Surgical Data Science

Surgical data science applies machine learning to analyze surgical videos and other data to improve surgical techniques, training, and patient outcomes. Current research focuses on developing robust computer vision models, often employing transformer and convolutional neural networks, to perform tasks like surgical gesture recognition, tool detection, and anatomical structure segmentation, often addressing challenges posed by noisy or limited data through techniques like self-supervised learning and data augmentation. These advancements aim to enhance surgical safety, efficiency, and the objectivity of surgical skills assessment, ultimately leading to better surgical training and improved patient care.