AI-Assisted Model Enhances MRI Heart Scans

A cardiac MRI can reveal critical information about the heart’s function and any abnormalities, but traditional scans take 30 to 90 minutes and often suffer from poor image quality due to patient movement. Clear taglines, which help track muscle movement, are crucial for identifying areas of the heart that may not be functioning properly. However, blurry images make it hard to recover fine details or accurately assess cardiac function. Researchers have now developed an artificial intelligence (AI)-assisted model that significantly shortens scan time and enhances image sharpness, allowing better visualization of the heart’s motion.

Developed by researchers from the University of Missouri (Columbia, MO, USA), the AI-assisted model—TagGen—restores the quality of blurry cardiac MRI scans and shortens scan duration by approximately 90%. TagGen enhances the sharpness of taglines within the images, which are essential for tracking muscle activity in the heart. By processing low-quality scans, the AI recovers missing visual details and provides clearer insight into how the heart beats, contracts, and pumps. This allows doctors to observe abnormal movement more effectively. The model also reduces the need for patients to hold their breath for extended periods, lowering discomfort and improving overall scan quality.

The improved scanning process enabled by TagGen helps clinicians capture clearer images in just three heartbeats, compared to over 20 heartbeats in standard scans. This not only enhances patient comfort and reduces costs but also improves the accuracy of diagnosis. Future plans for TagGen include refining the model’s motion-tracking capabilities and expanding its application to other types of cardiac MRI, CT scans, and even brain MRIs.

"During a heart MRI scan, patients are asked to hold their breath to reduce chest movement from breathing, which helps create clearer images,” said Changyu Sun, lead researcher, University of Missouri. “Some scans take more than 20 heartbeats, making it harder for patients to hold their breath. By using TagGen to maintain the taglines, doctors can see information they would have otherwise missed, and patients only need to hold their breath for three heartbeats. This technology will lead to better diagnoses and improved patient outcomes.”

Related Links:
University of Missouri

Visit expo >

Multi-Use Ultrasound Table

Clinton

Ultrasound-Guided Biopsy & Visualization Tools

Endoscopic Ultrasound (EUS) Guided Devices

Mammo DR Retrofit Solution

DR Retrofit Mammography

Ultrasound Table

Women’s Ultrasound EA Table