Clinical Application Study on Real-time Identification of Bleeding Points in Artificial Intelligence-assisted Laparoscopic Cholecystectomy
International Journal of Medical Innovation · 10.03.2026 · 1 (1) · 1-9
DOI: 10.66128/ijmi202601.1 ↗Kurz gefasst
Fachartikel in der Zeitschrift International Journal of Medical Innovation, online erschienen am 16.03.2026. Verfasst von Yangyang Sun und 3 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Objective
Advances in computer vision, a branch of artificial intelligence (AI), have enabled automated analysis of operative video. These advances are likely to benefit clinicians, healthcare systems, and patients alike, yet such benefits are yet to be realised [1]. This study aimed to explore the feasibility and accuracy of a deep learning-based AI system for real-time identification of bleeding points during laparoscopic cholecystectomy (LC), so as to provide a simplified and efficient technical support for intraoperative safety protection in general surgical minimally invasive procedures and offer technical assistance for junior physicians in handling intraoperative bleeding.
Methods
Clinical data and intraoperative laparoscopic video data of patients who underwent LC at Jinzhong First People's Hospital from January 2023 to December 2023 were collected retrospectively. Images were extracted from the videos to construct a dataset, which was divided into a training set and a test set at a ratio of 7:3. This study has been approved by the Medical Ethics Committee of Jinzhong First People's Hospital (Ethics Approval No.:2025-057-01), and informed consent from patients was waived due to the retrospective nature of the study. The YOLO series models reign supreme in real-time object detection due to their superior accuracy and computational efficiency. However, both the convolutional architectures of YOLOv11 and earlier versions and the area-based self-attention mechanism introduced in YOLOv12 are limited to local information aggregation and pairwise correlation modeling, lacking the capability to capture global multi-to-multi high-order correlations, which limits detection performance in complex scenarios. In this paper, we propose YOLOv13, an accurate and lightweight object detector [2]. A bleeding point identification model was constructed and validated using YOLOv13. The results of bleeding points determined by "intraoperative marking by surgeons + blind evaluation by 3 senior general surgery physicians after surgery" were used as the gold standard. Precision, recall, and average precision (AP) were adopted to evaluate the accuracy of detection and identification.
Results
A total of 1,233 instances were segmented, including 863 frames of images in the training set and 370 frames of images in the validation set. Training was conducted using the YOLOv13v-seg model. In the loss curve of the training set, the loss value decreased gradually and smoothly with the increase in the number of iterative calculation epochs. In the training set, the AP value for bleeding point detection and identification reached more than 0.90. In the validation set, the AP for bleeding point detection was 0.82.
Conclusion
The AI system based on YOLOv13 can quickly and accurately identify intraoperative bleeding points during LC. The experimental process is simple and data are easy to obtain. It can serve as a basic research direction for AI-assisted technologies in general surgical minimally invasive procedures and provide a reference for the promotion of intraoperative AI-assisted tools in primary hospitals.
Zitieren
APA
Sun, Y., Wang, B., Gao, Z., & Tian, X. (2026). Clinical Application Study on Real-time Identification of Bleeding Points in Artificial Intelligence-assisted Laparoscopic Cholecystectomy. International Journal of Medical Innovation, 1(1), 1-9. https://doi.org/10.66128/ijmi202601.1
Vancouver
Sun Y, Wang B, Gao Z, Tian X. Clinical Application Study on Real-time Identification of Bleeding Points in Artificial Intelligence-assisted Laparoscopic Cholecystectomy. International Journal of Medical Innovation. 2026;1(1):1-9. doi:10.66128/ijmi202601.1
BibTeX
@article{Sun2026,
author = {Sun, Yangyang and Wang, Baocai and Gao, Zhigang and Tian, Xuhao},
title = {Clinical Application Study on Real-time Identification of Bleeding Points in Artificial Intelligence-assisted Laparoscopic Cholecystectomy},
journal = {International Journal of Medical Innovation},
year = {2026},
volume = {1},
number = {1},
pages = {1--9},
doi = {10.66128/ijmi202601.1}
}
EUVIMED