UPMC’s stroke network is testing new ways to get patients to advanced stroke care faster. In a study published in the Journal of Neurology, Neurosurgery and Psychiatry, UPMC and University of Pittsburgh researchers found that an AI-assisted scan‑sharing and alert app can do just that.
Mohamed (Mo) Doheim, M.D.
Lead author of this research, Dr. Mohamed (Mo) Doheim, research assistant professor of neurology at Pitt and NIH StrokeNet fellow at the UPMC Stroke Institute, answers questions about the study’s findings and what they mean for patients and healthcare providers.
What problem was this study trying to solve?
Dr. Doheim: Each year, nearly 800,000 Americans have a stroke, which is a leading cause of long‑term disability and among the leading causes of death in the United States.
When someone has a severe stroke, it is essential to get them to a major hospital for advanced treatment quickly. For every 15‑minute delay in care, the chance of disability‑free survival drops by about 4%. But getting the patient transferred from a smaller hospital to a center that has more extensive stroke care capabilities isn’t instantaneous — it often requires coordinating limited resources, arranging transport and navigating logistical barriers.
The UPMC Stroke Institute team, guided by its director, Dr. Raul Nogueira, wanted to determine if an AI-powered phone app that instantly shares brain scans and alerts the stroke team at the receiving hospital could speed up this process and get patients faster treatment.
What were your most important findings?
Dr. Doheim: We looked at data from over 4,500 stroke patients across UPMC’s network of four stroke centers receiving transfers from 60 community hospitals. Then we compared what happened with the AI system versus without it to see whether it changed timing, transfers and treatment.
Our findings were very encouraging:
• More transferred patients received the life-saving clot-removal procedure.
• Patients transferred from hospitals using the AI system waited about a half hour less before transfer. Once they arrived at the main hospital, their treatment also started sooner.
- • Smaller community hospitals saw the biggest improvements in their treatment times.
- • And, finally, because AI helped doctors easily identify which patients truly needed the clot-removal procedure, it prevented unnecessary transfers.
What do these findings mean for patients and hospitals?
Dr. Doheim: For patients, this means faster, more equal access to life-saving treatments, no matter which hospital they go to first.
For hospitals, this shows that AI could be a powerful tool to make stroke care more efficient by helping doctors communicate instantly and look at scans together on their smart phones, rather than waiting for phone calls or faxes.
Saving time during a stroke is crucial to saving brain tissue, so these faster treatment times should ultimately lead to better, healthier recoveries.
What’s next?
Dr. Doheim: While this is the largest study of its kind in the United States showing that AI improves stroke care times, we still have more to learn. Next, we need to track how patients are doing in the long term after their stroke to prove that these faster times directly lead to better long-term health. We also want to look closer at the outcomes of the patients who weren’t transferred, to make sure that we are making the safest, fastest and most highly informed decisions.
Additional Resources:
UPMC Stroke Institute | Department of Neurology
About Stroke | American Stroke Association
Related Posts:
