Partnership Advances AI Tool for Predicting Malignant Cerebral Edema

Early identification remains difficult in neurocritical care, yet timely recognition can enable intensified monitoring or decompressive surgery before neurological deterioration occurs. Clinicians therefore need decision support that integrates early imaging with clinical data to help identify patients at highest risk. Innolitics and BEACON-Neuro.AI have advanced BEACONPredict, an AI tool designed to predict malignant cerebral edema risk.

BEACONPredict is being developed by BEACON-Neuro.AI, a startup from Washington University in St. Louis, with regulatory strategy and software engineering support from Innolitics. The clinical science originated in the neurocritical care program at WashU Medicine. Through this collaboration, the software has moved from a research prototype to U.S. Food and Drug Administration (FDA) Breakthrough Device status on the first review cycle.

The software analyzes early brain imaging changes together with clinical variables to generate a patient-level risk estimate for malignant cerebral edema after acute ischemic stroke. The output is intended to support clinician decision-making when considering escalation of monitoring or surgical intervention. The tool is positioned for use as clinical decision support within neurocritical care workflows.

The program is pursuing an FDA De Novo pathway to market. BEACONPredict has also been accepted into the FDA Total Product Life Cycle Advisory Program (TAP) pilot, enabling enhanced engagement with the agency during development. Participation may also help position the technology for a potentially accelerated Centers for Medicare & Medicaid Services (CMS) New Technology Add-on Payment (NTAP) pathway after authorization.

“BEACON-Neuro.AI brought new science to address a clinical need nobody was serving well. Our job was to give it a route the FDA would accept, and the designation came through on the first cycle. That happens when the clinical team and the regulatory team are working off one plan instead of passing documents back and forth,” said Yujan Shrestha, M.D., partner, Innolitics.

“We built the science at WashU Medicine, but taking it to patients means clearing the FDA, and that is not what a research lab is set up to do. Partnering with Innolitics, BEACON-Neuro.AI took BEACONPredict from lab code to an FDA Breakthrough Device in a single submission cycle, and we are building the product together now,” said Raj Dhar, M.D., co-founder of BEACON-Neuro.AI and professor of Neurology, WashU Medicine.

Related Links
BEACON-Neuro.AI
Innolitics