Efficient Segmentation of Intraoperative Anatomical Landmarks in Laparoscopic Cholecystectomy Based on Deep Learning

Background Laparoscopic cholecystectomy (LC) is the gold standard procedure for the treatment of benign gallbladder diseases, but there is the risk of intraoperative bile duct injury, which can lead to surgical failure and cause significant social and economic burden. When surgeons rely on visual inspection to identify tissue structures during LC, subjective factors such as experience, psychological factors, and fatigue can compromise the intraoperative recognition of anatomic landmarks. The positioning of anatomical landmarks by the surgeon in the pre-dissection phase of LC is relatively vague and requires step-by-step exploration as the surgery progresses, becoming clearer in the post-dissection phase.Method To alleviate the pressure on surgeons during procedures, this study aimed to achieve real-time intraoperative navigation during LC by dynamically identifying and annotating key anatomical landmarks, including the gallbladder, Calot's triangle, and common bile duct. The study proposed a novel semantic segmentation neural network called the Channel Attention Pyramid Scene Parsing Plus Network (CPPN). The network utilized pooling layers with different scales and assigned non-equal weights to extract feature information. Additionally, a spatial channel attention module was added to accurately capture important features or contextual information, improving the model's performance and effectiveness. Training was conducted using video frames from the pre-dissection phase, while testing used video frames from the post-dissection phase.Results All models were subjected to a 10-fold cross-validation on 1425 selected frames from 132 LC videos, with training and validation conducted in two separate LC stages. The proposed model CPPN achieved a mean intersection over union (mIoU) of 0.855 (±0.03), outperforming other segmentation neural networks. The model demonstrated optimal performance across most metrics, with an intersection over union (IoU) of 0.881 (±0.01) for the gallbladder, 0.769 (±0.03) for Calot's triangle, and 0.813 (±0.02) for the common bile duct.Conclusion The intelligent segmentation algorithm proposed in this study achieved the highest mIoU, outperforming other models, and may assist surgeons in the real-time assessment of critical anatomical landmarks of Calot's triangle. This could help prevent common bile duct injury by allowing for more intuitive dissection of Calot's triangle, aiding in the visual inspection of laparoscopic cholecystectomy procedures.