Artificial Intelligence Speeds Up Examination of Heart Images from Magnetic Resonance Imaging

Researchers have developed a ground-breaking approach to analyzing heart magnetic resonance imaging (MRI) scans using artificial intelligence (AI), potentially saving significant time and resources while enhancing patient care.

A collaborative effort involving researchers from the University of East Anglia (UEA, Norfolk, UK) has led to the creation of a sophisticated computer model that applies AI to assess heart images obtained from MRI scans in a specific view known as the four-chamber plane. This initiative builds on the team’s earlier development of cutting-edge 4D MRI imaging technology, which offers quicker, non-invasive, and more precise diagnostics for heart failure and other cardiac conditions. In a retrospective observational study, the team analyzed data from 814 patients to train the AI model. To validate the accuracy of the model, it was further tested using scans and data from an additional 101 patients. Unlike other studies, this AI model was trained with data sourced from multiple hospitals and various types of scanners, and it was tested on a diverse patient group from a different hospital.

Furthermore, this AI model analyzes the entire heart from a view that displays all four chambers, whereas most prior studies concentrated on just the two main chambers. The AI effectively assessed the size and function of the heart’s chambers, achieving results comparable to those performed manually by doctors but in a fraction of the time. Traditional manual analysis of an MRI can take up to 45 minutes, whereas the AI model completes the process in just a few seconds. This automated method could provide quick and reliable assessments of cardiac health, potentially improving patient care significantly. The researchers recommend that future studies should extend testing of the model to larger patient groups across different hospitals, with a variety of MRI scanners and including common conditions encountered in clinical settings, to evaluate its effectiveness in more diverse real-world scenarios.

“Automating the process of assessing heart function and structure will save time and resources and ensure consistent results for doctors,” said PhD student Dr. Hosamadin Assadi, of UEA’s Norwich Medical School. “This innovation could lead to more efficient diagnoses, better treatment decisions, and ultimately, improved outcomes for patients with heart conditions. Moreover, the potential of AI to predict mortality based on heart measurements highlights its potential to revolutionize cardiac care and improve patient prognosis.”

Related Links:
University of East Anglia