Scientists are hoping to identify brain wave patterns associated with the risk of dementia
AI can monitor brain health and detect early signs of dementia by analyzing data from brain scans, EEG tests, and even movement patterns. These technologies can identify subtle changes in brain activity or cognitive function long before symptoms appear, allowing for earlier diagnosis and intervention. Broadcast Retirement Network’s Jeffrey H. Snyder discusses applying this technology with NYU Langone’sArthur Caplan, PhD.,
Jeffrey H. Snyder, Broadcast Retirement Network
This morning on BRN, how AI can monitor brain health and find dementia sooner. Joining me now to discuss this, Dr. Arthur Kaplan is the professor of bioethics at NYU Langone Health. Dr. Kaplan, so great to see you. Thanks for joining us on the program this morning. Thank you very much for having me. This is a, as I was telling you in the green room, the virtual green room, this is a fascinating topic because on the one hand, I think everyone’s talking about artificial intelligence.
On the other hand, dementia, Alzheimer’s, some of the mental cognitive diseases have really gotten a lot of news and press in recent weeks, months, and in the last year. So let’s talk about, I want to, my first question to you is how can artificial intelligence in its current form be leveraged to assist with monitoring brain health?
Arthur Caplan, PhD., NYU Langone
Well, there are many, many ways in which artificial intelligence can contribute to our understanding of brain diseases like Alzheimer’s, like lowly body syndrome, other even diseases like Parkinsonism, which is a limited damage area to the brain that can have big physical consequences in terms of mobility and gait. First, artificial intelligence can be programmed to study patient observable behavior. You can take lots of tapes of many people and boil them down and start to say, you know what, there are patterns here that indicate the onset of Alzheimer’s, sometimes before human observers would notice, sometimes because you can get a big, big data set, a library, if you will, of people who are pre-dementia.
And doctors, while they see a lot of patients, they don’t see the range that AI can see. So contributing just through behavioral diagnosis, which we don’t have, which would be wonderful for a variety of reasons to be able to make an early diagnosis earlier, partly for peace of mind, so people would know what’s going on or isn’t going on for that matter. Secondly, AI can look at brain scans.
And while we have brain scans now, they’re not really good at picking up sort of the subtle changes in the brain that indicate Alzheimer’s. To put it simply, gunk building up in the brain around cells. And that gunk, by the time we can see it, it’s usually too late to do anything about it.
That’s why it’s hard to treat Alzheimer’s, because you can’t diagnosis through the best scans we’ve got today until the damage is done, or on an autopsy, which is clearly way too late. But if you could get AI to work with much more sensitive scanners, again, you might start to see subtle changes taking place in the brain. And that would allow you to do a couple of things.
You might then study those people with new drugs to see if you could slow the process of Alzheimer’s. Let me say that again, because by the time we figure out if you have Alzheimer’s, you are, in my opinion, not someone we could fix. It’s like saying my entire body is shot through with cancer.
Let me try some drugs and see if I can cure you. You’re not going to do that because the damage is done.
Jeffrey H. Snyder, Broadcast Retirement Network
Sure. So this sounds like, again, this is another application of the technology, and you need data. You need to be able to capture the data.
Do we have the data sets? I guess we’re talking prospectively, but do we have the data to be able to have the artificial intelligence look through and cull through to make those determinations? Does that exist today, or do we need to build that?
We have big, big data sets out there in genetics.
Arthur Caplan, PhD., NYU Langone
We don’t have if you will, from broad segments of humanity, brain scans. People have brain scans. They get them, but they tend to be stored at individual hospitals, not merged together into worldwide, international imagery and data sets.
That is something we have to do. One bit of good news, we’re mapping the brain. You’ve heard, many of the listeners will have heard, reviewers will have heard that we’ve mapped the human genome.
It’s partly true we haven’t mapped it down to the most precise detail, but we have a pretty good map. It’s like we do have a street map. I’m not sure we’ve got a map of the genes that goes house by house.
In the brain area, we’re just starting to map out the cities, and we’re starting to learn where the key areas of function are for the brain. You can see advances are going to be possible both by mapping what the average brain looks like, the normal brain. We need data sets from those people to compare and contrast with the scans that we tend to have now, which are of people who have illness or showing symptoms.
You can’t really use the big data set for the brain without having a baseline of the normal, if that makes sense.
Jeffrey H. Snyder, Broadcast Retirement Network
If we were able to detect this early, so AI could see the subtle changes, alert the practitioner, is the disease at a point where it’s preventable? You can take appropriate steps, for example, using medications or using other types of therapies to, in your terms, alleviate the gunk, the blockages that exist in the brain. Is detecting it early paramount to reversing or eliminating the risk?
