Parameter-efficient framework for surgical action triplet recognition
International Journal of Computer Assisted Radiology and Surgery · 30.04.2024 · 19 (7) · 1291-1299
DOI: 10.1007/s11548-024-03147-6 ↗Kurz gefasst
Fachartikel in der Zeitschrift International Journal of Computer Assisted Radiology and Surgery, online erschienen am 30.04.2024. Verfasst von Yuchong Li und 2 weiteren Autor:innen. Beteiligte Einrichtung: Shenzhen Institute of Advanced Technology, Chinese Academy of Sciences, Shenzhen, China. Hauptthemen laut MEDLINE: Neural Networks, Computer. Die Arbeit wurde bisher 11-mal zitiert (Crossref) und nennt 27 Literaturangaben. Gefördert von: National Natural Science Foundation of China.
Abstract (Originalsprache)
PURPOSE: Surgical action triplet recognition is a clinically significant yet challenging task. It provides surgeons with detailed information about surgical scenarios, thereby facilitating clinical decision-making. However, the high similarity among action triplets presents a formidable obstacle to recognition. To enhance accuracy, prior methods necessitated the utilization of larger models, thereby incurring a considerable computational burden. METHODS: We propose a novel framework known as the Lite and Mega Models (LAM). It comprises a CNN-based fully fine-tuned model (LAM-Lite) and a parameter-efficient fine-tuned model based on the foundation model using Transformer architecture (LAM-Mega). Temporal multi-label data augmentation is introduced for extracting robust class-level features. RESULTS: Our study demonstrates that LAM outperforms prior methods across various parameter scales on the CholecT50 dataset. Using fewer tunable parameters, LAM achieves a mean average precision (mAP) of 42.1%, a 3.6% improvement over the previous state of the art. CONCLUSION: Leveraging effective structural design and robust capabilities of the foundational model, our proposed approach successfully strikes a balance between accuracy and computational efficiency. The source code is accessible at https://github.com/Lycus99/LAM .
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Shenzhen Institute of Advanced Technology, Chinese Academy of Sciences, Shenzhen, China.
- Shenzhen College of Advanced Technology, University of Chinese Academy of Sciences, Shenzhen, China.
- University of Toronto, Toronto, ON, Canada.
- Shenzhen Institute of Advanced Technology, Chinese Academy of Sciences, Shenzhen, China. fc.jia@siat.ac.cn.
- The Key Laboratory of Biomedical Imaging Science and System, Chinese Academy of Sciences, Shenzhen, China. fc.jia@siat.ac.cn.
Förderung
- National Natural Science Foundation of China
Zitieren
APA
Li, Y., Bai, B., & Jia, F. (2024). Parameter-efficient framework for surgical action triplet recognition. International Journal of Computer Assisted Radiology and Surgery, 19(7), 1291-1299. https://doi.org/10.1007/s11548-024-03147-6
Vancouver
Li Y, Bai B, Jia F. Parameter-efficient framework for surgical action triplet recognition. International Journal of Computer Assisted Radiology and Surgery. 2024;19(7):1291-1299. doi:10.1007/s11548-024-03147-6
BibTeX
@article{Li2024,
author = {Li, Yuchong and Bai, Bizhe and Jia, Fucang},
title = {Parameter-efficient framework for surgical action triplet recognition},
journal = {International Journal of Computer Assisted Radiology and Surgery},
year = {2024},
volume = {19},
number = {7},
pages = {1291--1299},
doi = {10.1007/s11548-024-03147-6}
}
EUVIMED