Article intro - SDS trends and advances
Abstract
Surgical Data Science (SDS) represents a rapidly advancing domain at the intersection of clinical surgery and data science, aimed at enhancing surgical practices, planning, execution and patient care. This chapter delves into the core aspects of SDS, encompassing phase recognition, image segmentation, Surgical Process Modeling (SPM), and surgical skill assessment, presenting a systematic exploration of the key components: datasets, data acquisition techniques, surgical procedures, annotation tools, and processing techniques. Various specialized datasets are introduced, which are pivotal in training and evaluating computational models in SDS. Data acquisition methods, ranging from endoscopic video capture to multimodal sensory data collection, highlighting their crucial role in gathering diverse surgical data are also presented. The chapter also examines a range of surgical procedures, emphasizing their importance in understanding and improving surgical outcomes. Annotation tools are explored, both for temporal and spatial data, highlighting their significance in labeling and categorizing surgical data for Machine Learning applications. Advanced processing techniques in SDS are addressed, including Deep Learning methods, underscoring their impact on extracting meaningful insights from complex clinical data. By exploring the challenges and future directions in SDS, the chapter aims to provide a comprehensive understanding of the way innovative data-driven approaches are transforming the field of surgery.
Source: Springer
Comments