Core Insights - The article discusses the transformative impact of Foundation Models in the healthcare sector, particularly in ophthalmic surgery, highlighting the development of an ophthalmic video foundation model (OVFM) for surgical recognition and navigation [3][6]. Group 1: Research Development - A research team from Shanghai Jiao Tong University and other institutions developed OVFM, which is specifically designed for micro-ophthalmic surgery recognition and navigation, validated through wet-lab porcine eye experiments [3]. - The team constructed a large-scale dataset of 11,426 micro-surgery videos, covering 144 types of anterior and posterior segment surgeries, resulting in approximately 1.1 million surgical video segments [6]. Group 2: Model Performance - The OVFM model outperformed existing video foundation models across seven downstream tasks, including surgical step recognition and complication detection [6]. - A dual-stage knowledge distillation framework was designed to compress the model size by 15.8 times while maintaining about 95% of the original recognition accuracy, enabling real-time deployment in resource-constrained surgical environments [6]. Group 3: Surgical Navigation System - Based on the lightweight model, a smart surgical navigation system was developed, capable of automatically identifying current surgical steps and projecting personalized navigation information without human intervention [7]. - Clinical trials involving ten ophthalmologists showed that the system significantly improved key surgical metrics, with novice surgeons demonstrating greater performance enhancements compared to expert surgeons when assisted by the system [7]. Group 4: Future Implications - The research illustrates the potential of the ophthalmic video foundation model in scene understanding, real-time response, and enhancement of ophthalmic surgical skills, paving the way for the next generation of high-performance, intelligent micro-surgical navigation and robotic systems [9].