Learned video measurements associate surgical instrument use with operative experience
Source check: Source checked. Not independent scientific confirmation.
Finding: DeepLab-based recognition measured instrument appearance in 55 recorded robotic gastric-cancer operations. The study found shorter relative stapler and large-clip appearance times in the later-experience group than in the earlier cases. These are retrospective associations in four surgeons’ procedures, not a validated general skill score or evidence that AI improved surgery or patient outcomes.
Technology: DeepLab v3+ learned surgical-instrument image segmentation
Publication: Peer-reviewed paper · Test context: retrospective or benchmark. Catalog categories, not an evidence grade.
- AI contribution
- DeepLab v3+ learns instrument segmentation from annotated frames, then supplies per-instrument appearance-duration measurements for the retrospective experience-group comparison.
- Test and result
- An instrument-recognition model trained on 1234 annotated video images and assessed on 149 additional images was applied to 55 recorded operations. Comparing 22 earlier-experience cases with 33 later cases yielded associations between relative instrument-appearance time and experience.
- Limitation / still unknown
- Frames were randomly selected from the 55 videos; separation by operation or surgeon in image testing is not established. Instrument appearance is an estimated proxy for use, not independently validated surgical skill. Case mix and operation duration differed, surgeon-level clustering is not established, and outcome differences do not demonstrate an AI intervention effect. The model sometimes confused partly visible instruments.
Evidence sources and annotation context
Evidence context (descriptions follow the cited sources; test type is a catalog classification):
Evidence sources: https://pmc.ncbi.nlm.nih.gov/articles/PMC11216797/
Source checked · editorial details
The source page was checked against this entry. Independent confirmation is not assessed by this label.
Source type: paper. Paper: not linked in this catalog; this does not mean no paper exists.
Review date: 21 Nov 2026. Review scope: Exact primary front identity/date, clinical cohort, instrument-model/statistical Methods, image evaluation and experience-group instrument-duration Results/Table3. Surgical procedure instructions are outside the represented evidence; fresh primary notice metadata checked..
No follow-up recorded; this is not a claim that no follow-up exists. Follow-up review scope and date are not recorded. No source-supported relationships recorded.
Editorial assessment: Impact 2/5 · Validation 2/5 · Novelty 2/5. Impact and novelty are editorial judgments; validation follows the annotated test type where one exists. Not an evidence grade. Scoring criteria.