HEADLINER

ACS perspective  

Artificial intelligence (AI), specifically machine learning and computer vision, largely depends on vast amounts of annotated data for model training and validation. Data challenges are a well-established methodology in the computer science community for creating large-scale and diverse datasets reflective of a wide range of target features. In surgery, the imaging and video data produced during minimally-invasive procedures presents a unique opportunity to explore AI-based assessment of clinically relevant visual features. The Society of American Gastrointestinal Surgeons (SAGES) is conducting the first surgical data challenge initiated in collaboration between surgeons and computer scientists. The SAGES Critical View of Safety Challenge is creating a worldwide dataset and is fostering a large-scale competition around the development of AI models to detect essential surgical safety measures, such as the critical view of safety in laparoscopic cholecystectomy. This talk will elaborate on the key concept of data challenges, the SAGES CVS Challenge in particular, and the significance of surgical artificial intelligence.