Accelerated MRI sequence helps radiologists assess heart disease patients without breath-holding
An accelerated MRI sequence can help radiologists assess ischemic heart disease without requiring patients to hold their breath, according to new research published Wednesday.
Cardiac magnetic resonance imaging plays a central role in evaluating and diagnosing patients with such heart damage, caused by poor blood flow. However, these examinations can take a long time and require patients to repeatedly perform breath holds to avoid respiratory artifacts, researchers detailed in the American Journal of Roentgenology [1].
“These breath holds may be difficult in patients with [ischemic heart disease] and can lead to suboptimal image quality, resulting in possible measurement errors,” David Monteuuis, MD, with the Department of Radiology at Amiens University Hospital in France, and co-authors wrote Feb. 7.
To address this problem, scientists conducted a prospective study incorporating patients who underwent cardiac MRI for evaluation of ischemic heart disease between March and June of 2023. Examinations included an investigational, free-breathing, short-axis imaging sequence that used deep learning to reconstruct the images. Two radiologists assessed the image quality for both MRI approaches.
The analysis included a total of 26 patients. Acquisition times proved shorter for the deep learning-based method when compared to the standard sequence. And the accelerated approach also showed no significant difference when assessing left-ventricular ejection fraction. Both radiologists said the subjective image quality was better for the deep learning-based images. However, blurring artifacts were more frequent, the authors noted.
“The improved subjective image quality of the cine-[deep learning] sequence likely in part relates to the sequence’s respiratory synchronization, in comparison with the standard sequence’s reliance on adequate breath-holding, which may be challenging in dyspneic patients,” Monteuuis and colleagues wrote.
Read much more, including potential study limitations, in AJR at the link below.