O022 Real-time artificial intelligence instructor vs expert instruction in teaching of expert level tumour resection skills – a randomized controlled trial
Bibliographic record
Abstract
Abstract Introduction Competency-based approach in surgical training still lacks objective quantifiable methodologies to assess surgical technical skills and train residents, a limitation that can be addressed with the adaptation of artificial intelligence (AI). In this randomized controlled, the efficacy of learning by a real-time intelligent instruction system was compared to learning with in-person human instructor-mediated training. Methods The study was ethics approved. Ninety-eight medical students performed five virtually simulated brain tumour resections, randomly allocated into three feedback groups: (1) no-real-time feedback, (2) real-time intelligent instruction, and (3) in-person human instruction. The first task was considered as baseline performance, done with no feedback. Group-1 received expert benchmark feedback only after each procedure. Group-2 was instructed in real-time by the AI system. After each task, the students were shown their error-video clips generated by this system alongside the expert-level demonstrations relating to each error. Group-3 was instructed by human instructors during the tasks. After each task, instructors summarized the areas of improvement and demonstrated how to expertly perform the tumour resections. Participant data in all tasks were scored by the AI system to assess learning. Results Students in Group-2 and Group-3 significantly improved their performance score by the third and second task, respectively (p<0.01, p=0.01), compared to the baseline performance. Group-2 achieved significantly higher scores than Group-3 in the final/fifth task (p<0.01). Conclusion AI-powered systems may increase efficiency in learning by providing objective, and action-oriented real-time feedback. Such systems may aid the shift towards competency-based surgical curricula.
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 | 0.007 | 0.009 |
| Meta-epidemiology (narrow) | 0.002 | 0.001 |
| Meta-epidemiology (broad) | 0.006 | 0.002 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.003 |
| Scholarly communication | 0.002 | 0.002 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.005 | 0.004 |
| Insufficient payload (model declined to judge) | 0.013 | 0.002 |
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".