Article intro - SAGES recommendation on surgical videos
Abstract
Background – Operating rooms generate a vast amount of data from each procedure daily. Particularly video and imaging provides significant value for surgical research, clinical outcome assessment, quality control, and education. The data lifecycle is influenced by various factors, including data structure and its interplay with acquisition, storage, and sharing; data use through processing and annotation; data exploration for research, education, and clinical purposes; and data governance, which encompasses all policies and regulations associated with the data. As a result, there is a universal need among stakeholders in surgical data science to establish standardized frameworks that address all aspects of the data lifecycle to ensure data quality and purpose.
Methods – Four working groups were formed, from a pool of 48 representatives from academia and industry, including clinicians, computer scientists, engineers and industry representatives. These working groups focused on four themes: (1) Data Use, (2) Data Structure, (3) Data Exploration, and (4) Data Governance. After extensive working group and panel discussions among the experts, a modified Delphi process was conducted to obtain expert recommendations on the four themes.
Results – Following two Delphi rounds, the experts reached a consensus, conceptualizing and structuring each domain to generate concise guidelines for utilizing surgical visual data. We identified the key stakeholders influencing the critical principles associated with the data lifecycle and formulated proposed guidelines. Guidelines for data use should be comprehensive, easily understandable, and applicable to all use cases and stakeholders. Standardization of the data structure should encompass format and quality, data sources, documentation, metadata, and account for biases within the data. To foster scientific data exploration, acquisition, and processing should reflect data diversity and remain adaptable to future applications and uses ofthe data. And finally data governance must be accessible to all stakeholders, addressing legal and ethical considerations surrounding the data. Overall the expert panel agreed that surgical video data is vital for research, clinical applications, and education within the surgical community.
Conclusion – This consensus framework presents essential aspects around the generation of standardized and diverse surgical imaging databanks, accounting for multiple stakeholders involved in data generation and use throughout its lifecycle. Following the SAGES annotation framework, these data management recommendations lay the foundation for standardization for data use, structure, and exploration. In sequence a more detailed exploration of aspects required for adequate data governance will follow. Further considerations addressing the unique characteristics of the video data and the individual aspects of its lifecycle influencing research, education, and clinical quality improvement, remain to be done.
Key Words – AI, Education, Surgical Data Science, Surgical AI, Delphi Consensus
Source: SAGES
Comments