Key points are not available for this paper at this time.
Laparoscopic surgery is the method of choice for numerous surgical procedures, while it confronts a lot of challenges. Computer vision exerts a vital role in addressing these challenges and has become a research hotspot, especially in the classification, segmentation, and target detection of abdominal anatomical structures. This study presents a comprehensive review of the last decade of research in this area. At first, a categorized overview of the core subtasks is presented regarding their relevance and applicability to real-world medical scenarios. Second, the dataset used in the experimental validation is statistically analyzed. Subsequently, the technical approaches and trends of classification, segmentation, and target detection tasks are explored in detail, highlighting their advantages, limitations, and practical implications. Additionally, evaluation methods for the three types of tasks are discussed. Finally, gaps in current research are identified. Meanwhile, the great potential for development in this area is emphasized.
Zhou et al. (Mon,) studied this question.