[Artificial intelligence in cardiac imaging: status and future developments].
文献摘要
Introduction Artificial intelligence (AI) has rapidly gained importance in medicine over recent years, particularly in medical imaging. AI is transforming cardiac imaging along the entire workflow - from image acquisition and reconstruction to interpretation and diagnosis. In echocardio-graphy, computed tomography (CT), magnetic resonance imaging (MRI), and nuclear medicine, AI systems have the potential to automatically capture and standardize measurements, reduce noise and artifacts, shorten acquisition and analysis times, and improve reproducibility. Applications range from real-time automated analysis of transthoracic echocardiograms and CT-based calcium scoring and plaque quantification to the detection of rare coronary anomalies, AI-assisted MRI planning and the identification of subtle pathological patterns of cardiac amyloidosis. In the future, multimodal AI models combining imaging, clinical, laboratory, and genetic data will enable highly precise risk stratification and individualized therapies. However, challenges remain in terms of generalizability, prospective validation, explainability, and integration into real-world workflows. Clinical validation, quality control, and physician oversight remain essential cornerstones for the responsible use of AI in cardiac imaging.