Part of the Endoscopic Vision Challenge¶

Surgical tool detection is attracting increasing attention from the medical image processing community [1]. The goal generally is not to precisely locate tools in images, but rather to indicate which tools are being used by the surgeon at each instant. The main motivation for annotating tool usage is to design efficient solutions for surgical workflow analysis, with potential applications in report generation, surgical training and real-time decision support.

In 2017, we organized a challenge on surgical tool detection in cataract surgery videos. With 14 participating teams, that first edition of CATARACTS was a success. Therefore, we decided to repeat this experience in 2018 with new data and new technical challenges. In particular, the 2018 edition provides two synchronized video streams per surgery: one showing the patient's eye (like in the 2017 edition), the other one showing the surgical tray. By knowing which tools exit or enter the surgical tray, we know which tools are likely being used by the surgeon and which tools surely are not. This second edition of CATARACTS is organized as a sub-challenge of the MICCAI 2018 EndoVis challenge. It should be noted that CATARACTS does not rely on endoscopic videos, but rather on microscopic videos. However, these two modalities share many similarities that justify a joint event.

[1] Bouget D, Allan M, Stoyanov D, Jannin P. Vision-based and marker-less surgical tool detection and tracking: a review of the literature. Med Image Anal. 2017 Jan;35:633-654.