At the 27th Annual Cardiovascular Conference of the Chinese Medical Association held recently, China's first large-scale cardiovascular AI-OCT (Optical Coherence Tomography) model, the Qiguang AI-OCT model, was officially unveiled. It integrates AI big models with advanced imaging technology, relies on high-quality cardiovascular databases, significantly improves the analysis efficiency and accuracy of OCT images, greatly reduces the threshold for technology use, and is regarded as an important breakthrough in achieving precise vascular intervention therapy (PCI) surgery in the industry. Endovascular imaging technology, especially OCT, has become an indispensable "optical biopsy" tool in PCI surgery for cardiovascular diseases due to its ultra-high resolution of 10-20 microns, which can clearly present the nature of vascular plaques, stent adhesion and other subtle structures. However, its image interpretation is highly dependent on the operator's experience, and the learning curve is long, which limits the popularity of this technology in primary hospitals. In this field, AI is expected to become a "diagnosis and treatment partner" for doctors, but issues such as the professionalism of AI technology constrain its further promotion and application. The reliability of medical AI lies in the quality and compliance of data. ”Wu Dengsheng, a distinguished professor at Shenzhen University and the general manager of Shenzhen Dongbi Technology Data Co., Ltd. (hereinafter referred to as "Dongbi Technology Data"), said, "In the past, many AI models experienced hallucinations or misjudgments, with the core being chaotic training data or detachment from clinical practice. ”To address this issue, Dongbi Technology has established the first professional cardiovascular corpus in China, the Dongbi Cardiovascular Corpus. The corpus systematically integrates 129 top journals, authoritative guidelines, and clinical data from around the world, and has been rigorously validated to form a high-quality data foundation. In early 2024, Dongbi Technology Data collaborated with Shenzhen Zhongke Micro Light Medical Equipment Technology Co., Ltd. to launch the research and development of AI-OCT large models, deeply integrating high-quality data with advanced imaging equipment. The Qiguang AI-OCT large model has overcome multiple technical difficulties. It not only achieves intelligent recognition of key features such as vulnerable plaques, calcification distribution, and stent adhesion, but also real-time access to similar cases and evidence-based data to automatically generate structured reports. The misdiagnosis rate of this system is lower than the average level of human doctors, and it can achieve results in seconds, "said Zhu Rui, founder of Shenzhen Zhongke Micro Light Medical Equipment Technology Co., Ltd." This greatly reduces the threshold for the use of OCT. ”Clinical trials have shown that the AI-OCT system performs outstandingly in improving diagnostic consistency, shortening interpretation time, and assisting in the development of surgical strategies. Professor Wu Xiaofan from Beijing Anzhen Hospital affiliated with Capital Medical University commented that AI-OCT has achieved a leap from "viewing images" to "developing solutions", truly helping clinical practice enter the era of intelligent diagnosis and treatment. Yuan Zuyi, Vice President and Director of the Department of Cardiology at the First Affiliated Hospital of Xi'an Jiaotong University, said that AI-OCT can help doctors manage coronary heart disease in a more systematic and evidence-based manner, especially providing key insights in complex lesions. The value of AI-OCT lies not only in technological breakthroughs, but also in promoting the universal access of medical resources. We have long realized that the diagnostic capabilities of grassroots hospitals are insufficient, and the medical resources of tertiary hospitals are tight. How to transform the experience of top surgeons into universal tools is the key to medical equality, "said Zhu Rui." Our goal is to transfer the experience and precise diagnosis and treatment capabilities of top hospitals to grassroots through AI. ”(New Society)