Arthur Caplan, PhD., NYU Langone
We don’t have gunk busters yet, but that’s because we haven’t really been able to study them in people with incipient or just early onset problems. When you identify those people, you put them in trials, you put your drug development there, you stop trying to do things that reverse Alzheimer’s, because that’s proven to be futile. Basically, it hasn’t worked.
All the drug claims that are out there now, I’m not persuaded that there’s anything that’s of any real value to anybody. You might get something that slows the process down a little bit in a severely impacted person for months, but that doesn’t do you much good. Identify the beginning of the process.
That’s where the clinical studies can go. I’ll be optimistic. I have no doubt we can identify that group using sophisticated scanning plus AI to analyze vast volumes of data on the normal and the people who are becoming sick.
We will get drugs that slow that process. Maybe people are always saying, next year, two years. Well, this is probably a seven or eight year project.
It’s not next year. I have a lot of faith that if you could just get the disease at its early stages and figure out who’s got it, that’s where the studies will deliver.
Jeffrey H. Snyder, Broadcast Retirement Network
Yeah. It’s similar to other diseases like cancer and heart disease. If you can detect it early, then you have the ability to prevent it and alleviate damage to those.
Arthur Caplan, PhD., NYU Langone
Here’s an area where AI is making a contribution, breast cancer. We know that early detection is great. You ask women to self-examine and look for lumps.
Obviously, they come in for their mammograms. If you have more sensitive breast cancer scanners and bigger pools of data, you again start to see earlier changes in the breast that indicate cancer. You can intervene and we’ve got much better survival rates.
You’re right, Jeff. That’s exactly what you want to do with Alzheimer’s.
Jeffrey H. Snyder, Broadcast Retirement Network
Let me just ask you, I’m not suggesting this in any way, but are there negatives or are there ethical concerns? Because everyone brings up artificial intelligence. In general, we just saw a strike, the dock worker strike, where they said it’s going to take our jobs.
When people work in organizations, AI is going to take our jobs. There are ethical concerns about having all this data. How do you look at the ethics around what we’re talking about?
I see a lot of the benefit, but there may be some concerns there.
Arthur Caplan, PhD., NYU Langone
The first problem will be recruiting people into studies to see if they have early Alzheimer’s. You might say, why? Why wouldn’t they?
Well, they may not want to know in a world in which there’s still no cure. I do meet people who say, boy, I don’t want to find out that I’m starting down the road to Alzheimer’s. It’s great.
I hope science finds an answer, but I don’t want to live my life that way. I’m not going to sign up for that. I don’t want to do that.
It’s going to freak me out. I’ll have this from the Greek days, sort of Damocles hanging over my head all the time. Is it tomorrow that I’m not going to be able to function?
So there’s that. Second is who owns the data? It would be great if we had publicly accessible, huge databases with everybody contributing their data.
There’s a lot of private equity. There’s a lot of for-profit healthcare that looks at that and says, that’s going to be very profitable. We’re going to own that data.
You want to use it to find cures? Pay us. Well, a lot of the people who had the scans paid for them because they were trying to figure out what was wrong with them.
And it seems that asking them to pay twice is a little bit much. Third is the price of what it’s going to cost. Oh, scanning in AI is not cheap.
It takes a lot of technical expertise, it takes a lot of machinery. By the way, it takes a lot of power. There’s a push in some ways to bring back nuclear power because AI and the computers behind it in the cloud are sucking up so much electricity that we can’t power what we need.
So these are costs and they have to be shared. But in the American healthcare system, it’s usually the rich who get benefits far more quickly than the poor. But Alzheimer’s doesn’t discriminate, so you have an equity issue.
Are we really going to go through a system where for 10 years, the rich are getting the first identified cures, the poor are just waiting?
Jeffrey H. Snyder, Broadcast Retirement Network
This is analogous to so many other aspects of AI and data. Probably more to come in terms of how that all sorts itself out in terms of privacy. Dr. Kaplan, we’re going to have to leave it there. Great to meet you. Great to see you. Thanks so much for joining us, and we look forward to having you back on the program again very soon.
My pleasure. Thank you. And don’t forget to subscribe to our daily newsletter, The Morning Pulse, for all the news in one place.
Details, of course, at our website. And we’re back again tomorrow for another edition of BRN. Until then, I’m Jeff Snyder.
Stay safe, keep on saving, and don’t forget, roll with the changes.
#Monitor #Brain #Health #Find #Dementia #Sooner