Meta-learning reduces the amount of data needed to build AI models in oncology
Open accessAbstract
Meta-learning is showing promise in recent genomic studies in oncology. Meta-learning can facilitate transfer learning and reduce the amount of data that is needed in a target domain by transferring knowledge from abundant genomic data in different source domains enabling the use of AI in data scarce scenarios.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
Few- or zero-shot learning approaches, such as meta-learning and transfer learning, have been proposed to enable adaption in small-size datasets.
“To overcome this problem, few- or zero-shot learning approaches, such as meta-learning and transfer learning, have been proposed to enable adaption in small-size datasets [178].”
From Challenges in AI-driven Biomedical Multimodal Data Fusion and Analysis (2025) · cites this paper as [178]
Cites
Ma, J., Fong, S. H., Luo, Y., Bakkenist, C. J., Shen, J. P., Mourragui, S., Wessels, L. F. A., Hafner, M., Sharan, R., Peng, J., et al. (2021) · Nature Cancer · doi:10.1038/s43018-020-00169-2 · Cited by 180 · see citing works
Flagged first, then by citations
Cited by
Sorted by citations
Challenges in AI-driven Biomedical Multimodal Data Fusion and Analysis
Liu, J., Cen, X., Yi, C., Wang, F.-A., Ding, J., Cheng, J., Wu, Q., Gai, B., Zhou, Y., He, R., et al. (2025) · Genomics, Proteomics & Bioinformatics · doi:10.1093/gpbjnl/qzaf011 · Cited by 75 · see citing works
“Techniques such as meta-learning and transfer learning can be employed to learn general data representations from other large datasets [178,181].”
Interpretable meta-learning of multi-omics data for survival analysis and pathway enrichment
Cho, H. J., Shu, M., Bekiranov, S., Zang, C., & Zhang, A. (2023) · Bioinformatics · doi:10.1093/bioinformatics/btad113 · Cited by 42 · see citing works
“It has been shown that meta-learning methods often require fewer samples than more traditional deep learning models (Triantafillou et al. 2020; Gevaert 2021).”
MethodAI-driven nanomedicine for cancer theranostics
Tiwari, A., Widodo, Krisnawati, D. I., Chen, C.-Y., & Kuo, T. (2026) · Molecular Cancer · doi:10.1186/s12943-025-02563-9 · Cited by 41 · see citing works
Generalizing Few-Shot Classification of Whole-Genome Doubling Across Cancer Types
Chao, S., & Belanger, D. (2021) · 2021 IEEE/CVF International Conference on Computer Vision Workshops (ICCVW) · doi:10.1109/iccvw54120.2021.00377 · Cited by 15 · see citing works
One-shot skill assessment in high-stakes domains with limited data via meta learning
Yanik, E., Schwaitzberg, S., Yang, G., Intes, X., Norfleet, J., Hackett, M., & De, S. (2022) · Comput. Biol. Medicine · doi:10.1016/j.compbiomed.2024.108470 · Cited by 12 · see citing works
“This generated 38, 20, and 20 samples for Novice, Intermediate, and Expert classes in robotic suturing, 22, 16, and 18 for needle passing and 32, 20, 20 for knot tying.”
BackgroundInfluentialTefera, M. A., Dehnaw, A., Manie, Y., Yao, C.-K., Bogale, S. D., & Peng, P. (2024) · Future Internet · doi:10.3390/fi16080280 · Cited by 11 · see citing works
“Moreover, real-time data collection for chronic diseases such as diabetes, cancer, cardiovascular diseases, and others is not only time-consuming and tedious but also resource-intensive due to the constant need for continuous or frequent monitoring across various sensors and devices [12–17].”
BackgroundSuccessive model-agnostic meta-learning for few-shot fault time series prognosis
Su, H., Hu, J., Yu, S.-S., Liu, J.-H., & Qin, X. (2024) · Neurocomputing · doi:10.1016/j.neucom.2024.127879 · Cited by 11 · see citing works
Park, Y.-H., Hauschild, A., & Heider, D. (2021) · bioRxiv · doi:10.1093/nargab/lqab104 · Cited by 11 · see citing works
“This meta-learning approach has already gained attention in cancer genomics, where many large-scale datasets have already been generated compared to other fields (12).”
