Unpaired deep adversarial learning for multi‐class segmentation of instruments in robot‐assisted surgical videos
The International Journal of Medical Robotics and Computer Assisted Surgery · 2023-08 · 19 (4)
DOI: 10.1002/rcs.2514 ↗Kurz gefasst
Fachartikel in der Zeitschrift The International Journal of Medical Robotics and Computer Assisted Surgery, online erschienen am 28.03.2023. Verfasst von Shubhangi Nema und 1 weiteren Autor:innen. Beteiligte Einrichtung: Systems and Control Group, Indian Institute of Technology Bombay, Mumbai, Maharastra, India. Hauptthemen laut MEDLINE: Robotic Surgical Procedures, Robotics. Die Arbeit wurde bisher 9-mal zitiert (Crossref) und nennt 35 Literaturangaben. Gefördert von: Prime Minister's Research Fellows scheme.
Abstract (Originalsprache)
Purpose
Image segmentation of instruments in the raw surgical videos is a critical component of intraoperative assistance softwares. Challenges include addressing rendered overlays occluding the instrument while providing pivotal input to instrument tracking frameworks and, train the segmentation process with limited labelled data available from surgical videos.
Method
The proposed adversarial network, InstruSegNet uses unpaired training (eliminating need for massive paired data) for automated multi‐class surgical instrument segmentation in raw surgical videos with complex backgrounds. The proposed method is applied for single/multiple robotic and rigid instruments and optimised on least square Generative Adversarial Networks loss.
Result
Promising validation has been conducted on the publicly available dataset. Proposed approach for multi‐class segmentation of robotic and rigid instruments meets outstanding performance in terms of accuracy and surpasses the existing methods. Application This work facilitates segmenting instrument information without manual interventions from raw videos providing means to code surgeon's actions for developing intelligent assistance software.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Systems and Control Group, Indian Institute of Technology Bombay, Mumbai, Maharastra, India.
ORCID-Profile: Shubhangi Nema, Leena Vachhani
Förderung
- Prime Minister's Research Fellows scheme
Zitieren
APA
Nema, S., & Vachhani, L. (2023). Unpaired deep adversarial learning for multi‐class segmentation of instruments in robot‐assisted surgical videos. The International Journal of Medical Robotics and Computer Assisted Surgery, 19(4). https://doi.org/10.1002/rcs.2514
Vancouver
Nema S, Vachhani L. Unpaired deep adversarial learning for multi‐class segmentation of instruments in robot‐assisted surgical videos. The International Journal of Medical Robotics and Computer Assisted Surgery. 2023;19(4). doi:10.1002/rcs.2514
BibTeX
@article{Nema2023,
author = {Nema, Shubhangi and Vachhani, Leena},
title = {Unpaired deep adversarial learning for multi‐class segmentation of instruments in robot‐assisted surgical videos},
journal = {The International Journal of Medical Robotics and Computer Assisted Surgery},
year = {2023},
volume = {19},
number = {4},
doi = {10.1002/rcs.2514}
}
EUVIMED