How AI should be used in radiology: assessing ambiguity and completeness of intended use statements of commercial AI products
Open accessAbstract
Background: Intended use statements (IUSs) are mandatory to obtain regulatory clearance for artificial intelligence (AI)-based medical devices in the European Union. In order to guide the safe use of AI-based medical devices, IUSs need to contain comprehensive and understandable information. This study analyzes the IUSs of CE-marked AI products listed on AIforRadiology.com for ambiguity and completeness. Methods: We retrieved 157 IUSs of CE-marked AI products listed on AIforRadiology.com in September 2022. Duplicate products (n = 1), discontinued products (n = 3), and duplicate statements (n = 14) were excluded. The resulting IUSs were assessed for the presence of 6 items: medical indication, part of the body, patient population, user profile, use environment, and operating principle. Disclaimers, defined as contra-indications or warnings in the IUS, were identified and compared with claims. Results: Of 139 AI products, the majority (n = 78) of IUSs mentioned 3 or less items. IUSs of only 7 products mentioned all 6 items. The intended body part (n = 115) and the operating principle (n = 116) were the most frequently mentioned components, while the intended use environment (n = 24) and intended patient population (n = 29) were mentioned less frequently. Fifty-six statements contained disclaimers that conflicted with the claims in 13 cases. Conclusion: The majority of IUSs of CE-marked AI-based medical devices lack substantial information and, in few cases, contradict the claims of the product. Critical relevance statement: To ensure correct usage and to avoid off-label use or foreseeable misuse of AI-based medical devices in radiology, manufacturers are encouraged to provide more comprehensive and less ambiguous intended use statements. Key points: • Radiologists must know AI products’ intended use to avoid off-label use or misuse. • Ninety-five percent (n = 132/139) of the intended use statements analyzed were incomplete. • Nine percent (n = 13) of the intended use statements held disclaimers contradicting the claim of the AI product. • Manufacturers and regulatory bodies must ensure that intended use statements are comprehensive. Graphical Abstract:
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
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?
Ambiguities can originate during different applications of AI tools from the intended use statement provided by developers, increasing the risk of off-label or erroneous applications of AI in clinical practice.
“Ambiguities can also originate during the different applications of the AI tools from the intended use statement provided by the AI developers, increasing the risk of off-label or erroneous applications of AI in clinical practice.119”
From Bias in artificial intelligence for medical imaging: fundamentals, detection, avoidance, mitigation, challenges, ethics, and prospects (2024) · cites this paper as [119]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Cites
High-performance medicine: the convergence of human and artificial intelligence
Topol, E. J. (2019) · Nature Medicine · doi:10.1038/s41591-018-0300-7 · Cited by 10,450 · see citing works
Muehlematter, U. J., Daniore, P., & Vokinger, K. N. (2021) · The Lancet Digital Health · doi:10.1016/s2589-7500(20)30292-2 · Cited by 665 · see citing works
van Leeuwen, K. G., Schalekamp, S., Rutten, M. J. C. M., van Ginneken, B., & de Rooij, M. (2021) · European Radiology · doi:10.1007/s00330-021-07892-z · Cited by 461 · see citing works
Self-citationHow does artificial intelligence in radiology improve efficiency and health outcomes?
van Leeuwen, K. G., de Rooij, M., Schalekamp, S., van Ginneken, B., & Rutten, M. J. C. M. (2022) · Pediatric Radiology · doi:10.1007/s00247-021-05114-8 · Cited by 236 · see citing works
Self-citationClark, P., Kim, J., & Aphinyanaphongs, Y. (2023) · JAMA Network Open · doi:10.1001/jamanetworkopen.2023.21792 · Cited by 70 · see citing works
Petrick, N., Chen, W., Delfino, J. G., Gallas, B. D., Kang, Y., Krainak, D. M., Sahiner, B., & Samala, R. K. (2023) · Journal of medical imaging · doi:10.1117/1.jmi.10.5.051804 · Cited by 51 · see citing works
Artificial intelligence tools in clinical neuroradiology: essential medico-legal aspects
Hedderich, D. M., Weisstanner, C., Cauter, S. V., Federau, C., Edjlali, M., Radbruch, A., Gerke, S., & Haller, S. (2023) · Neuroradiology · doi:10.1007/s00234-023-03152-7 · Cited by 37 · see citing works
Self-citationNelson, B., Zeng, R., Sammer, M. B. K., Frush, D. P., & Delfino, J. G. (2023) · Journal of the American College of Radiology · doi:10.1016/j.jacr.2023.06.004 · Cited by 13 · see citing works
Flagged first, then by citations
Cited by
Sorted by citations
Koçak, B., Ponsiglione, A., Stanzione, A., Bluethgen, C., Santinha, J., Ugga, L., Huisman, M., Klontzas, M. E., Cannella, R., & Cuocolo, R. (2024) · Diagnostic and Interventional Radiology · doi:10.4274/dir.2024.242854 · Cited by 168 · see citing works
“119 For example, AI software trained for adult fracture detection is at risk of erroneous results if applied in a pediatric population.”
