As artificial intelligence continues to dominate Wall Street's attention, conversations typically revolve around productivity, among other more existential concerns.
But can AI be used to treat the most serious and prevalent cognitive diseases?
Linguist and cognitive scientist Dr. Coral Ho certainly thinks so, having dedicated 30 years of her career to developing Dysolve, an AI-powered system specifically designed to treat dyslexia.
Hoh explains how the innovative technology works and why artificial intelligence is needed to fill the treatment gap in the video above.
Transcription:
Dr. Coral Ho: Please try to imagine. You need an expert for something like this to diagnose and correct brain conditions. So what we've done is actually put experts into the machine. So what’s a professional to do?
Let's just say it's the child who needs to be diagnosed. As you know, the intervention must be given to the child. Professionals need to interact with children in some way. Observe what is happening correctly. So Dysolve creates games. Generate a game or build a game and give it to your child through your device. And the kids play games.
Dysolve AI now has a way to observe what's happening. Was the child able to do this language activity or was it a mistake? Well, if it's correct, break up. Let's explore another game. Maybe other regions. However, if there is an error in this particular activity, please disband. I want to figure out what's going on and investigate further.
If the results are ambiguous and ambiguous, there is a problem with Dysolve. What should I do next? But unlike me, we have to sit there and think for a while. Dissolve requires immediate action. So we'll break up.
Just generate one game after another so your child doesn't have to wait. When it comes to language processing, it needs to run super fast.
So the question is, why build a computer expert system? Why weren't human experts enough to do this?
So there are three elements. One is complexity. Another factor is speed. And the third thing is capacity. Now, let's deal with them one by one.
Complexity: This is a language processing issue. So we need to consider the brain's language system. It's like a super-complex system, like a large computer operating system with millions of lines of code. Therefore, varying degrees of inefficiency can occur in any part of the system.
How do humans find them? That's one thing, right? The other thing, as I said earlier, is that you shouldn't keep your child waiting. right? So it's a matter of speed. And natural language is processed in parallel in the brain at lightning speed within a few hundred milliseconds, less than a second. Many, many things are happening at the same time.
Therefore, there is no way for humans to record, adjust, or calculate it. The third thing is of course the scale, right? Even if one person can be in charge of one child, what about the 30 million students across the country? At least 30 million for this kind of thing. Part of the problem is cost and accessibility.
