FDA clears software using AI to evaluate CT scans for stroke

Artificial Intelligence healthcare developer Viz.ai said this week it won approval from the FDA for its LVO Stroke platform designed to analyze computed tomography results and notify providers of potential strokes, touting it as the first such system to win FDA clearance. The company’s LVO Stroke platform is designed using deep learning and connects to a hospital CT scanner to alert stroke specialists that a suspected LVO stroke has been identified, with the ability to send radiological images directly to the physician’s phones. “Timely LVO stroke identification and transfer to a specialist hospital that can perform mechanical thrombectomy is the top priority for LVO stroke patients. This software is designed to help identify appropriate patients earlier and more often, and enable faster transfer and treatment, which, ultimately may result in improved patient outcomes,” Dr. Raul Nogueira of Emory University said in a prepared statement. “This 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,” FDA Center for Devices and Radiological Health’s radiological health acting deputy director Robert Ochs said in a prepared release. Approval came supported by a 300-patient performance study of the system, with results indicating an AUC of 0.91, a 90% sensitivity to identifying LVOs and alerting specialists, and a...
Source: Mass Device - Category: Medical Devices Authors: Tags: Food & Drug Administration (FDA) Regulatory/Compliance Software / IT vizai Source Type: news