Ultromics has received clearance from the US Food and Drug Administration (FDA) for its artificial intelligence (AI)-based platform, EchoGo Heart Failure, which enables the detection of heart failure with preserved ejection fraction (HFpEF) accurately.
Echocardiography and AI experts from Ultromics, a spin-out of the University of Oxford in the UK, developed the device in partnership with Mayo Clinic.
Using AI, the EchoGo Heart Failure device can precisely detect HFpEF, a heterogeneous syndrome, from a single echocardiogram image.
The clearance from the FDA comes weeks after Ultromics proposed to address this area of significant unmet medical need in collaboration with the Foundation for the National Institutes of Health (FNIH) Accelerating Medicines Partnership Heart Failure (AMP HF) programme.
Managed by FNIH, the $37m multi-stakeholder partnership is a collaboration between the National Institutes of Health (NIH), the National Heart Lung and Blood Institute (NHLBI), and the FDA that will last for five years.
The parties seek to develop precise strategies and targeted therapies for HFpEF through the partnership.
Ultromics CEO and founder Dr Ross Upton said: “We are delighted that the FDA has recognised EchoGo Heart Failure as a breakthrough device and has cleared the technology to provide reliable detection of HFpEF.
“The technology improves the accuracy of HFpEF detection, enabling more patients to receive treatment, which will reduce the significant burden on patients and healthcare systems alike.”
Through better HFpEF detection, EchoGo Heart Failure aims to prevent hospitalisations of patients with HF and reduce mortality.
By analysing several pixels within a single echocardiogram, the platform delivers precise detection of HFpEF.
In the validation dataset, EchoGo Heart Failure demonstrated 90% accuracy in detecting HFpEF.
It also showed 87.8% and 83% sensitivity and specificity, respectively, in the independent testing dataset.
Additionally, 68% more patients with HFpEF were correctly identified by EchoGo Heart Failure than clinical algorithms in the dataset.