Researchers have developed a groundbreaking AI method that improves the accuracy and efficiency of analysing MRI heart scans.
Developed by experts from the Universities of East Anglia (UEA), Sheffield and Leeds, the AI model analyses heart scans in a specific view known as the four-chamber plane.
This innovative method could provide faster, non-invasive and more accurate diagnosis of heart failure and other cardiac conditions, saving the NHS valuable time and resources.
Testing the AI model
The team performed a retrospective observational study using data from 814 patients at Sheffield and Leeds Teaching Hospitals to train an AI model.
To ensure accuracy, the model was tested using scans and data from 101 patients at Norfolk and Norwich University Hospitals.
This AI model stands out because it was trained using data from multiple hospitals and various scanner types and tested on a diverse patient group.
Unlike previous studies that focused on the heart’s two main chambers, this model provides a complete analysis of the entire heart, showing all four chambers.
Advancing the analysis of MRI heart scans
The study revealed that the novel AI method was comparable to the manual analysis traditionally performed by doctors and could lead to better treatment outcomes for patients.
Dr Pankaj Garg, leader of the research from the University of East Anglia’s Norwich Medical School, explained: “The AI model precisely determined the size and function of the heart’s chambers and demonstrated outcomes comparable to those acquired by doctors manually but much quicker.
“Unlike a standard manual MRI analysis, which can take up to 45 minutes or more, the new AI model takes just a few seconds.
“This automated technique could offer speedy and dependable evaluations of heart health, with the potential to enhance patient care.”
The researchers recommend that future studies should test the model with larger patient groups from different hospitals.
These studies should use various types of MRI scanners and include other common diseases seen in medical practice to determine if the model performs well in a broader range of real-world situations.
