Deep learning can rapidly detect stroke causing blockages and potentially speed the onset of lifesaving treatment, according to a study in Radiology.Large vessel occlusions, which block arteries to the brain, account for many ischemic strokes. Prompt diagnosis is critical to begin opening the blocked artery through endovascular therapy.
Matthew Stib RES’21, MD, a diagnostic radiology resident and lead author, used deep learning to help spot large vessel occlusions on CT angiography (CTA) and reduce time to treatment. When the researchers tested the model in 62 patients, it detected all 31 large vessel occlusions.
The model could be a useful asset to medical centers that don’t have the expertise for reading large vessel occlusion CTA images.
“This algorithm is not replacing the ability of radiologists to do their job,” Stib says. “It’s trying to speed up the time to diagnosis.”