AI-Enhanced MRI Improves Image Quality in Arrhythmia Patients

Arrhythmia, an irregular heartbeat, can degrade cardiac magnetic resonance imaging by inducing motion and mistriggering. Accurate, repeatable assessment of left ventricular function is essential for treatment selection, response monitoring, and risk stratification, yet many patients struggle with the repeated breath holds required for standard cine imaging. Failed or low-quality exams can delay care and complicate clinical decision-making. Researchers have now evaluated an artificial intelligence-enhanced MRI approach designed to improve image quality and exam success in this population.

The technology, called deep learning–enhanced Compressed SENSE (AI-CS) single-shot cine MRI, was assessed by investigators at Zhongshan Hospital of Fudan University in Shanghai. Single-shot cine captures an entire cardiac cycle in two heartbeats, which shortens breath-hold duration and reduces the impact of rhythm irregularity on image formation. The work was published in Radiology: Cardiothoracic Imaging on March 26, 2026, and represents the first application of this technique to patients with arrhythmia.

The study enrolled 25 healthy volunteers and 45 patients with suspected arrhythmias. Each participant underwent both conventional balanced steady-state free precession cine MRI and AI-CS single-shot cine sequences. Left ventricular volumetric and strain parameters included end-diastolic volume, end-systolic volume, stroke volume, ejection fraction, radial, longitudinal, and circumferential peak strain, and the standard deviation of peak strain. Three cardiovascular radiologists, blinded to clinical data, independently graded artifacts and structure visibility.

AI-CS single-shot cine produced significantly better image quality than conventional cine, especially in participants with arrhythmia. It yielded fewer mistrigger events and fewer motion artifacts. Quantitative agreement with conventional cine was good to excellent for biventricular volumes and left ventricular mass, supporting interchangeability for core measurements. When conventional cine failed, AI-CS provided ejection fraction values comparable to echocardiography. The AI-CS sequence achieved a 100% success rate versus 88% for conventional cine and demonstrated a shorter mean acquisition time.

These findings indicate AI-CS can serve as a practical alternative in clinical settings where long acquisitions and breath-holding limit cardiac MRI reliability. The investigators noted that refinements in image contrast and artifact suppression could further ease routine adoption. Consistent performance across challenging rhythms may help standardize ventricular assessment in patients who are otherwise difficult to image.

“The AI-CS sequence effectively avoided the cardiac motion artifacts commonly caused by mistriggering in conventional cine,” said Nan Zhang, M.S., supervisor radiologic technologist, Department of Radiology, Zhongshan Hospital of Fudan University in Shanghai. “It also demonstrated a shorter mean acquisition time while providing improved image quality, particularly in the visualization of the endocardial border, epicardial border, papillary muscles and cardiac motion” 

“The AI-CS framework offers a promising alternative for cardiac MRI examinations in the clinical setting, where long acquisition time remains a major challenge,” Zhang said. “Further optimization of the AI-CS framework, particularly regarding image contrast and artifact reduction, will enhance its applicability in routine clinical practice.”