AI发现特定心脏病准确率已超过人类专家？心电图迎来技术飞跃

Core Insights - The article discusses the potential of artificial intelligence (AI) in early detection of structural heart disease, highlighting a new AI tool named EchoNext that analyzes electrocardiogram (ECG) data to identify patients at risk [1][3][4] Group 1: AI and Structural Heart Disease Detection - Early detection of structural heart disease is crucial for improving patient outcomes, but traditional screening methods like echocardiography are limited by cost and accessibility [1][3] - EchoNext, the AI screening tool, was trained on over 1.2 million ECG-echocardiogram pairs from 230,000 patients, demonstrating a detection accuracy of 77.3% for structural heart disease [3][4] - The AI tool identified over 7,500 high-risk patients from a cohort of 85,000 who had not previously undergone echocardiography, with more than half of these patients subsequently receiving the necessary echocardiogram [3][4] Group 2: Comparison with Human Experts - In a study comparing EchoNext with human cardiologists, the accuracy of cardiologists without AI assistance was 64%, which improved to 69.2% with AI support, but still fell short of EchoNext's performance [4] - The findings suggest that AI can significantly enhance the screening process for structural heart disease, potentially leading to better patient management and resource allocation [4] Group 3: Future Directions and Challenges - The research team envisions integrating AI with other diagnostic modalities, such as chest X-rays and laboratory results, for a more comprehensive risk assessment of heart disease [5] - Challenges remain regarding the integration and adoption of AI models in clinical settings, including issues related to generalization and the complexity of implementation [5]