Generalizability of Machine Learning Models: Quantitative Evaluation of Three Methodological Pitfalls
Open accessThe DOI matches its Crossref record; 14 of the 14 references it cites were found in a registry or an index.
Abstract
Purpose To investigate the impact of the following three methodological pitfalls on model generalizability: (a) violation of the independence assumption, (b) model evaluation with an inappropriate performance indicator or baseline for comparison, and (c) batch effect. Materials and Methods The authors used retrospective CT, histopathologic analysis, and radiography datasets to develop machine learning models with and without the three methodological pitfalls to quantitatively illustrate their effect on model performance and generalizability. F1 score was used to measure performance, and differences in performance between models developed with and without errors were assessed using the Wilcoxon rank sum test when applicable. Results Violation of the independence assumption by applying oversampling, feature selection, and data augmentation before splitting data into training, validation, and test sets seemingly improved model F1 scores by 71.2% for predicting local recurrence and 5.0% for predicting 3-year overall survival in head and neck cancer and by 46.0% for distinguishing histopathologic patterns in lung cancer. Randomly distributing data points for a patient across datasets superficially improved the F1 score by 21.8%. High model performance metrics did not indicate high-quality lung segmentation. In the presence of a batch effect, a model built for pneumonia detection had an F1 score of 98.7% but correctly classified only 3.86% of samples from a new dataset of healthy patients. Conclusion Machine learning models developed with these methodological pitfalls, which are undetectable during internal evaluation, produce inaccurate predictions; thus, understanding and avoiding these pitfalls is necessary for developing generalizable models. Keywords: Random Forest, Diagnosis, Prognosis, Convolutional Neural Network (CNN), Medical Image Analysis, Generalizability, Machine Learning, Deep Learning, Model Evaluation Supplemental material is available for this article. Published under a CC BY 4.0 license.
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
QUADAS-2: a revised tool for the quality assessment of diagnostic accuracy studies
Whiting, P. F., Rutjes, A. W. S., Westwood, M. E., Mallett, S., Deeks, J. J., Reitsma, J. B., Leeflang, M. M. G., Sterne, J. A. C., Bossuyt, P. M. M., & the QUADAS-2 Group* (2011) · Annals of Internal Medicine · doi:10.7326/0003-4819-155-8-201110180-00009 · Cited by 14,199
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
Decoding tumour phenotype by noninvasive imaging using a quantitative radiomics approach
Aerts, H. J. W. L., Velazquez, E. R., Leijenaar, R. T. H., Parmar, C., Grossmann, P., Carvalho, S., Bussink, J., Monshouwer, R., Haibe-Kains, B., Rietveld, D., et al. (2014) · Nature Communications · doi:10.1038/ncomms5006 · Cited by 5,132 · cited 2×
“We used a CT dataset of 137 head and neck squamous cell carcinoma (HNSCC) patients treated by radiotherapy.(15,16) Hereafter, we refer to this dataset as the “HNSCC” dataset.”
MethodIdentifying Medical Diagnoses and Treatable Diseases by Image-Based Deep Learning
Kermany, D. S., Goldbaum, M., Cai, W., Valentim, C. C. S., Liang, H., Baxter, S. L., McKeown, A., Yang, G., Wu, X., Yan, F., et al. (2018) · Cell · doi:10.1016/j.cell.2018.02.010 · Cited by 4,842
The Cancer Imaging Archive (TCIA): Maintaining and Operating a Public Information Repository
Clark, K., Vendt, B., Smith, K., Freymann, J., Kirby, J., Koppel, P., Moore, S., Phillips, S., Maffitt, D., Pringle, M., et al. (2013) · Journal of Digital Imaging · doi:10.1007/s10278-013-9622-7 · Cited by 4,599 · cited 2×
“We used the images from the Lung CT Segmentation Challenge 2017, extracted from TCIA, containing 120 CT scan series from 60 patients (19-21).”
MethodKey challenges for delivering clinical impact with artificial intelligence
Kelly, C. J., Karthikesalingam, A., Suleyman, M., Corrado, G., & King, D. (2019) · BMC Medicine · doi:10.1186/s12916-019-1426-2 · Cited by 3,136
PROBAST: A Tool to Assess the Risk of Bias and Applicability of Prediction Model Studies
Wolff, R. F., Moons, K. G. M., Riley, R. D., Whiting, P. F., Westwood, M., Collins, G. S., Reitsma, J. B., Kleijnen, J., Mallett, S., & for the PROBAST Group† (2019) · Annals of Internal Medicine · doi:10.7326/m18-1376 · Cited by 2,984 · cited 2×
“designed PROBAST as a series of questions to facilitate systematic review and assessment of potential bias in clinical prediction models (16).”
BackgroundAlbumentations: fast and flexible image augmentations
Buslaev, A., Iglovikov, V. I., Khvedchenya, E., Parinov, A., Druzhinin, M., & Kalinin, A. A. (2020) · Information · doi:10.3390/info11020125 · Cited by 2,714
Collins, G. S., Reitsma, J. B., Altman, D. G., & Moons, K. (2015) · BMC Medicine · doi:10.1186/s12916-014-0241-z · Cited by 1,927
Checklist for Artificial Intelligence in Medical Imaging (CLAIM): a guide for authors and reviewers
Mongan, J., Moy, L., & Kahn, C. E. (2020) · Radiology: Artificial Intelligence · doi:10.1148/ryai.2020200029 · Cited by 1,443 · cited 3×
“published the CLAIM checklist to aid authors and reviewers with best practices in artificial intelligence research in medical imaging (14).”
BackgroundMethod
Cited by
Scientific evidence for clinical text summarization using large language models: scoping review
Bednarczyk, L., Reichenpfader, D., Gaudet-Blavignac, C., Ette, A. K., Zaghir, J., Zheng, Y., Bensahla, A., Bjelogrlic, M., & Lovis, C. (2025) · Journal of Medical Internet Research · doi:10.2196/68998 · Cited by 83
Deep learning in rheumatological image interpretation
Stoel, B. C., Staring, M., Reijnierse, M., & van der Helm-van Mil, A. H. M. (2024) · Nature Reviews Rheumatology · doi:10.1038/s41584-023-01074-5 · Cited by 25
“Errors in training and validation design can result in considerable over-estimation of the performance of deep-learning methods 77 .”
Assessing retrieval-augmented large language models for medical coding
Klang, E., Tessler, I., Apakama, D. U., Abbott, E., Glicksberg, B. S., Arnold, M., Moses, A., Sakhuja, A., Soroush, A., Charney, A. W., et al. (2025) · NEJM AI · doi:10.1056/aics2401161 · Cited by 6
Metadata
- Authors
- Maleki, F., Ovens, K., Gupta, R., Reinhold, C., Spatz, A., & Forghani, R.
- Venue
- Radiology: Artificial Intelligence
- Published
- 2023-01-01
- Volume
- 5
- Issue
- 1
- Pages
- e220028-e220028
- ISSN
- 2638-6100 (ISSN-L)
- OpenAlex
- W4309321702
- Cited by
- 150
- References
- 14
- FWCI
- 13.52
- Type
- journal article
- Language
- en
- Open access
- hybrid
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: Generalizability of Machine Learning Models: Quantitative Evaluation of Three Methodological Pitfalls. https://app.cochat.ai/papers/10.1148/ryai.220028
- The paper
- Maleki, F., Ovens, K., Gupta, R., Reinhold, C., Spatz, A., & Forghani, R. (2023). Generalizability of Machine Learning Models: Quantitative Evaluation of Three Methodological Pitfalls. Radiology: Artificial Intelligence. https://doi.org/10.1148/ryai.220028
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.