Automation bias in AI-assisted detection of cerebral aneurysms on time-of-flight MR angiography
Kim, S. H., Schramm, S., Riedel, E. O., Schmitzer, L., Rosenkranz, E., Kertels, O., Bodden, J. H., Paprottka, K. J., Sepp, D., Renz, M., et al. (2025) · La radiologia medica · doi:10.1007/s11547-025-01964-6 · Cited by 36 · see citing works
Self-citationSosna, J., Joskowicz, L., & Saban, M. (2025) · Radiology · doi:10.1148/radiol.240982 · Cited by 31 · see citing works
Recognising errors in AI implementation in radiology: A narrative review
Stogiannos, N., Cuocolo, R., D’Antonoli, T. A., dos Santos, D. P., Harvey, H., Huisman, M., Koçak, B., Kotter, E., Lekadir, K., Shelmerdine, S. C., et al. (2025) · European Journal of Radiology · doi:10.1016/j.ejrad.2025.112311 · Cited by 31 · see citing works
Self-citationSafe for kids? AI medical devices in radiology overlook paediatric suitability
Kelly, B. S., Lee, J., Antram, E., Arthurs, O. J., & Shelmerdine, S. C. (2025) · European Radiology · doi:10.1007/s00330-025-11975-6 · Cited by 10 · see citing works
Howard, J. P., Zhang, Q., Salih, A. M., Petersen, S. E., Lekadir, K., & Raisi‐Estabragh, Z. (2025) · Heart · doi:10.1136/heartjnl-2024-324612 · Cited by 9 · see citing works
Building Trustworthy AI in Healthcare
Amadi, C., & Ojo, A. (2025) · IEEE Access · doi:10.1109/access.2025.3648410 · Cited by 9 · see citing works
Managing expectations and challenges of AI in radiology
Meijer, F. J. A. (2024) · European Radiology · doi:10.1007/s00330-024-10790-9 · Cited by 5 · see citing works
Vendors’ perspectives on AI implementation in medical imaging and oncology: a cross-sectional survey
Stogiannos, N., Skelton, E., van Leeuwen, K. G., Edgington, S., Shelmerdine, S. C., & Malamateniou, C. (2025) · European Radiology · doi:10.1007/s00330-025-12013-1 · Cited by 4 · see citing works
Mascagni, P., Alapatt, D., Sestini, L., Yu, T., Alfieri, S., Morales‐Conde, S., Padoy, N., & Perretta, S. (2024) · Cirugía Española · doi:10.1016/j.ciresp.2024.04.006 · Cited by 3 · see citing works
Definition of the System, Operational Design Domain, and Concept of Operation
Myklebust, T., Stålhane, T., & Vatn, D. M. K. (2025) · SpringerBriefs in computer science · doi:10.1007/978-3-031-80504-2_3 · Cited by 2 · see citing works
Mastrodicasa, D., & van Assen, M. (2024) · BJR|Open · doi:10.1093/bjro/tzaf015 · Cited by 2 · see citing works
Mascagni, P., Alapatt, D., Sestini, L., Yu, T., Alfieri, S., Morales‐Conde, S., Padoy, N., & Perretta, S. (2024) · Cirugía Española (English Edition) · doi:10.1016/j.cireng.2024.04.009 · Cited by 1 · see citing works
Kumbhakarna, V. M., & Kulkarni, S. B. (2026) · SSRN Electronic Journal · doi:10.2139/ssrn.6409036
Gurabi, A., Hosseini-Begtary, S. S., Hegedus, P., Gál, A. V., Maurovich‐Horvat, P., & Marton, N. (2026) · European Radiology Experimental · doi:10.1186/s41747-026-00763-6
Kao, C.-H. (2026) · SSRN Electronic Journal · doi:10.2139/ssrn.7485558
- Authors
- van Leeuwen, K. G., Hedderich, D. M., Harvey, H., & Schalekamp, S.
- Affiliations
- TUM Klinikum (DE)
- Radboud University Nijmegen (NL)
- Corresponding author
- Schalekamp, Steven
- Venue
- Insights into Imaging
- Published
- 2024-02-16
- Volume
- 15
- Issue
- 1
- Pages
- 51-51
- ISSN
- 1869-4101 (ISSN-L)
- PMC
- PMC10873256
- OpenAlex
- W4391885246
- Cited by
- 15
- References
- 8
- Citations by year
- 2026: 2 · 2025: 8 · 2024: 5
- FWCI
- 1.40
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Locations
- Insights into Imaging (publishedVersion, cc-by)
- PubMed Central (publishedVersion, cc-by)
- PubMed (publishedVersion)
- Radboud Repository (Radboud University) (submittedVersion)
- Informationszentrum Lebenswissenschaften (ZB MED) (submittedVersion, cc-by)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion)
- mediaTUM (Technical University of Munich) (submittedVersion)
- Topics
- Health Sciences › Medicine › Health Informatics › Artificial Intelligence in Healthcare and Education
- Health Sciences › Health Professions › Health Information Management › Healthcare Regulation
- Social Sciences › Law › Digitalization, Law, and Regulation
- Keywords
- AI-based medical devices, artificial intelligence in radiology, CE marking, regulatory approval, completeness, disclaimers, off-label drug use, working principle, patient populations
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
