This survey paper explores the application of data-centric foundation models within the field of computational healthcare. It highlights the challenges in acquiring and processing high-quality clinical data, such as quantity, annotation, and privacy concerns. The paper reviews various data-centric strategies for foundation models, from pre-training to inference, aiming to improve healthcare workflows and patient outcomes. It also touches upon AI security, assessment, and alignment with human values, while providing a list of relevant healthcare foundation models and datasets. AI
Summary written by gemini-2.5-flash-lite from 1 source. How we write summaries →
IMPACT Provides a comprehensive overview of data-centric approaches for foundation models in healthcare, potentially guiding future research and development in clinical AI applications.
RANK_REASON This is a survey paper on a specific application area of AI.