Optimizing risk-based breast cancer screening policies with reinforcement learning
Open accessPublished version of a preprint
The DOI matches its Crossref record; 53 of the 53 references it cites were found in a registry or an index.
Abstract
Screening programs must balance the benefit of early detection with the cost of overscreening. Here, we introduce a novel reinforcement learning-based framework for personalized screening, Tempo, and demonstrate its efficacy in the context of breast cancer. We trained our risk-based screening policies on a large screening mammography dataset from Massachusetts General Hospital (MGH; USA) and validated this dataset in held-out patients from MGH and external datasets from Emory University (Emory; USA), Karolinska Institute (Karolinska; Sweden) and Chang Gung Memorial Hospital (CGMH; Taiwan). Across all test sets, we find that the Tempo policy combined with an image-based artificial intelligence (AI) risk model is significantly more efficient than current regimens used in clinical practice in terms of simulated early detection per screen frequency. Moreover, we show that the same Tempo policy can be easily adapted to a wide range of possible screening preferences, allowing clinicians to select their desired trade-off between early detection and screening costs without training new policies. Finally, we demonstrate that Tempo policies based on AI-based risk models outperform Tempo policies based on less accurate clinical risk models. Altogether, our results show that pairing AI-based risk models with agile AI-designed screening policies has the potential to improve screening programs by advancing early detection while reducing overscreening.
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.
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?
The source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“As shown in Figure 2A, AI agents possess the capacity to process extensive patient data, including medical histories, diagnostic test results, and imaging, thereby aiding healthcare professionals in making more informed and precise diagnostic decisions.62,63,64,65,66,67,68,69”
From A foundational architecture for AI agents in healthcare (2025) · cites this paper as [63]
Experimental — accuracy not yet measured
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
Human-level control through deep reinforcement learning
Mnih, V., Kavukcuoglu, K., Silver, D., Rusu, A. A., Veness, J., Bellemare, M. G., Graves, A., Riedmiller, M., Fidjeland, A. K., Ostrovski, G., et al. (2015) · Nature · doi:10.1038/nature14236 · Cited by 30,357 · cited 3×
“RL algorithms are commonly used in game playing [17, 18] and robotics [19-23].”
MethodLearning Phrase Representations using RNN Encoder–Decoder for Statistical Machine Translation
Cho, K., van Merrienboer, B., Gulcehre, C., Bahdanau, D., Bougares, F., Schwenk, H., & Bengio, Y. (2014) · Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP) · doi:10.3115/v1/d14-1179 · Cited by 25,545 · cited 1×
“For our RNN model, we utilized a Gated Recurrent Unit (GRU) [47] with an additive hazard layer [12] to estimate future risk.”
MethodReinforcement Learning: An Introduction
Johnson, J. D., Li, J., & Chen, Z. (2000) · Neurocomputing · doi:10.1016/s0925-2312(00)00324-6 · Cited by 8,414
Grandmaster level in StarCraft II using multi-agent reinforcement learning
Vinyals, O., Babuschkin, I., Czarnecki, W. M., Mathieu, M., Dudzik, A., Chung, J., Choi, D. H., Powell, R., Ewalds, T., Georgiev, P., et al. (2019) · Nature · doi:10.1038/s41586-019-1724-z · Cited by 3,455 · cited 1×
“RL algorithms are commonly used in game playing [17, 18] and robotics [19-23].”
MethodGail, M. H., Brinton, L. A., Byar, D. P., Corle, D. K., Green, S. B., Schairer, C., & Mulvihill, J. J. (1989) · JNCI Journal of the National Cancer Institute · doi:10.1093/jnci/81.24.1879 · Cited by 3,334 · cited 2×
“Our study is complementary to a rich body of work surrounding risk-based screening [7, 28-30].”
BackgroundResultLearning Phrase Representations using RNN Encoder-Decoder for Statistical Machine Translation
Cho, K., van Merrienboer, B., Gulcehre, C., Bahdanau, D., Bougares, F., Schwenk, H., & Bengio, Y. (2014) · arXiv · doi:10.48550/arxiv.1406.1078 · Cited by 3,317 · cited 1×
“For our RNN model, we utilized a Gated Recurrent Unit (GRU) [47] with an additive hazard layer [12] to estimate future risk.”
