The DOI matches its Crossref record.
FDA Review of Radiologic AI Algorithms: Process and Challenges
Abstract
A Food and Drug Administration (FDA)-cleared artificial intelligence (AI) algorithm misdiagnosed a finding as an intracranial hemorrhage in a patient, who was finally diagnosed with an ischemic stroke. This scenario highlights a notable failure mode of AI tools, emphasizing the importance of human-machine interaction. In this report, the authors summarize the review processes by the FDA for software as a medical device and the unique regulatory designs for radiologic AI/machine learning algorithms to ensure their safety in clinical practice. Then the challenges in maximizing the efficacy of these tools posed by their clinical implementation are discussed.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
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?
Recent work has provided insights into subsets of devices, but a comprehensive characterization of FDA-authorized AI-enabled devices is currently lacking.
“Recent work has provided insights into subsets of devices6–11, but a comprehensive characterization of FDA-authorized AI-enabled devices is currently lacking.”
From How AI is used in FDA-authorized medical devices: a taxonomy across 1,016 authorizations (2025) · cites this paper as [8]
Citations
Cites
No references could be resolved yet.
Research works printed but not linked: 17
- Ojeda P, Zawaideh M, Mossa-Basha M, Haynor D. The utility of deep learning: evaluation of a convolutional neural network for detection of intracranial bleeds on non-contrast head computed tomography studies. In: Proc SPIE 10949, Medical Imaging 2019: Image Processing, 109493J (15 March 2019). SPIE, 2019; 899–906. · DOI
- 10.1016/j.acra.2020.01.035 · DOI
- 10.1148/radiol.212830 · DOI
- 10.1148/ryai.2019180089 · DOI
- 10.1148/ryai.210064 · DOI
- 10.1148/radiol.212151 · DOI
- 10.1038/s41591-021-01312-x · DOI · Paper page
- 10.1016/j.ijmedinf.2022.104828 · DOI · Paper page
- 10.1016/S2589-7500(20)30292-2 · DOI · Paper page
- 10.1007/s00234-021-02874-w · DOI
- 10.3174/ajnr.A7179 · DOI
- 10.1148/radiol.213063 · DOI
- 10.1007/s11060-022-04080-x · DOI
- Seyyed-Kalantari L, 2021, Pac Symp Biocomput, 26, 232
- 10.1016/j.acra.2021.09.002 · DOI · Paper page
- 10.1148/ryai.220010 · DOI
- 10.1148/radiol.222217 · DOI
Cited by
Sorted by citations
How AI is used in FDA-authorized medical devices: a taxonomy across 1,016 authorizations
Singh, R., Bapna, M., Diab, A. R., Ruiz, E. S., & Lotter, W. (2025) · npj Digital Medicine · doi:10.1038/s41746-025-01800-1 · Cited by 61
“Thepossibleclassesforeach factorwerede fi ned through the review process to encompass major axes of variationacrossthedevices,andwerealsoinformedbyexistingcategoriesof medicaldevices 7,8 , such as computer-aided detection ( “ CADe ” ) and computer-aided diagnosis ( “ CADx ” ).”
Metadata
- Authors
- Zhang, K., Khosravi, B., Vahdati, S., & Erickson, B. J.
- Affiliations
- Mayo Clinic (US)
- Venue
- Radiology
- Published
- 2024-01-01
- Volume
- 310
- Issue
- 1
- Pages
- e230242-e230242
- ISSN
- 0033-8419 (ISSN-L)
- 1527-1315
- OpenAlex
- W4390507795
- Cited by
- 46
- References
- 0
- Citations by year
- 2026: 21 · 2025: 18 · 2024: 7
- FWCI
- 4.35
- Type
- journal article
- Language
- en
- Open access
- closed
- Locations
- Radiology (publishedVersion)
- PubMed (publishedVersion)
- Topics
- Health Sciences › Medicine › Health Informatics › Artificial Intelligence in Healthcare and Education
- Health Sciences › Medicine › Public Health, Environmental and Occupational Health › Healthcare Regulation
- Physical Sciences › Engineering › Safety, Risk, Reliability and Quality › Safety Systems Engineering in Autonomy
- Keywords
- FDA regulatory review, artificial intelligence algorithms, software as a medical device, machine learning, human-machine interaction, clinical implementation, false positives, ischemic stroke, intracranial hemorrhage
- MeSH terms
- Machine Learning, Algorithms, Artificial Intelligence, Humans, Software, United States, United States Food and Drug Administration
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
