US regulators have approved AI-based software that sends a text message to healthcare professionals if their patients are judged at risk of suffering a stroke.
The Viz.AI Contact application analyses CT (computed tomography) results and texts neurovascular specialists if a suspected large vessel blockage is identified.
The computer-aided triage software uses an artificial intelligence algorithm to analyse images for indicators associated with a stroke, running at the same time as a patient is seen by a first-line care provider for a standard review of the images.
The notification can be sent to a mobile device, such as a smart phone or tablet, but the specialist still needs to review the images on a clinical workstation and the system is not intended to replace a full patient evaluation.
Robert Ochs, is acting deputy director for radiological health at the Office of In Vitro Diagnostics and Radiological Health within the FDA’s Center for Devices and Radiological Health.
He said: “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.”
The FDA granted marketing authorisation for the Viz.AI Contact system under its De Novo premarket review pathway. The work-in-progress regulatory system has so far seen the FDA approve its first substance use mobile medical application and a wireless contact lens sensor.
The regulatory framework aims to encourage developers take into account the needs of healthcare providers when it comes to diagnosing and treating diseases and conditions.
When it came to Viz.AI Contact, FDA approval was based a retrospective study of how the image analysis algorithm and notification functionality for 300 CT images. The test saw the Viz.AI Contact application’s ability to detect large vessel blockages in the brain compared with the performance of two trained neuro-radiologists. Real-world evidence was then used to demonstrate the application could notify a neurovascular specialist sooner in cases where a blockage was suspected.
