Blink of an eye to measuring stress and attentiveness: AI for healthcare and education
Can Artificial Intelligence understand us better than we understand ourselves?
AI is now the main power of public healthcare systems transformation. As a result of AI-related technology, self-care became easier and more accurate than ever before.
More and more of us start using a variety of gadgets and apps that allow tracking health status. Some companies take it to the next level. This week we chatted with a nice guy from South Korea, who is working on a contactless health analysis gadget. The app's name is CardiVu, and their main tech is called Automatic Vital Sign Extraction Algorithm, promising over 90% of the accuracy of vital signs. It appears that regular cameras from your smartphone or laptop can reveal your stress level just by analyzing your eye. Particularly, by the way, your iris moves. You don’t even have to interrupt your work for the analysis — just keep doing what you’re doing, and the app will tell you how stressed you are.
Oh, and something about your health as well. So, how does it really work? As usual, it starts with collecting data. Lots of data. And finding out what’s the correlation between the vital signs and human eye iris micromovements. The more data we gather, the more accurate the results are. The data monitors include heartbeat rate, fluctuations, low-frequency to high-frequency band ratio, and others.
Turns out, when the iris is too calm, you’re in trouble. And some time later these measurements are applied to you, and you can learn a lot about your current self. If you’ve been using the app for a while, you can also see the history of your past measurements. Not bad for contactless technology, right?
Heart Rate Variability, or HRV, remains one of the core indicators. The Smart Diagnosis researchers also believe that “HRV has been proven to be a very powerful biomarker”. Also, it’s an accurate, non-invasive measure of the Autonomous Nervous System.
Healthy people are supposed to experience large and complex heart rate changes, but complexity is significantly reduced if they have a disease or stress. Having got the approval of KFDA, Smart Diagnosis went further and applied the algorithm to the education industry.
The rise of online educational courses & platforms leads to a question on how to measure the involvement and attentiveness of students.
Even if students’ cameras are turned on, who knows what’s going on in their head, do they really try to grasp the material? Or maybe there’s a game open in a separate browser window?
Of course, it’s hard to achieve involvement similar to personal lessons and discussions, but we could at least start by tracking it. In this case, the camera is more focused on gaze tracking, as opposed to iris contractions. The app tracks time spent with focus, as well as time spent without focus on the screen. The app can also show us a heat map of the student’s attention, highlighting the main areas of the screen where the student was focused on most during a certain time.
Some other tracked factors are blinking frequency, pupil reaction, and head position. What do you guys think about technology? Is it useful? Dangerous? Ethical enough? Would you use it if you could? These guys are currently in the alpha stage, so we still have some time to think about the options before it hits the market.
Let us know!
Also available in audio format here.
Created by Zfort Group.
