The Impact of the Algorithm: Q&A with Endovascular Neurosurgeon Gal Yaniv

In November, we announced the FDA-clearance of our latest vessel occlusion (VO) algorithm, making Aidoc the first and only healthcare AI provider to gain clearance to flag intracranial large vessel occlusions (LVOs) and medium vessel occlusions (MeVOs). However, the benefits don’t stop at flagging suspected positive cases. In fact, the value of such a solution is proving to enhance interventionist workflows, leading to improved patient outcomes. 

To learn about the downstream value of the VO clearance and what it entails for VO workflows, read the Q&A below with Dr. Gal Yaniv, Director of Endovascular Neurosurgery at Sheba Medical Center and Aidoc’s Co-Founder and Chief Medical Officer:

What role can AI play in improving patient care for distal occlusions?

It’s important to understand that the more distal the occlusion, the less pronounced stroke symptoms may be when compared to proximal occlusions. Even M2 or M3 occlusions can, on the surface, appear to be more similar to confusion or inebriation than a stroke. The lack of presenting symptoms can impact the level of suspicion of an occlusion for the radiologist. This makes prioritization, or the lack thereof, problematic. 

Second, many stroke workflows are done on cell phones now via apps, including Aidoc. Obviously, a cell phone screen isn’t nearly as big as a workstation, making it more difficult to see a small VO. The AI can help prioritize a patient with a suspected VO finding and, in turn, help interventionalists decide more quickly and accurately about next steps.

What value does AI provide in not only radiologist’s MeVO workflows but also downstream for neuro interventionalists?

It’s not only more difficult to identify more distal occlusions on the image, but once identified, the intervention on this kind of occlusion has a chance of being more dangerous for the patient, meaning the risk for complication gets higher. You run a risk of creating more bleeding or damage than before you intervened. That’s another reason why imaging is so crucial. It’s essential to get an exact understanding of the anatomy and occlusion site of the patient so you can more confidently decide what to do with them and whether surgical intervention is the best option. Under some conditions, blood thinners can be a better course of action. So having the AI there with its ability to flag potential suspected findings for smaller occlusions gives radiologists a better chance at diagnosing these often hard-to-find occlusions in a timely manner, translating into a benefit in which the interventionalist has a better chance of choosing the right treatment path.

What does it mean for neuroradiologist workflows to have the assistance of AI in flagging large and medium occlusions?

For positive scans, the benefits are obviously flagging the occlusion in the first place, then you can detect it for yourself, write a report and move onto the next steps. Stroke care is exceedingly complicated, and it takes a lot of time to make sense of relevant patient data. This is a huge value point for AI: its ability to dig into the EMR, find and extract relevant data, helps the physician(s) to understand whether this patient has a real LVO, MeVO or something else. An intelligent EMR extraction by AI will improve that workflow and efficiency of the radiologist, spreading that benefit to the neurologist, endovascular neurosurgeon and everyone who is involved with the treatment of the patient. 

What sort of downstream impact do you anticipate with patient outcomes and revenue generating opportunities?

Since patients will be diagnosed more quickly and, potentially, patient outcomes will be improved. Time is brain, so every minute saved during a stroke can result in drastic improvements for patient outcomes. 

As far as revenue generating opportunities, the return for stroke and endovascular intervention is very high. So the fact that we have a technology that can help radiologists identify most MeVOs in a timely manner can increase the numbers of mechanical and endovascular interventions. It’s a win-win no matter how you look at it.

Explore the Latest AI Insights, Trends and Research