From Acquisition to Prognosis: The Role of AI in Cardiac Magnetic Resonance Imaging Evaluation of Ischemic Cardiomyopathy

Abstract

Acute and chronic ischemic cardiomyopathy (ICM) still represents a leading cause of morbidity and mortality. Cardiac magnetic resonance (CMR) imaging plays a central role in the diagnosis and management of ICM, offering detailed visualization of cardiac structures and function. The evolving role of artificial intelligence (AI) in enhancing CMR exams, from acquisition to prognosis, is rapidly expanding in clinical practice, particularly in CMR of patients with ICM, emphasizing the integration of AI algorithms to optimize imaging workflows in standard protocols. Advanced AI models enable more efficient and faster image acquisition, reducing artifacts and enhancing accuracy, even offering free-breathing sequences. In post-processing, AI allows for the segmentation and quantification of cardiac parameters, facilitating precise assessment of volumes, myocardial scarring, and perfusion abnormalities, which are critical parameters in ICM. Moreover, AI-driven analysis provides robust prognostic insights by predicting adverse outcomes, such as heart failure and arrhythmias, through comprehensive data integration and pattern recognition. Looking forward, the future of AI in CMR promises further advancements in personalized medicine, with AI algorithms continually improving in accuracy and clinical applicability. This review will analyze the role of AI in increasing diagnostic accuracy, optimizing workflows, and improving prognosis in patients with ICM.

Keywords: artificial intelligence; cardiac magnetic resonance; deep learning; ischemic cardiomyopathy; late gadolinium enhancement; machine learning.