Toward Generalizability in the Deployment of Artificial Intelligence in Radiology: Role of Computation Stress Testing to Overcome Underspecification
Open accessThe DOI matches its Crossref record; 75 of the 76 references it cites were found in a registry or an index, and 1 are not checked yet.
Abstract
The clinical deployment of artificial intelligence (AI) applications in medical imaging is perhaps the greatest challenge facing radiology in the next decade. One of the main obstacles to the incorporation of automated AI-based decision-making tools in medicine is the failure of models to generalize when deployed across institutions with heterogeneous populations and imaging protocols. The most well-understood pitfall in developing these AI models is overfitting, which has, in part, been overcome by optimizing training protocols. However, overfitting is not the only obstacle to the success and generalizability of AI. Underspecification is also a serious impediment that requires conceptual understanding and correction. It is well known that a single AI pipeline, with prescribed training and testing sets, can produce several models with various levels of generalizability. Underspecification defines the inability of the pipeline to identify whether these models have embedded the structure of the underlying system by using a test set independent of, but distributed identically, to the training set. An underspecified pipeline is unable to assess the degree to which the models will be generalizable. Stress testing is a known tool in AI that can limit underspecification and, importantly, assure broad generalizability of AI models. However, the application of stress tests is new in radiologic applications. This report describes the concept of underspecification from a radiologist perspective, discusses stress testing as a specific strategy to overcome underspecification, and explains how stress tests could be designed in radiology—by modifying medical images or stratifying testing datasets. In the upcoming years, stress tests should become in radiology the standard that crash tests have become in the automotive industry. Keywords: Computer Applications-General, Informatics, Computer-aided Diagnosis © RSNA, 2021
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 cited source says what the claim attributes to it.
In the citing paper
“A critical barrier to the clinical implementation of AI decision tools is their frequent failure to generalize across heterogeneous populations and imaging protocols (Eche et al., 2021).”
From Multimodal neuroimaging and AI integration in cognitive disorders: advances, challenges, and future directions for precision medicine (2026) · cites this paper as Eche et al., 2021
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
Lecture Notes in Computer Science 1205
(1999) · Industrial Robot: An International Journal · doi:10.1108/ir.1999.04926fae.001 · Cited by 38,662
Federated Machine Learning: Concept and Applications
Yang, Q., Liu, Y., Chen, T., & Tong, Y. (2019) · ACM Transactions on Intelligent Systems and Technology · doi:10.1145/3298981 · Cited by 6,322
Radiomics: the bridge between medical imaging and personalized medicine
Lambin, P., Leijenaar, R. T. H., Deist, T. M., Peerlings, J., de Jong, E. E. C., van Timmeren, J., Sanduleanu, S., Larue, R. T. H. M., Even, A. J. G., Jochems, A., et al. (2017) · Nature Reviews Clinical Oncology · doi:10.1038/nrclinonc.2017.141 · Cited by 5,688
Zwanenburg, A., Vallières, M., Abdalah, M. A., Aerts, H. J. W. L., Andrearczyk, V., Apte, A., Ashrafinia, S., Bakas, S., Beukinga, R. J., Boellaard, R., et al. (2020) · Radiology · doi:10.1148/radiol.2020191145 · Cited by 4,177
Simple and Scalable Predictive Uncertainty Estimation using Deep Ensembles
Lakshminarayanan, B., Pritzel, A., & Blundell, C. (2016) · arXiv · doi:10.48550/arxiv.1612.01474 · Cited by 3,666
Neural Networks and the Bias/Variance Dilemma
