterize the important role feature reuse plays in its success.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Deep Residual Learning for Image Recognition
He, K., Zhang, X., Ren, S., & Sun, J. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.90 · Cited by 235,474
Adam: A Method for Stochastic Optimization
Kingma, D. P., & Ba, J. (2014) · arXiv · doi:10.48550/arxiv.1412.6980 · Cited by 98,946
ImageNet: A large-scale hierarchical image database
Deng, J., Dong, W., Socher, R., Li, L.-J., Li, K., & Fei-Fei, L. (2009) · 2009 IEEE Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2009.5206848 · Cited by 65,333
An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale
Dosovitskiy, A., Beyer, L., Kolesnikov, A., Weissenborn, D., Zhai, X., Unterthiner, T., Dehghani, M., Minderer, M., Heigold, G., Gelly, S., et al. (2020) · arXiv · doi:10.48550/arxiv.2010.11929 · Cited by 47,472
Swin Transformer: Hierarchical Vision Transformer using Shifted Windows
Liu, Z., Lin, Y., Cao, Y., Hu, H., Wei, Y., Zhang, Z., Lin, S., & Guo, B. (2021) · 2021 IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv48922.2021.00986 · Cited by 35,534
Representation Learning: A Review and New Perspectives
Bengio, Y., Courville, A., & Vincent, P. (2013) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2013.50 · Cited by 13,001
Decoupled Weight Decay Regularization
Loshchilov, I., & Hutter, F. (2017) · arXiv · doi:10.48550/arxiv.1711.05101 · Cited by 12,522
GANs Trained by a Two Time-Scale Update Rule Converge to a Local Nash Equilibrium
Heusel, M., Ramsauer, H., Unterthiner, T., Nessler, B., & Hochreiter, S. (2017) · arXiv · doi:10.48550/arxiv.1706.08500 · Cited by 8,365
How transferable are features in deep neural networks?
Yosinski, J., Clune, J., Bengio, Y., & Lipson, H. (2014) · arXiv · doi:10.48550/arxiv.1411.1792 · Cited by 6,061 · cited 2×
“[43] found early features tend to be more general, and as we move deeper into the network the features become more task-specific.”
BackgroundEmerging Properties in Self-Supervised Vision Transformers
Caron, M., Touvron, H., Misra, I., Jegou, H., Mairal, J., Bojanowski, P., & Joulin, A. (2021) · 2021 IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv48922.2021.00951 · Cited by 5,638 · cited 2×
“This fits with previous works that have shown that local attention, which is crucial for good performance, is learned in early layers [7, 11].”
MethodResultEhteshami Bejnordi, B., Veta, M., Johannes van Diest, P., van Ginneken, B., Karssemeijer, N., Litjens, G., van der Laak, J. A. W. M., and the CAMELYON16 Consortium, Hermsen, M., Manson, Q. F., et al. (2017) · JAMA · doi:10.1001/jama.2017.14585 · Cited by 3,293
Convolutional Neural Networks for Medical Image Analysis: Full Training or Fine Tuning?
Tajbakhsh, N., Shin, J. Y., Gurudu, S. R., Hurst, R. T., Kendall, C. B., Gotway, M. B., & Liang, J. (2016) · IEEE Transactions on Medical Imaging · doi:10.1109/tmi.2016.2535302 · Cited by 3,277 · cited 1×
“The lack of large public datasets has led to the widespread adoption of transfer learning from IMAGENET [10] to improve performance on medical tasks [27, 28, 39].”
BackgroundCodella, N. C. F., Gutman, D., Celebi, M. E., Helba, B., Marchetti, M. A., Dusza, S. W., Kalloo, A., Liopyris, K., Mishra, N., Kittler, H., et al. (2018) · 2018 IEEE 15th International Symposium on Biomedical Imaging (ISBI 2018) · doi:10.1109/isbi.2018.8363547 · Cited by 2,620
Geirhos, R., Rubisch, P., Michaelis, C., Bethge, M., Wichmann, F. A., & Brendel, W. (2018) · arXiv · doi:10.48550/arxiv.1811.12231 · Cited by 1,490 · cited 1×
“Recent investigations have looked into which situations transfer learning works and when it fails [12,14,22,27,31].”