MethodInternational evaluation of an AI system for breast cancer screening
McKinney, S. M., Sieniek, M., Godbole, V., Godwin, J., Antropova, N., Ashrafian, H., Back, T., Chesus, M., Corrado, G. S., Darzi, A., et al. (2020) · Nature · doi:10.1038/s41586-019-1799-6 · Cited by 3,287
Screening for Lung Cancer: U.S. Preventive Services Task Force Recommendation Statement
Moyer, V. A., & on behalf of the U.S. Preventive Services Task Force* (2014) · Annals of Internal Medicine · doi:10.7326/m13-2771 · Cited by 2,632
Screening for Colorectal Cancer: US Preventive Services Task Force Recommendation Statement
US Preventive Services Task Force, Bibbins-Domingo, K., Grossman, D. C., Curry, S. J., Davidson, K. W., Epling, J. W., García, F. A. R., Gillman, M. W., Harper, D. M., Kemper, A. R., et al. (2016) · JAMA · doi:10.1001/jama.2016.5989 · Cited by 2,509
Screening for breast cancer: U.S. Preventive Services Task Force recommendation statement
Siu, A. L., & on behalf of the U.S. Preventive Services Task Force (2016) · Annals of Internal Medicine · doi:10.7326/m15-2886 · Cited by 2,094
Cited by
Swanson, K., Wu, E., Zhang, A., Alizadeh, A. A., & Zou, J. (2023) · Cell · doi:10.1016/j.cell.2023.01.035 · Cited by 597
“28 Yala et al. demonstrated that reinforcement learning, an area of machine learning that involves deciding which actions to take to maximize a reward, can be used in conjunction with DL risk prediction models to automatically design an optimal screening schedule for each patient individually.”
BackgroundMethodBreast cancer detection using deep learning: Datasets, methods, and challenges ahead
din, N. M. U., Dar, R. A., Rasool, M., & Assad, A. (2022) · Computers in Biology and Medicine · doi:10.1016/j.compbiomed.2022.106073 · Cited by 363
Empowering biomedical discovery with AI agents
Gao, S., Fang, A., Huang, Y., Giunchiglia, V., Noori, A., Schwarz, J. R., Ektefaie, Y., Kondic, J., & Zitnik, M. (2024) · Cell · doi:10.1016/j.cell.2024.09.022 · Cited by 315
A guide to artificial intelligence for cancer researchers
Perez-Lopez, R., Ghaffari Laleh, N., Mahmood, F., & Kather, J. N. (2024) · Nature Reviews Cancer · doi:10.1038/s41568-024-00694-7 · Cited by 211
A Primer on Reinforcement Learning in Medicine for Clinicians
Jayaraman, P., Desman, J. M., Sabounchi, M., Nadkarni, G. N., & Sakhuja, A. (2024) · npj Digital Medicine · doi:10.1038/s41746-024-01316-0 · Cited by 193
“Implicit Q-Learning (IQL) 34 IQLaddressesthechallengesoftraditionalQ-learningmethodsbyleveragingimplicitestimationtechniquesforimprovedpolicy optimization.”
Influential
Metadata
- Authors
- Yala, A., Mikhael, P. G., Lehman, C., Lin, G., Strand, F., Wan, Y.-L., Hughes, K., Satuluru, S., Kim, T., Banerjee, I., Gichoya, J., Trivedi, H., & Barzilay, R.
- Affiliations
- Massachusetts Institute of Technology (US)
- Harvard University (US)
- Chang Gung University (TW)
- Karolinska University Hospital (SE)
- Medical University of South Carolina (US)
- Emory University (US)
- Georgia Institute of Technology (US)
- Mayo Clinic (US)
- Corresponding author
- Yala, Adam
- ORCID
- Venue
- Nature Medicine
- Published
- 2022-01
- Volume
- 28
- Issue
- 1
- Pages
- 136-143
- ISSN
- 1078-8956 (ISSN-L)
- 1546-170X
- OpenAlex
- W4226269440
- Cited by
- 103
- References
- 51
- Citations by year
- 2026: 18 · 2025: 23 · 2024: 27 · 2023: 22 · 2022: 13
- FWCI
- 6.97
- Type
- journal article
- Language
- en
- Open access
- green
- doi.org
- Licence
- cc-by
- www.springer.com/tdm (tdm)
- www.springer.com/tdm (vor)
- Locations
- Nature Medicine (publishedVersion)
- Research Square (acceptedVersion, cc-by)
- PubMed (publishedVersion)
- DSpace@MIT (Massachusetts Institute of Technology) (submittedVersion, cc-by-nc-sa)
- Funders
- Susan G. Komen
- MIT J-Clinic Quanta Computing Anonymous Foundation
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › AI in cancer detection
- Health Sciences › Medicine › Oncology › Global Cancer Incidence and Screening
- Health Sciences › Medicine › Radiology, Nuclear Medicine and Imaging › Digital Radiography and Breast Imaging
- Keywords
- breast cancer screening, reinforcement learning, personalized screening, overscreening, early detection, mammography, screening frequency, clinical risk model, screening costs
- MeSH terms
- Artificial Intelligence, Breast Neoplasms, Female, Humans, Mammography, Early Detection of Cancer
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
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: Optimizing risk-based breast cancer screening policies with reinforcement learning. https://app.cochat.ai/papers/10.1038/s41591-021-01599-w
- The paper
- Yala, A., Mikhael, P. G., Lehman, C., Lin, G., Strand, F., Wan, Y.-L., Hughes, K., Satuluru, S., Kim, T., Banerjee, I., Gichoya, J., Trivedi, H., & Barzilay, R. (2022). Optimizing risk-based breast cancer screening policies with reinforcement learning. Nature Medicine. https://doi.org/10.1038/s41591-021-01599-w
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.