HomeFDA Approves New AI Software to Help Doctors Take Echocardiograms

FDA Approves New AI Software to Help Doctors Take Echocardiograms

The FDA has approved software that will make it easier for healthcare staff to take echocardiograms.

The technology, developed by San Francisco-based artificial intelligence startup Caption Health, will help more hospitals take ultrasound pictures of the heart, a skill that is usually limited.

Echocardiograms are one of the most widely used tools to help in the diagnosis and treatment of heart disease. Killing around 647,000 people every year, heart disease is the leading cause of death in the United States.

Deputy director of the Office of In Vitro Diagnostics and Radiological Health in the FDA’s Centre for Devices and Radiological health praised the move.

He said: “Today’s marketing authorization enables medical professionals who may not be experts in ultrasonography, such as a registered nurse in a family care clinic or others, to use this tool.

“This is especially important because it demonstrates the potential for artificial intelligence and machine learning technologies to increase access to safe and effective cardiac diagnostics that can be life-saving for patients.”

Boston Children's Hospital and GE Healthcare partner to develop radiology AI

The new software, called Caption Guidance, will help doctors and nurses to carry out the echocardiogram procedure before a trained cardiologist reviews the images.

It is the first software authorized to guide users through ultrasound images of the heart.

Caption Guidance was developed using machine learning, which trained the software to establish a difference between acceptable and unacceptable echocardiogram image quality.

The knowledge, equipped with an interactive AI user face, provides guidance to untrained healthcare professionals on how to maneuver the ultrasound probe for the best results. The software provides real-time feedback, can auto-capture video clips and can select and automatically save the best clip for review.

Caption Guidance is currently used with an FDA-cleared ultrasound system developed by Teratech Corporation. However, it has the potential to be rolled out across more systems.

Studies carried out by the FDA to test the technology proved 50 trained sonographers could obtain similar images both with and without the help of the Caption Guidance software.

Leidos, Mayo Clinic Team Up To Build Tech Accelerator For Innovation

Following this, eight registered nurses, who are not experts in sonography, used the software to obtain images. All images captured were ‘of diagnostic quality’, according to five cardiologists who assessed the work.

The results showed that Caption Guidance would enable registered nurses to successfully acquire echocardiography images and videos.

The software was reviewed through the De Novo premarket pathway, created by the FDA to regulate low to moderate risk devices

Captions CEO Andy Page said in a statement: “No patient should have to forgo a potentially lifesaving cardiac ultrasound. Through the power of artificial intelligence, Caption Guidance will provide patients with unprecedented access to ultrasound when and where they need it most.”

The FDA has said it will soon be sharing more information on the role of artificial intelligence in the healthcare industry.

Along with the authorization, the FDA is working to release special controls for devices of this type which will include requirements related to labeling and performance testing.

[Podcast] How Would An Interoperable Healthcare System in Action Look Like?

The move will create a new regulatory classification so that subsequent similar devices could go through the administration’s 510(k) premarket process. This would allow companies developing new devices to obtain authorization by demonstrating their similarities to technology already on the market.

In a statement, the FDA said it was ‘dedicated to ensuring medical device regulation keeps pace with technological advancements’ and will be hosting a workshop titled “Evolving Role of Artificial Intelligence (AI) in Radiological Imaging”.

It also wants to discuss emerging AI in radiological imaging to focus on best practices for the validation of AI-automated radiological imaging software.