Appendix300: A multi-institutional laparoscopic appendectomy video dataset for computational modeling tasks

The limited availability of diverse and representative training data poses a critical barrier to the development of clinically relevant computational tools for intraoperative surgical decision support. Surgical procedures are not routinely recorded, and annotation requires domain expertise, resulting in a scarcity of open-access surgical video datasets with high-quality annotations. Existing datasets are typically limited to single institutions and specific procedures, such as cholecystectomy, and rarely comprise patient-level metadata like demographic characteristics, disease history, or laboratory parameters.

The Appendix300 dataset comprises 330 laparoscopic surgery recordings, including 325 full-length laparoscopic appendectomies and 5 control recordings from non-appendectomy procedures in pediatric and adult patients treated at five German centers. The dataset includes patient-level clinical metadata (demographics, medical history, clinical symptoms, laboratory parameters, and histopathological findings), as well as standardized expert annotations of the laparoscopic grade of appendicitis.

Appendix300 currently represents the largest publicly available collection of surgical video data with patient metadata and the first curated dataset of laparoscopic appendectomies. It enables novel validation tasks for computer vision in surgery, including the classification of appendicitis severity and the detection of appendiceal perforation. Technical validation of the laparoscopic appendicitis grade annotations showed substantial interrater agreement (weighted Cohen's κ = 0.615).

The Appendix300 dataset expands the scope of surgical data science by integrating video data with clinical and pathological metadata across institutions. It enables new and clinically relevant patient-level validation tasks for computer vision in laparoscopic surgery and facilitates decentralized learning approaches, overall enhancing the breadth and translational relevance of AI-based surgical video analysis.