AI turns low-quality MRI scans into helpful heart images
A new advanced artificial intelligence (AI) model is capable of restoring low-quality cardiac MRI scans so that they can still deliver considerable value to cardiologists and other clinicians.
A team of biomedical engineers, radiologists and cardiologists developed the AI model, TagGen, and shared its early findings in Magnetic Resonance in Medicine.[1]
“If you have a blurry image, you have very few ways to recover the fine details or quality of the image,” lead author Changyu Sun, PhD, an MRI researcher with the University of Missouri School of Medicine, said in a statement. “The sharpness reveals very important information for the clinical diagnosis, like if there’s abnormal movement or any dysfunction.”
TagGen was trained, validated and tested using more than 2,000 images captured at a single hospital from August 2022 to June 2024. Sun et al. found that it could restore low-resolution heart images and deliver much better visualization, giving cardiologists the detail needed to make important diagnoses. The key to TagGen’s success is its ability produce sharper taglines, which help capture muscle movements.
According to the group, one of the major benefits of this AI model is its potential to speed up MRI scans. This improves patient satisfaction while helping radiology practices scan more patients in a given day.
“During a heart MRI scan, patients are asked to hold their breath to reduce chest movement from breathing, which helps create clearer images.” Sun said. “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.”
Looking ahead, the team behind this work plans on fine-tuning TagGen as necessary and bringing the technique to other modalities.
Click here to read the full analysis in Magnetic Resonance in Medicine.
Michael has more than 18 years of experience as a professional writer and editor. He has written at length about cardiology, radiology, artificial intelligence and other key healthcare topics.