Is FDA Getting More Cozy with AI?
The use of artificial intelligence (AI) is undoubtedly on the rise in healthcare. From fledging medtech companies to established giants like Watson IBM—many firms are riding the current AI wave. Not only are companies embracing technology, but so is FDA.
Recently the agency gave approval to San Francisco-based Viz.ai’s Contact application, a type of clinical decision support software designed to analyze CT results that could notify providers of a potential stroke in their patients. “What it’s doing is comparing the imaging features in that patient with the millions of images it has been trained on in the past, in order to identify a specific disease, which is a large vessel occlusion,” Chris Mansi MD, president and CEO of Viz.ai, told MD+DI.
Once the Contact Application analyzes the CT images it sends a text notification to a neurovascular specialist if a suspected large vessel blockage has been identified. The algorithm then automatically notifies the specialist during the same time the first-line provider is conducting a standard review of the images, potentially involving the specialist sooner than the usual standard of care in which patients wait for a radiologist to review the images.
According to the Centers for Disease Control and Prevention, stroke is the fifth leading cause of death in the U.S. and is a major cause of serious disability for adults. About 795,000 people in the U.S. have a stroke each year.
“Strokes can cause serious and irreversible damage to patients. The software device could benefit patients by notifying a specialist earlier thereby decreasing the time to treatment. Faster treatment may lessen the extent or progression of a stroke,” Robert Ochs, Ph.D., acting deputy director for radiological health, Office of In Vitro Diagnostics and Radiological Health in FDA’s Center for Devices and Radiological Health, said in a release.
To gain approval Viz.ai submitted a retrospective study of 300 CT images that assessed the independent performance of the image analysis algorithm and notification functionality of the Viz.ai Contact application against the performance of two trained neuro-radiologists for the detection of large vessel blockages in the brain. Real-world evidence was used with a clinical study to demonstrate that the application could notify a neurovascular specialist sooner in cases where a blockage was suspected.
The application received approval through the De Novo premarket review pathway, a regulatory pathway for some new types of medical devices that are low to moderate risk and have no legally marketed predicate device to base a determination of substantial equivalence.
“The beautiful thing from deep learning is that it learns from every example,” Mansi said. “Absolutely it can get better in time. But it’s important to realize that from an FDA perspective if you do make improvements you have to resubmit to another performance test.”
Mansi’s company isn’t the only firm that has approval for an AI-based technology that could serve as a potential diagnostic. In July of 2017, FDA approved Cardiolog Technologies’ ECG analysis platform. The technology is a cloud-based cardiac monitoring- analysis web service designed to help physicians screen for atrial fibrillation and other arrhythmias using long-term ambulatory ECG monitoring recordings.
FDA is currently creating a regulatory framework for these products that encourages developers to create, adapt and expand the functionalities of their software to aid providers in diagnosing and treating diseases and conditions.
The agency recognized the existing commercialization paradigm for these technologies has become too burdensome to continue to innovate at such a rapid pace and has established the Digital Health Innovation Action Plan. With this plan FDA can address concerns and create a new regulatory pathway for these emerging technologies.
Mansi said if companies are to thrive in the AI space, then they must not try to supplant the role of the physician.
“Doctors are the people that make the diagnosis to the treatment,” Mansi said. “Companies that try and replace doctors are going to fail. We are a very long way from doing that.”
He added that AI will eventually become common place in healthcare.
“I use AI in my everyday life, from my software to my email,” Mansi said. “I just don’t realize it’s happening. We want to bring that to healthcare to increase efficiency and patient outcomes. That’s what’s about to happen. We’ll be the first company to do that, but I don’t think we’ll be the only company.”