AI IDs candidates for endovascular thrombectomy

Researchers have developed an AI algorithm that can help identify patients who have suffered a stroke and would benefit from an endovascular thrombectomy. The team shared its findings in Stroke.

“With endovascular thrombectomy, we now have a treatment for ischemic stroke that is really revolutionary,” corresponding author Sunil A. Sheth, MD, an assistant professor at the University of Texas Health Science Center in Houston, said in a prepared statement. “It allows us to take stroke patients from severe disability and return them to an almost normal life. Unfortunately, the advanced imaging techniques used currently to identify which patients benefit from this procedure are not widely available outside of large referral hospitals.”

To address this issue, the authors developed a machine learning-based algorithm, which analyzes CT angiogram (CTA) findings and learns to pick up on specific image patterns. The AI tool was tested on data from more than 200 patients who had experienced a stroke.

The algorithm, known as DeepSymNet, studied CTA imaging results of those patients and was able to identify blocked cerebral blood vessels. Its area under the ROC curve was 0.88.

“The advantage is you don't have to be at an academic health center or a tertiary care hospital to determine whether this treatment would benefit the patient. And best of all, CT angiogram is already widely used for patients with stroke,” Sheth said in the same prepared statement.