AI Is Better at Recognizing Stroke Symptoms Than Humans
Howdy! I'm Keith Ward, and this is my newsletter about the ups and downs of growing old. If you like it, please subscribe (and pass it along to others, too):
Let’s get to it…
——
Medical experts agree that the single most important factor in treating stroke victims is to get them help fast. Really fast. And it’s looking like artificial intelligence may be better than humans at recognizing the signs of a stroke and getting victims that help more quickly.
That’s what Danish researchers found in a recent study using an analysis of about 1.5 million calls made to Copenhagen Emergency Medical Services between 2015 and 2020. Those calls included about 7,000 stroke-related calls.
The findings, which were recently released at the European Stroke Organisation Conference (ESOC) 2023, showed that AI was about 10% better at identifying a potential stroke than human phone responders.
That has big implications for all of us. I didn’t realize this, but strokes are the second-leading cause of death in Europe. And as the population over there (as well as in America) continues to get older, strokes will affect a larger share of people.
I have definite concerns about AI and the unforeseen ramifications in many areas of life. But there’s also no doubt that it can be a good thing if properly applied, and this is one of those cases.
Personal Experience
I have a personal interest in this as well, in that I may have suffered a TIA (sometimes called a “mini-stroke”) a little while ago. There is some disagreement about this, and conflicting evidence. But ever since the event, I’m more aware of the issue than ever. And anything that improves the ability to spot a stroke is huge.
I’ll be writing more about my alleged TIA in a future column. In the meantime, I have to say I’m all for using AI where it makes sense. And this is one case where it makes perfect sense.
One of the study’s authors said that the tech will only improve. “As with any new tool, further research and development are necessary to improve the framework’s accuracy and expand its capabilities. In the future, it may be possible to train the framework directly from the call audio, bypassing the transcription step, as well as incorporating non-word audio – such as a slurred voice – into the training data. However, given the promising results of this study, it is already clear that technologies like this have the capability to completely transform stroke diagnosis and care.”
Amen.
(If you’ve had a stroke, let us know in the comments below. And let us know how you’re doing now).