Researchers have devised an innovative approach using artificial intelligence to analyze heart MRI scans in a way that could optimize time, resources, and patient care.
Researchers have come up with an advanced technique to analyze heart MRI scans with the assistance of artificial intelligence, potentially saving significant time and resources for healthcare systems like the NHS, while also enhancing patient care.
Collaborating researchers from the Universities of East Anglia (UEA), Sheffield, and Leeds have developed a smart computer model powered by AI to scrutinize heart images from MRI scans, focusing on a specific view called the four-chamber plane.
Dr. Pankaj Garg, the lead researcher from the University of East Anglia’s Norwich Medical School and a consultant cardiologist at the Norfolk and Norwich University Hospital, is spearheading a team that has introduced innovative 4D MRI imaging technology. This breakthrough technology is laying the foundation for quicker, non-invasive, and more precise diagnoses of heart failure and other cardiac conditions.
Dr. Garg explained: “The AI model accurately assessed the size and functionality of the heart’s chambers, producing results comparable to manual assessments by doctors but in a significantly shorter time frame.
“In contrast to a traditional manual MRI analysis that could take 45 minutes or longer, the new AI model processes results within seconds.
“This automated method holds the potential to provide swift and reliable evaluations of heart health, ultimately improving patient care.”
An observational study involving data from 814 patients from Sheffield Teaching Hospitals NHS Foundation Trust and Leeds Teaching Hospitals NHS Trust was used to train the AI model.
To ensure the accuracy of the model’s outcomes, data from an additional 101 patients from the Norfolk and Norwich University Hospitals NHS Foundation Trust were used for testing purposes.
Unlike previous studies focusing on specific hospital data and particular scanner types, this AI model was trained using data from multiple hospitals and various scanners, with testing carried out on a diverse patient group. Moreover, the model offers a comprehensive assessment of the entire heart by visualizing all four chambers, compared to earlier studies concentrating only on the heart’s two main chambers.
PhD candidate Dr. Hosamadin Assadi from UEA’s Norwich Medical School stated: “Automating the evaluation of heart structure and function will save time, resources, and ensure consistent results for physicians.
“This innovation has the potential to lead to more efficient diagnoses, better treatment choices, and ultimately, improved patient outcomes in cases of heart conditions.
“Furthermore, the AI’s capacity to predict mortality based on heart measurements underscores its potential to transform cardiac care and enhance patient prognosis.”
The researchers suggest that future studies should test the model on larger patient groups from different hospitals, using various MRI scanners and including other prevalent medical conditions to assess its performance in broader real-world scenarios.
Recent research from the UEA, Leeds, and Sheffield teams has also refined the use of heart MRI scans in diagnosing female patients, particularly those with early or borderline heart disease, resulting in a 16.5% increase in the number of females accurately diagnosed.
This collaborative research involved the University of East Anglia, the University of Leeds, the University of Sheffield, Leiden University Medical Centre, the Norfolk and Norwich University Hospitals NHS Foundation Trust, Sheffield Teaching Hospitals NHS Foundation Trust, and Leeds Teaching Hospitals NHS Trust.
The study received funding for Dr. Pankaj Garg from the Wellcome Trust’s Clinical Research Career Development Fellowship.
‘Development and validation of AI-derived segmentation of four-chamber cine CMR’ has been published in the European Radiology Experimental journal.
