Artificial intelligence for diagnostics in radiology practice: a rapid systematic scoping reviewResearch in context
Summary: Background: The aim of this review was to evaluate evidence on the use of Artificial Intelligence (AI) to support diagnostics in radiology, including implementation, experiences, perceptions, quantitative, and cost outcomes. Methods: We conducted a systematic scoping review (PROSPERO regis...
Saved in:
| Main Authors: | , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-05-01
|
| Series: | EClinicalMedicine |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2589537025001609 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Summary: Background: The aim of this review was to evaluate evidence on the use of Artificial Intelligence (AI) to support diagnostics in radiology, including implementation, experiences, perceptions, quantitative, and cost outcomes. Methods: We conducted a systematic scoping review (PROSPERO registration: CRD42024537518) and discussed emerging findings with relevant stakeholders (radiology staff, public members) using workshops. We searched four databases and the grey literature for articles published between 1st January 2020 and 31st January 2025. Articles were screened for eligibility (N = 8013), resulting in 140 included studies. Studies evaluated implementation (N = 7), perceptions (N = 74), experiences (N = 14), effectiveness (N = 53), and cost (N = 6). Findings: Factors influencing AI adoption were identified, including the high technical demand, lack of guidance, training/knowledge, transparency, and expert engagement. Evidence demonstrated improvements in diagnostic accuracy and reductions in interpretation time. However, evidence was mixed regarding experiences of using AI, the risk of increasing false positives, and the wider impact of AI on workflow efficiency and cost-effectiveness. Interpretation: The potential benefits of AI are evident, but there is a paucity of evidence in real-world settings, supporting cautiousness in how AI is perceived (e.g., as a complementary tool, not a solution). We outline wider implications for policy and practice and summarise evidence gaps. Funding: This project is funded by the National Institute for Health and Care Research, Health and Social Care Delivery Research programme (Ref: NIHR156380). NJF and AIGR are supported by the National Institute for Health Research (NIHR) Central London Patient Safety Research Collaboration and NJF is an NIHR Senior Investigator. The views expressed are those of the author(s) and not necessarily those of the NIHR or the Department of Health and Social Care. |
|---|---|
| ISSN: | 2589-5370 |