Zero-shot Meta-learning for Tabular Prediction Tasks with Adversarially Pre-trained Transformer
Wu, Y.-L., & Bergman, D. (2025) · International Conference on Machine Learning · doi:10.48550/arxiv.2502.04573 · Cited by 10
Federated Learning in Glaucoma
Hallaj, S., Chuter, B., Lieu, A. C., Singh, P., Kalpathy–Cramer, J., Xu, B. Y., Christopher, M. A., Zangwill, L. M., Weinreb, R. N., & Baxter, S. L. (2024) · Ophthalmology Glaucoma · doi:10.1016/j.ogla.2024.08.004 · Cited by 10 · see citing works
A meta-learning approach to improving radiation response prediction in cancers
Zhang, Y.-N., Qiu, L., Ren, Y.-Y., Cheng, Z., Li, L., Yao, S.-Q., Zhang, C.-D., Luo, Z., & Lu, H. (2022) · Comput. Biol. Medicine · doi:10.1016/j.compbiomed.2022.106163 · Cited by 8 · see citing works
Successive Model-Agnostic Meta-Learning for Few-Shot Fault Time Series Prognosis
Su, H., Hu, J., & Yu, S.-S. (2023) · arXiv.org · doi:10.48550/arxiv.2311.02300 · Cited by 7
“Previous studies have shown that continuous time series data contain more temporal features and heritage, which can improve the performance of neural networks in the prediction process [19].”
BackgroundJiachu, D., Luo, L., Xie, M., Xie, X., Guo, J., Ye, H., Cai, K., Zhou, L., Song, G., Jiang, F., et al. (2024) · Translational Vision Science & Technology · doi:10.1167/tvst.13.9.22 · Cited by 6
“12 Meta-learning enables learning model weights by leveraging prior knowledge from various tasks and can be implemented in different task objectives such as few-shot learning 13 or multi-task learning.”
MethodFang, Y.-Y., Qian, J.-H., Wang, X.-K., Cooper, L. A., & Zhou, B. (2025) · IEEE journal of biomedical and health informatics · doi:10.48550/arxiv.2507.16886 · Cited by 2
“Future improvements could explore (1) distilled network architectures maintaining accuracy while reducing parameters(Moslemi et al., 2024), or (2) meta-learning approaches enabling knowledge transfer across similar tissue specimens(Gevaert, 2021).”
Mahmood, R. A., Mahin, M., Alam, R., Jeong, S., Kang, H., Cho, D., & Kim, S. (2026) · Journal of Analytical Science and Technology · doi:10.1186/s40543-026-00564-x
“Advanced algorithm such as stacked-ensemble learning can be alternative method for the accurate predictions directly from the small data sample size (Gevaert 2021).”
- Authors
- Gevaert, O.
- Affiliations
- Stanford Medicine (US)
- Corresponding author
- Gevaert, Olivier
- ORCID
- Gevaert, Olivier
- Venue
- British Journal of Cancer
- Published
- 2021-03-29
- Volume
- 125
- Issue
- 3
- Pages
- 309-310
- ISSN
- 0007-0920 (ISSN-L)
- 1532-1827
- PMC
- PMC8329303
- OpenAlex
- W3145099584
- Cited by
- 13
- References
- 0
- Citations by year
- 2026: 2 · 2025: 1 · 2024: 6 · 2023: 1 · 2022: 2 · 2021: 1
- FWCI
- 0.88
- Type
- journal article
- Language
- en
- Open access
- green
- www.ncbi.nlm.nih.gov
- Locations
- British Journal of Cancer (publishedVersion)
- PubMed Central (publishedVersion)
- PubMed (publishedVersion)
- Funders
- U.S. Department of Health & Human Services | NIH | National Institute of Biomedical Imaging and Bioengineering (R56 EB020527)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Domain Adaptation and Few-Shot Learning
- Health Sciences › Medicine › Health Informatics › Artificial Intelligence in Healthcare and Education
- Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Gene expression and cancer classification
- Keywords
- meta-learning, transfer learning, data scarcity, oncology, genomic data, artificial intelligence in oncology, cross-domain knowledge transfer
- MeSH terms
- Metadata, Artificial Intelligence, Humans, Neoplasms, Computational Biology, Genomics
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
