Surgical Action Recognition and Prediction with Transformers

Le résumé fourni par la source

Surgical action recognition and prediction play a critical role in clinical decision support. The start-of-the-art surgical action analysis methods mainly focus on one of these two tasks. In this paper, an online surgical action recognition and prediction model to aid contextual decision-making in the operating room is proposed, termed SA. By learning the relationship between endoscopic video frames, the relationship between surgical actions, and the relationship between images and actions, the two tasks of recognition and prediction are realized. Experiments show that the proposed SA is effective in online surgical action recognition and prediction.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.