Image recognition method, device and storage medium for laparoscopic surgery

Abstract

The invention discloses an image recognition method, device and storage medium for laparoscopic surgery. The method includes: first acquiring a real-time surgical video stream, and intercepting key frame images in the real-time surgical video stream; The frame image is divided into multiple image blocks; then the temporal and spatial positions of the multiple image blocks are encoded to obtain the time dimension vector and the space dimension vector; finally, the time dimension vector and the space dimension vector are input into the surgical stage recognition model to obtain the surgical stage recognition result. The present invention realizes precise identification of laparoscopic surgery stages, and can cope with image recognition of complex and multi-stage laparoscopic surgery stages, has high efficiency and high accuracy, and can be widely used in the field of smart medical technology.

Description

technical field [0001] Embodiments of the present invention relate to the field of smart medical technology, and more specifically, embodiments of the present invention relate to image recognition methods, devices, and storage media in laparoscopic surgery. Background technique [0002] Surgical workflow identification is an important topic in the field of computer-assisted surgery. Automated surgical workflow identification can explain the specific activities currently being performed and facilitate the standardization of surgical procedures. With the application of deep learning in the medical field, the use of convolutional neural networks has been proposed in recent years to extract more visual information from surgical videos to identify surgical stages. Given that surgical videos are actually sequential data, in order to address these challenges, many studies have focused on first extracting discriminative visual features from video frames, and then modeling the tempor...

AI Technical Summary

Problems solved by technology

However, the automatic surgical stage method based on the convolutional neural network is limited by the inherent translation invariance of the convolution operation itself and can only perceive local information. In the field of video understanding, it cannot meet the requirements of the system to accurately distinguish the surgical stage.

Moreover, the existing surgical stage recognition method is aimed at laparoscopic cholecystectomy with fewer simple surgical stages. For complex, multi-stage laparoscopic surgery, its recognition efficiency and accuracy are greatly reduced.

Images