Surgeons versus computer vision: a comparative analysis on surgical phase recognition capabilities
International Journal of Computer Assisted Radiology and Surgery · 31.05.2025 · 20 (6) · 1283-1291
DOI: 10.1007/s11548-025-03383-4 ↗Kurz gefasst
Vergleichsstudie in der Zeitschrift International Journal of Computer Assisted Radiology and Surgery, online erschienen am 31.05.2025. Verfasst von Marco Mezzina und 5 weiteren Autor:innen. Beteiligte Einrichtung: Orsi Academy, Ghent, Belgium. marco.mezzina@orsi.be. Hauptthemen laut MEDLINE: Artificial Intelligence, Robotic Surgical Procedures, Nephrectomy, Clinical Competence. Die Arbeit wurde bisher 8-mal zitiert (Crossref) und nennt 19 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: VLAIO.
Abstract (Originalsprache)
Purpose
Automated surgical phase recognition (SPR) uses artificial intelligence (AI) to segment the surgical workflow into its key events, functioning as a building block for efficient video review, surgical education as well as skill assessment. Previous research has focused on short and linear surgical procedures and has not explored if temporal context influences experts’ ability to better classify surgical phases. This research addresses these gaps, focusing on robot-assisted partial nephrectomy (RAPN) as a highly nonlinear procedure.
Methods
Urologists of varying expertise were grouped and tasked to indicate the surgical phase for RAPN on both single frames and video snippets using a custom-made web platform. Participants reported their confidence levels and the visual landmarks used in their decision-making. AI architectures without and with temporal context as trained and benchmarked on the Cholec80 data set were subsequently trained on this RAPN data set.
Results
Video snippets and presence of specific visual landmarks improved phase classification accuracy across all groups. Surgeons displayed high confidence in their classifications and outperformed novices, who struggled discriminating phases. The performance of the AI models is comparable to the surgeons in the survey, with improvements when temporal context was incorporated in both cases.
Conclusion
SPR is an inherently complex task for expert surgeons and computer vision, where both perform equally well when given the same context. Performance increases when temporal information is provided. Surgical tools and organs form the key landmarks for human interpretation and are expected to shape the future of automated SPR.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Orsi Academy, Ghent, Belgium. marco.mezzina@orsi.be.
- Faculty of Engineering Science, KU Leuven, Leuven, Belgium. marco.mezzina@orsi.be.
- Orsi Academy, Ghent, Belgium.
- Department of Urology, OLV Aalst Hospital, Aalst, Belgium.
- School of BMEIS, King's College London, London, UK.
- Faculty of Engineering Science, KU Leuven, Leuven, Belgium.
ORCID-Profile: Pieter De Backer, Tom Vercauteren, Matthew Blaschko, Alexandre Mottrie, Tinne Tuytelaars
Förderung
- VLAIO
Zitieren
APA
Mezzina, M., De Backer, P., Vercauteren, T., Blaschko, M., Mottrie, A., & Tuytelaars, T. (2025). Surgeons versus computer vision: a comparative analysis on surgical phase recognition capabilities. International Journal of Computer Assisted Radiology and Surgery, 20(6), 1283-1291. https://doi.org/10.1007/s11548-025-03383-4
Vancouver
Mezzina M, De Backer P, Vercauteren T, Blaschko M, Mottrie A, Tuytelaars T. Surgeons versus computer vision: a comparative analysis on surgical phase recognition capabilities. International Journal of Computer Assisted Radiology and Surgery. 2025;20(6):1283-1291. doi:10.1007/s11548-025-03383-4
BibTeX
@article{Mezzina2025,
author = {Mezzina, Marco and De Backer, Pieter and Vercauteren, Tom and Blaschko, Matthew and Mottrie, Alexandre and Tuytelaars, Tinne},
title = {Surgeons versus computer vision: a comparative analysis on surgical phase recognition capabilities},
journal = {International Journal of Computer Assisted Radiology and Surgery},
year = {2025},
volume = {20},
number = {6},
pages = {1283--1291},
doi = {10.1007/s11548-025-03383-4}
}
EUVIMED