Topic: Foundation Models and Multimodal Intelligence for Surgery

Recent advances in foundation models are reshaping artificial intelligence in surgery by enabling generalizable, data-efficient, and multimodal solutions across surgical imaging, video, language, sensor data, and robotic systems. These models offer new opportunities to advance surgical perception, workflow understanding, clinical decision support, personalized surgical planning, and robotic assistance across the preoperative, intraoperative, and postoperative continuum. Emerging applications include surgical video understanding, image-guided intervention, surgical data science, digital twins, and autonomous or semi-autonomous surgical systems.

Despite this progress, significant challenges remain in data quality, model generalization, interpretability, reliability, safety, privacy, and clinical validation. Translating foundation models into real-world surgical practice also requires rigorous benchmarking, external validation, human–AI collaboration, and integration into clinical workflows.

This Special Issue aims to bring together methodological innovations, clinical applications, reviews, and perspectives on foundation models and multimodal intelligence for surgery. Topics of interest include, but are not limited to, large-scale pretraining, vision-language and large language models, surgical video understanding, generative AI, surgical robotics and embodied intelligence, multimodal data integration, data-efficient adaptation, personalized surgical planning, digital twins, benchmarking, model safety and trustworthiness, and clinical translation across surgical specialties.