Canonical text (the exact bytes fingerprinted)

TRUVACE RECORD VERSION
record: TRV-2026-2032
version: 1
kind: certified
reason: Certified into the record
timestamp: 2026-10-20T18:13:54.231567Z
status: published
lens: g_space
sector: health
headline: AI enhanced surgeons' faster responses to intraoperative bleeding in laparoscopic surgery: a preclinical randomized controlled trial
dek: Objective To develop and validate an artificial intelligence (AI) model for real-time detection and tracking of intraoperative bleeding sites during laparoscopic surgery, and to evaluate whether AI assistance improves surgeons' bleeding‑identification performance. Methods From January to September 2023, laparoscopic surgical videos were retrospectively collected from 20 medical centers in China. Senior surgeons identified bleeding events, extracted relevant video clips, and manually annotated bleeding frames to…
gain_title: In a randomized preclinical trial of 77 surgeons, AI assistance shortened the time to identify a bleeding site in laparoscopic video and lowered the surgeons' negative emotional burden.
problem_title: (none)
trace_subject: (none)
gain_reading: In a randomized preclinical trial of 77 surgeons, AI assistance shortened the time to identify a bleeding site in laparoscopic video and lowered the surgeons' negative emotional burden.
gain_evidence: the experimental group demonstrated a significantly shorter reaction time (2.53 s vs. 3.23 s | AI‑assisted support significantly shortens surgeons' reaction time for bleeding‑site identification
problem_reading: (none)
problem_evidence: (none)
quick_read: Researchers built an AI model that detects and tracks bleeding sites in laparoscopic surgery video. They trained it on footage from 20 medical centers in China, then ran a randomized preclinical trial to test whether it helped surgeons spot bleeding faster.

Faster recognition of bleeding is relevant to surgical safety, and the model also eased surgeons' negative emotional response to bleeding events. The evidence comes from a preclinical setting using recorded video, so effects on live operations and patient outcomes remain untested in this report.
limitation: The trial was preclinical, so the reported benefit applies to bleeding identification in surgical videos rather than to live operations or patient outcomes.
tag: Evidence-backed gain
key_points: The model was trained on 257 laparoscopic videos from 20 medical centers, split into training, validation, and test sets. | On the test set the model reached an F1-score of 0.860 (95% CI 0.838-0.878). | 77 surgeons were randomized: 38 to AI assistance and 39 to a control group. | Reaction time averaged 2.53 s with AI versus 3.23 s in controls.
rundown: The videos were collected retrospectively between January and September 2023. Senior surgeons identified bleeding events, extracted clips, and manually annotated bleeding frames to train the model.

Performance was judged by F1-score, mean absolute error (3.315), and mean distance error (0.087). A separate randomized trial then measured surgeon performance.

The primary outcome was reaction time, defined as the time from bleeding onset to the participant's first successful click on the bleeding site. The trial is registered with the Chinese Clinical Trial Registry (ChiCTR2600118185).
sources:
- peer_reviewed | Surgical Endoscopy | https://doi.org/10.1007/s00464-026-13490-1 | 2026-10-13
prev: 0000000000000000000000000000000000000000000000000000000000000000

sha256

33b353fdc41d196001251424d5c1c6b90f5328f19b71e154a4477d3ec655f535

previous

0000000000000000000000000000000000000000000000000000000000000000