BackgroundDo Better ImageNet Models Transfer Better?
Kornblith, S., Shlens, J., & Le, Q. V. (2019) · 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2019.00277 · Cited by 1,325
Pros and cons of GAN evaluation measures
Borji, A. (2019) · Computer Vision and Image Understanding · doi:10.1016/j.cviu.2018.10.009 · Cited by 963
A curated mammography data set for use in computer-aided detection and diagnosis research
Lee, R. S., Gimenez, F., Hoogi, A., Miyake, K. K., Gorovoy, M., & Rubin, D. L. (2017) · Scientific Data · doi:10.1038/sdata.2017.177 · Cited by 859
Similarity of Neural Network Representations Revisited
Kornblith, S., Norouzi, M., Lee, H., & Hinton, G. (2019) · arXiv · doi:10.48550/arxiv.1905.00414 · Cited by 724 · cited 1×
“Looking more closely at feature reuse within the network, we ask the questions: how are features organized before and after fine-tuning – are they similar? Can feature similarity reveal feature reuse, or lack thereof? To answer these questions, we use Centered Kernel Alignment (CKA) to compute similarity between featu…”
MethodA scoping review of transfer learning research on medical image analysis using ImageNet
Morid, M. A., Borjali, A., & Del Fiol, G. (2021) · Computers in Biology and Medicine · doi:10.1016/j.compbiomed.2020.104115 · Cited by 508
Rotation Equivariant CNNs for Digital Pathology
Veeling, B. S., Linmans, J., Winkens, J., Cohen, T., & Welling, M. (2018) · Lecture Notes in Computer Science · doi:10.1007/978-3-030-00934-2_24 · Cited by 476
Factors of Transferability for a Generic ConvNet Representation
Azizpour, H., Razavian, A. S., Sullivan, J., Maki, A., & Carlsson, S. (2016) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2015.2500224 · Cited by 301 · cited 1×
“They also showed, along with Azizpour et al ., that bene-fits from transfer learning diminish as the distance between the source and target domain increases [1, 43].”
BackgroundNguyen, T., Raghu, M., & Kornblith, S. (2020) · arXiv · doi:10.48550/arxiv.2010.15327 · Cited by 112
Cited by
Benchmarking Self-Supervised Learning on Diverse Pathology Datasets
Kang, M., Song, H., Park, S., Yoo, D., & Pereira, S. (2023) · 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr52729.2023.00326 · Cited by 187
A systematic review of intermediate fusion in multimodal deep learning for biomedical applications
Guarrasi, V., Aksu, F., Caruso, C. M., Di Feola, F., Rofena, A., Ruffini, F., & Soda, P. (2025) · Image and Vision Computing · doi:10.1016/j.imavis.2025.105509 · Cited by 95
Assessing the limits of zero-shot foundation models in single-cell biology
Kedzierska, K. Z., Crawford, L., Amini, A. P., & Lu, A. X. (2023) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2023.10.16.561085 · Cited by 61
Feature Reuse and Scaling: Understanding Transfer Learning with Protein Language Models
Li, F.-Z., Amini, A. P., Yue, Y., Yang, K. K., & Lu, A. X. (2024) · doi:10.1101/2024.02.05.578959 · Cited by 51
Pre-training in Medical Data: A Survey
Qiu, Y., Lin, F., Chen, W., & Xu, M. (2023) · Machine Intelligence Research · doi:10.1007/s11633-022-1382-8 · Cited by 30
Metadata
- Authors
- Matsoukas, C., Haslum, J. F., Sorkhei, M., Soderberg, M., & Smith, K.
- Venue
- 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
- Publisher
- IEEE
- Published
- 2022-06
- Pages
- 9215-9224
- ISSN
- 1063-6919 (ISSN-L)
- OpenAlex
- W4312713789
- Cited by
- 113
- References
- 22
- FWCI
- 14.38
- Type
- conference paper
- Language
- en
- Open access
- green
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.