LVO strokes can be detected by AI

The study, presented at the Society of NeuroInterventional Surgery’s nineteenth annual meeting, detects gaze deviation from a non-contrast computed tomography (NCCT scan) using AI. These scans identify whether a patient has an LVO, a type of ischemic stroke occurring when a significant artery in the brain is blocked. EVT (endovascular therapy) can be used to treat if a patient has a stroke.

EVT uses catheters to rapidly reopen blocked arteries in the brain and is a minimally invasive procedure. The number of neurons the brain loses is equivalent to the amount that would be lost in almost three and a half years of regular ageing with each hour in which treatment fails to occur. This is why patients need to undergo this procedure as soon as possible.

The AI algorithm in the study was trained using a set of two hundred scans to identify gaze direction. Then clinical symptoms of ipsiversive gaze deviation in one hundred and sixteen stroke patients with LVO, treated with EVT. The angle between the brain’s midline and gaze direction was measured to calculate the deviation. The algorithm correctly recognised seventy-nine per cent of proximal occlusions with an ipsiversive gaze deviation but was less effective in recognising cases with less extreme symptoms.

The study’s lead author, Jason Tarpley, MD, PhD, stated: “Endovascular thrombectomy for acute ischemic stroke is one of the most effective therapies in medicine, and its efficacy is critically dependent on time. Simply put, the faster we act, the better our stroke patients’ outcomes will be. Our results represent an advance that has the potential to speed up the identification of LVO stroke during the triage process at the hospital.”