Randomized Controlled Trials Evaluating AI in Clinical Practice: A Scoping Evaluation
Bibliographic record
Abstract
ABSTRACT Background Artificial intelligence (AI) has emerged as a promising tool in healthcare, with numerous studies indicating its potential to perform as well or better than clinicians. However, a considerable portion of these AI models have only been tested retrospectively, raising concerns about their true effectiveness and potential risks in real-world clinical settings. Methods We conducted a systematic search for randomized controlled trials (RCTs) involving AI algorithms used in various clinical practice fields and locations, published between January 1, 2018, and August 18, 2023. Our study included 84 trials and focused specifically on evaluating intervention characteristics, study endpoints, and trial outcomes, including the potential of AI to improve care management, patient behavior and symptoms, and clinical decision-making. Results Our analysis revealed that 82·1% (69/84) of trials reported positive results for their primary endpoint, highlighting AI’s potential to enhance various aspects of healthcare. Trials predominantly evaluated deep learning systems for medical imaging and were conducted in single-center settings. The US and China had the most trials, with gastroenterology being the most common field of study. However, we also identified areas requiring further research, such as multi-center trials and diverse outcome measures, to better understand AI’s true impact and limitations in healthcare. Conclusion The existing landscape of RCTs on AI in clinical practice demonstrates an expanding interest in applying AI across a range of fields and locations. While most trials report positive outcomes, more comprehensive research, including multi-center trials and diverse outcome measures, is essential to fully understand AI’s impact and limitations in healthcare.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.180 | 0.467 |
| Meta-epidemiology (narrow) | 0.003 | 0.002 |
| Meta-epidemiology (broad) | 0.015 | 0.023 |
| Bibliometrics | 0.010 | 0.011 |
| Science and technology studies | 0.002 | 0.004 |
| Scholarly communication | 0.007 | 0.007 |
| Open science | 0.004 | 0.003 |
| Research integrity | 0.007 | 0.004 |
| Insufficient payload (model declined to judge) | 0.010 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".