The paper "Data-Centric Foundation Models in Computational Healthcare: A Survey" by YUNKUN ZHANG et al. explores the application of foundation models (FMs) in computational healthcare, emphasizing the data-centric approach. FMs, which have gained prominence in visual recognition, language understanding, and knowledge discovery, are now being leveraged in healthcare to improve clinical workflows and patient outcomes. The authors discuss the challenges in obtaining and processing high-quality clinical data, such as data quantity, annotation, patient privacy, and ethical considerations. They review various data-centric approaches, including large-scale model pre-training, fine-tuning, and in-context learning, and their applications in healthcare. The paper also highlights the importance of multi-modal FMs for data fusion and the use of FMs to address data quantity and annotation challenges. Additionally, it discusses the role of FMs in enhancing data efficiency and the curation of large-scale healthcare datasets from the Internet. The authors conclude by offering a promising outlook on FM-based analytics to improve patient outcomes and streamline clinical workflows, while emphasizing the need for responsible AI alignment with human values.The paper "Data-Centric Foundation Models in Computational Healthcare: A Survey" by YUNKUN ZHANG et al. explores the application of foundation models (FMs) in computational healthcare, emphasizing the data-centric approach. FMs, which have gained prominence in visual recognition, language understanding, and knowledge discovery, are now being leveraged in healthcare to improve clinical workflows and patient outcomes. The authors discuss the challenges in obtaining and processing high-quality clinical data, such as data quantity, annotation, patient privacy, and ethical considerations. They review various data-centric approaches, including large-scale model pre-training, fine-tuning, and in-context learning, and their applications in healthcare. The paper also highlights the importance of multi-modal FMs for data fusion and the use of FMs to address data quantity and annotation challenges. Additionally, it discusses the role of FMs in enhancing data efficiency and the curation of large-scale healthcare datasets from the Internet. The authors conclude by offering a promising outlook on FM-based analytics to improve patient outcomes and streamline clinical workflows, while emphasizing the need for responsible AI alignment with human values.