Canonical text (the exact bytes fingerprinted)

TRUVACE RECORD VERSION
record: TRV-2026-0791
version: 1
kind: certified
reason: Certified into the record
timestamp: 2026-08-16T06:23:25.933056Z
status: published
lens: g_space
sector: health
headline: Real-time artificial intelligence-based anatomy recognition in single-port transvesical enucleation of the prostate
dek: To evaluate the feasibility and accuracy of an artificial intelligence (AI) model to assist surgeons through automated real-time detection and segmentation of key anatomical structures during robot-assisted single-port transvesical enucleation of the prostate (STEP). This retrospective single-centre study utilised surgical videos from patients undergoing single-port robot-assisted transvesical prostate enucleation performed by a single expert surgeon. Selected frames extracted from these surgical videos were man…
gain_title: An AI model trained on surgical video frames achieved automated real-time detection and segmentation of bladder neck, adenoma and peripheral zone during robot-assisted prostate enucleation with high Dice scores and >60 fps inference.
problem_title: (none)
trace_subject: (none)
gain_reading: An AI model trained on surgical video frames achieved automated real-time detection and segmentation of bladder neck, adenoma and peripheral zone during robot-assisted prostate enucleation with high Dice scores and >60 fps inference.
gain_evidence: The model demonstrated strong detection performance, with class-specific F1-scores of 0.88 (adenoma), 0.69 (bladder neck), and 0.70 (peripheral zone). | The model consistently operated at >60 frames/s, confirming real-time applicability without perceptible lag.
problem_reading: (none)
problem_evidence: (none)
quick_read: In a retrospective single-centre pilot, investigators developed a YOLOv11-based model to automatically detect and segment three key landmarks during robot-assisted single-port transvesical enucleation of the prostate using 611 annotated frames from 37 procedures performed by one expert surgeon.

The work matters because real-time anatomical guidance could reduce disorientation and improve safety in a complex endoscopic prostate procedure, but as of the August 2026 publication date the evidence is limited to offline video analysis without prospective clinical deployment, multi-surgeon validation, or patient outcome data.
limitation: Pilot feasibility study was retrospective, single-centre, and based on videos from a single expert surgeon with only 611 annotated frames, limiting generalizability.
tag: Evidence-backed gain
key_points: Retrospective study used 611 annotated frames from 37 surgical videos of single-port robot-assisted transvesical prostate enucleation. | Convolutional neural network based on You Only Look Once version 11 architecture trained to detect bladder neck, prostatic adenoma, and peripheral zone. | Segmentation accuracy by Dice Similarity Coefficient was 0.86 for adenoma, 0.83 for bladder neck, and 0.82 for peripheral zone with mean AP 0.43 across classes.
rundown: Researchers extracted and manually annotated frames to label bladder neck, prostatic adenoma and peripheral zone during the key step of STEP, then trained a YOLO version 11 convolutional network and evaluated it with recall, precision, F1-score, Intersection over Union, Dice Similarity Coefficient and mean average precision.

Quantitative results showed F1-scores of 0.88, 0.69 and 0.70 for adenoma, bladder neck and peripheral zone respectively, Dice scores of 0.86, 0.83 and 0.82, and mean AP of 0.43, with qualitative visual inspection and frame inference speed measurement confirming real-time operation.
sources:
- peer_reviewed | BJU International | https://doi.org/10.1111/bju.70424 | 2026-08-13
prev: 0000000000000000000000000000000000000000000000000000000000000000

sha256

223c0ff270653891bf6d70109614d0a1ea1efb9f4f28a05a705d5afe97c6dd43

previous

0000000000000000000000000000000000000000000000000000000000000000