A team at Sheffield Teaching Hospitals NHS Foundation Trust has received a double award nomination from HSJ Digital Awards 2023 for developing an AI tool that is able to spot heart damage in seconds.
The technology, which was shown to have a high degree of accuracy when tested in a large trial involving 2,000 patient scans, performs complex mathematical calculations to measure blood flow in and out of the heart without the need for the manual analysis that can leave results from MRI scans pending for days. It is also said to be the first AI tool to assess the heart muscle responsible for pumping blood from the body into the lungs.
The software, named the AI segmentation of cardiac MRI to automate the measurement of cardiac function and volume technology, is a finalist in the Digital Innovator of the Year and Optimising Clinical Pathways through Digital categories.
Consultant cardiothoracic radiologist Dr Andrew Swift worked alongside Dr Samer Alabed, Dr Kavita Karunasaagarar and Dr Pete Metherall to develop and test the software. It was developed with support from MRI radiographers and clinical scientists at the Sheffield 3DLab and in collaboration with Dr Rob van der Geest at Leiden University, the Netherlands.
Dr Swift said: “We are absolutely delighted with this double award nomination, which recognises the hard work of our team in bringing this potentially game-changing software to the NHS. The new software provides a quick and comprehensive analysis of the heart’s function and has the potential to free up clinician time to focus on direct patient care rather than having to do this analysis.
“The technology has been shown to be particularly useful in observing changes in the heart’s function caused by pulmonary hypertension, a rare condition that constricts blood flow in the arteries of the lung. This in turn causes a strain on the heart, eventually leading to heart failure.”
Picture: Dr Kavita Karunasaagarar, Dr Andrew Swift, Dr Samer Alabed and Dr Pete Metherall with the segmented results of an MRI heart scan generated by their AI technology.
Published on page 5 of the June 2023 issue of RAD Magazine.