Developing an artificial intelligence model for phase recognition in robot‐assisted radical prostatectomy
BJU International · 2025
DOI: 10.1111/bju.16862 ↗Kurz gefasst
Fachartikel in der Zeitschrift BJU International, 2025. Verfasst von Hideto Ueki und 13 weiteren Autor:innen. Beteiligte Einrichtung: Department of Urology, Kobe University Graduate School of Medicine, Kobe, Japan. Hauptthemen laut MEDLINE: Prostatectomy, Robotic Surgical Procedures, Artificial Intelligence, Prostatic Neoplasms. Die Arbeit wurde bisher 8-mal zitiert (Crossref) und nennt 32 Literaturangaben.
Abstract (Originalsprache)
Objectives
To develop and evaluate a convolutional neural network (CNN)‐based model for recognising surgical phases in robot‐assisted laparoscopic radical prostatectomy (RARP), with an emphasis on model interpretability and cross‐platform validation.
Methods
A CNN using EfficientNet B7 was trained on video data from 75 RARP cases with the hinotori robotic system. Seven phases were annotated: bladder drop, prostate preparation, bladder neck dissection, seminal vesicle dissection, posterior dissection, apical dissection, and vesicourethral anastomosis. A total of 808 774 video frames were extracted at 1 frame/s for training and testing. Validation was performed on 25 RARP cases using the da Vinci robotic system to assess cross‐platform generalisability. Gradient‐weighted class activation mapping was used to enhance interpretability by identifying key regions of interest for phase classification.
Results
The CNN achieved 0.90 accuracy on the hinotori test set but dropped to 0.64 on the da Vinci dataset, thus indicating cross‐platform limitations. Phase‐specific F1 scores ranged from 0.77 to 0.97, with lower performance in the phase of seminal vesicle dissection, and apical dissection. Gradient‐weighted class activation mapping visualisations revealed the model's focus on central pelvic structures rather than transient instruments, enhancing interpretability and insights into phase classification.
Conclusions
The model demonstrated high accuracy on a single robotic platform but requires further refinement for consistent cross‐platform performance. Interpretability techniques will foster clinical trust and integration into workflows, advancing robotic surgery applications.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of Urology, Kobe University Graduate School of Medicine, Kobe, Japan.
- Laboratory of Translational Research for Biologics, Department of Advanced Medical Science, Kobe University Graduate School of Science, Technology and Innovation, Kobe, Japan.
- Department of International Clinical Cancer Research and Promotion, Kobe University Graduate School of Medicine, Kobe, Japan.
ORCID-Profile: Hideto Ueki, Kotaro Suzuki, Takuto Hara
Zitieren
APA
Ueki, H., Uemura, M., Chinzei, K., Takahashi, K., Wakita, N., Okamura, Y., Suzuki, K., Bando, Y., Hara, T., Terakawa, T., Yao, A., Teishima, J., Chiba, K., & Miyake, H. (2025). Developing an artificial intelligence model for phase recognition in robot‐assisted radical prostatectomy. BJU International. https://doi.org/10.1111/bju.16862
Vancouver
Ueki H, Uemura M, Chinzei K, Takahashi K, Wakita N, Okamura Y, et al. Developing an artificial intelligence model for phase recognition in robot‐assisted radical prostatectomy. BJU International. 2025. doi:10.1111/bju.16862
BibTeX
@article{Ueki2025,
author = {Ueki, Hideto and Uemura, Munenori and Chinzei, Kiyoyuki and Takahashi, Kosuke and Wakita, Naoto and Okamura, Yasuyoshi and Suzuki, Kotaro and Bando, Yukari and Hara, Takuto and Terakawa, Tomoaki and Yao, Akihisa and Teishima, Jun and Chiba, Koji and Miyake, Hideaki},
title = {Developing an artificial intelligence model for phase recognition in robot‐assisted radical prostatectomy},
journal = {BJU International},
year = {2025},
doi = {10.1111/bju.16862}
}
EUVIMED