HiSWA-RLLS2024

HiSWA-RLLS挑战赛旨在通过分层设计理解手术工作流程。该挑战赛是MICCAI 2024 EndoVis挑战赛的一部分，使用公开可用的HiSWA-RLLS数据集的非公开测试拆分进行评估。数据集由临床程序中的手术图像组成，并由专家外科医生进行注释。通过识别手术过程中的步骤、任务和活动，自主机器人手术已经取得了显著进展。然而，来自领域知识的关系线索仍需进一步研究。该数据集中的所有帧都由三位具有5年以上临床经验的外科医生进行了分层手术注释，以确保时间和内容的准确性。

The HiSWA-RLLS Challenge aims to comprehend surgical workflows through hierarchical analysis. As a part of the MICCAI 2024 EndoVis Challenge, this challenge conducts evaluations using the private test split of the publicly available HiSWA-RLLS dataset. The dataset comprises surgical images captured during clinical surgical procedures, which are annotated by expert surgeons. Autonomous robotic surgery has made significant progress by recognizing steps, tasks, and activities during surgical procedures. However, relational cues derived from domain knowledge still require further research. All frames in this dataset have been hierarchically annotated for surgical procedures by three surgeons with over five years of clinical experience, to ensure the accuracy of both temporal and contextual information.