Viz.ai’s stroke-detecting CDS app receives FDA nod

By Dave Muoio
Share

The FDA announced yesterday that it will permit marketing of Viz.ai’s Contact, a clinical decision support (CDS) tool that analyzes CT results and highlights cases that may have experienced a stroke.

The CDS application was submitted through the agency’s de novo premarket review pathway, and as such creates a new regulatory classification for similar AI-drived platforms to come. The FDA’s decision comes just a couple of months after the agency released its long-awaited, and somewhat controversial, draft guidance on CDS outlining which tools would and would not be regulated by the agency.

"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, acting deputy director for radiological health within the FDA’s Center for Devices and Radiological Health, said in a statement.

Viz.ai’s software uses an AI algorithm to scan CT images for indicators associated with stroke, and then sends a text notification to a neurovascular specialist if it identifies a potential large vessel blockage. Because the tool alerts the specialist during the time a first-line provider is reviewing the images, patients may receive attention from a specialist earlier than they would normally. The agency stressed in its announcement that the application should not be a stand in for a full patient evaluation when diagnosing a case.

The premarket approval was supported by a retrospective study of 300 CT images submitted by Viz.ai, in which the algorithm’s analysis performance was tested against those of two trained neuoro-radiologists. The agency said it also received additional data that demonstrated how the CDS tool could notify specialists of a suspected blockage quicker than standard protocol.

Viz.ai also announced the receipt of a CE mark for its technology in January.

Regulation of CDS tools have been something of a focus for the FDA. In accordance with the 21st Century Cures Act, the agency released draft guidance in December that outlined regulated devices based on the degree of human involvement, as opposed to doing so based on risk.

Reactions from industry experts and analysts were mixed at the time, but comments submitted to the FDA last week by the CDS Coalition — a group that includes representatives from clinical societies, payers, providers, software developers, and other parties — has since framed the draft as a misstep. The letter urged the FDA to exempt low-risk CDS software from oversight, called for better alignment with the goals of the Cures Act, and stressed that the current draft would remove many existing CDS software companies from the market.

On a particularly relevant note, the letter also expressed concerns over the handling of machine learning CDS technology, as even low-risk products would be subject to FDA regulation under the current draft’s language.

“There is non-low risk CDS software that should be permitted to use machine learning without FDA oversight,” Bradley Merrill Thompson, an attorney with Epstein Becker Green, wrote on behalf of the CDS Coalition. “We believe that such software can meet the exemption under the Cures Act (despite the language in the Draft Guidance), as long as the relevant portions of our Coalition Industry Guidelines are followed.”

Whether or not Viz.ai’s recently greenlit software would not be considered low risk, yesterday’s decision and its resulting regulatory pathway establish a clear precedent for similar products hoping to reach the market.