Us2.ai HFpEF evaluation compared to gold standard invasive hemodynamic measurement

New research using Us2.ai in comparison to gold standard invasive hemodynamic measurement showed that in patients with HFpEF, AI measurements of echo parameters are interchangeable with manual core-lab measures to diagnose increased filling pressures.

“Evaluation of left ventricular filling pressures using non-invasive techniques is the holy grail of cardiac imaging in the setting of heart failure,” said Dr Peder Langeland Myhre, MD, PhD. “In this study, the investigators demonstrate that the traditional parameter, E/e’-ratio, was indeed associated with invasively measured filling pressures, but was outperformed by a more novel parameter: left atrial reservoir strain. Moreover, a deep learning-based algorithm, which immediately analyses the images without human input, had similar performance to human analyses. These findings highlight the potential use of AI to democratize echo for even advanced parameters such as LA reservoir strain. This may improve diagnosis and monitoring of heart failure – especially for HFpEF, which is too often missed.”