Geman, S., Bienenstock, E., & Doursat, R. (1992) · Neural Computation · doi:10.1162/neco.1992.4.1.1 · Cited by 3,578
Shortcut learning in deep neural networks
Geirhos, R., Jacobsen, J.-H., Michaelis, C., Zemel, R., Brendel, W., Bethge, M., & Wichmann, F. A. (2020) · Nature Machine Intelligence · doi:10.1038/s42256-020-00257-z · Cited by 2,224
Li, L., Qin, L., Xu, Z., Yin, Y., Wang, X., Kong, B., Bai, J., Lu, Y., Fang, Z., Song, Q., et al. (2020) · Radiology · doi:10.1148/radiol.2020200905 · Cited by 1,927
Neural Architecture Search: A Survey
Elsken, T., Metzen, J. H., & Hutter, F. (2018) · arXiv · doi:10.48550/arxiv.1808.05377 · Cited by 1,763
Preparing Medical Imaging Data for Machine Learning
Willemink, M. J., Koszek, W. A., Hardell, C., Wu, J., Fleischmann, D., Harvey, H., Folio, L. R., Summers, R. M., Rubin, D. L., & Lungren, M. P. (2020) · Radiology · doi:10.1148/radiol.2020192224 · Cited by 1,020
Research works printed but not linked: 1
- Elsken T, 2019, J Mach Learn Res, 20, 1
Cited by
Methods for clinical evaluation of artificial intelligence algorithms for medical diagnosis
Park, S. H., Han, K., Jang, H. Y., Park, J. E., Lee, J.-G., Kim, D. W., & Choi, J. (2023) · Radiology · doi:10.1148/radiol.220182 · Cited by 163
Rahimi, A. K., Pienaar, O., Ghadimi, M., Canfell, O. J., Pole, J. D., Shrapnel, S., van der Vegt, A. H., & Sullivan, C. M. (2024) · Journal of Medical Internet Research · doi:10.2196/49655 · Cited by 164
Deep learning in cancer genomics and histopathology
Unger, M., & Kather, J. N. (2024) · Genome Medicine · doi:10.1186/s13073-024-01315-6 · Cited by 150
Generalizability of FDA-Approved AI-Enabled Medical Devices for Clinical Use
Windecker, D., Baj, G., Shiri, I., Kazaj, P. M., Kaesmacher, J., Gräni, C., & Siontis, G. C. M. (2025) · JAMA Network Open · doi:10.1001/jamanetworkopen.2025.8052 · Cited by 142
Mitigating Bias in Radiology Machine Learning: 3. Performance Metrics
Faghani, S., Khosravi, B., Zhang, K., Moassefi, M., Jagtap, J. M., Nugen, F., Vahdati, S., Kuanar, S. P., Rassoulinejad-Mousavi, S. M., Singh, Y., et al. (2022) · Radiology: Artificial Intelligence · doi:10.1148/ryai.220061 · Cited by 94
Metadata
- Authors
- Eche, T., Schwartz, L. H., Mokrane, F.-Z., & Dercle, L.
- Affiliations
- NewYork–Presbyterian Hospital (US)
- Venue
- Radiology: Artificial Intelligence
- Published
- 2021-11-01
- Volume
- 3
- Issue
- 6
- Pages
- e210097-e210097
- ISSN
- 2638-6100 (ISSN-L)
- PMC
- PMC8637230
- OpenAlex
- W3210784269
- Cited by
- 163
- References
- 71
- Citations by year
- 2026: 33 · 2025: 40 · 2024: 47 · 2023: 30 · 2022: 13
- FWCI
- 6.56
- Type
- journal article
- Language
- en
- Open access
- green
- www.ncbi.nlm.nih.gov
- Locations
- Radiology Artificial Intelligence (publishedVersion)
- PubMed Central (publishedVersion)
- PubMed (publishedVersion)
- Funders
- Alain Rahmouni
- Topics
- Health Sciences › Medicine › Health Informatics › Artificial Intelligence in Healthcare and Education
- Health Sciences › Medicine › Radiology, Nuclear Medicine and Imaging › Radiomics and Machine Learning in Medical Imaging
- Health Sciences › Medicine › Radiology, Nuclear Medicine and Imaging › Radiology practices and education
- Keywords
- underspecification, stress testing, generalizability, AI deployment, diagnostic imaging, overfitting, model evaluation, test set design, artificial intelligence in radiology, computer-aided diagnosis
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: Toward Generalizability in the Deployment of Artificial Intelligence in Radiology: Role of Computation Stress Testing to Overcome Underspecification. https://app.cochat.ai/papers/10.1148/ryai.2021210097
- The paper
- Eche, T., Schwartz, L. H., Mokrane, F.-Z., & Dercle, L. (2021). Toward Generalizability in the Deployment of Artificial Intelligence in Radiology: Role of Computation Stress Testing to Overcome Underspecification. Radiology: Artificial Intelligence. https://doi.org/10.1148/ryai.2021210097
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.