The emergence of artificial intelligence (AI) in healthcare has been groundbreaking, reshaping the way clinicians diagnose, treat, and monitor patients. There are a few different types of AI, including machine learning, which gives an ability to learn from experience without being explicitly programmed, and deep learning, which learns from raw/nearly raw data without the need for feature engineering. Through AI in healthcare, medical professionals can make more decisions based on accurate information, ultimately saving time, reducing costs, and improving medical records management overall.
At the 2023 Americas Committee for Treatment and Research in Multiple Sclerosis (ACTRIMS) Forum, held February 23-25, in San Diego, California, attendee Michael Dwyer, PhD, presented a talk on the use of AI and MRI for multiple sclerosis (MS) care. In his presentation, he focused on how these approaches help improve the care and accessibility for patients, through image segmentation, anomaly detection, and facilitating the transition of novel MRI biomarkers, among others. Additionally, he spoke about ChatGPT, a recently launched AI chatbot trained to follow an instruction and provide a detailed a detailed response.
Following his presentation, Dwyer sat down with NeurologyLive® to discuss some of the recent successes with AI, and how the field is quickly adapting to it. As part of a new iteration of NeuroVoices, Dwyer, director of IT and Neuroinformatics Development at the Buffalo Neuroimaging Analysis Center, provided insight on how these approaches can be beneficial in the diagnosis, treatment, and management of MS going forward.
NeurologyLive®: What are some of the recent successes with AI?
Michael Dwyer, PhD. AI is obviously a very hot topic right now. We’re all hearing about ChatGPT, and how these things are taking the world by storm. It’s amazing, there’s a lot of hype, but we have to be careful, especially trying to pull these things into the clinical side. It’s one thing to have a student use ChatGPT to help with their term paper, it’s another thing to try to leverage AI to say, “what should we do to treat somebody for a medical disease?” Obviously, the stakes are higher for making mistakes. So yes, we need to be very careful.
It is an incredible time to be seeing what’s happening in AI because this whole deep learning revolution has really changed the way everything works. The groundwork has been there for decades—the math, the theory of how to do things—but the ability to have these deep learning tools, and the data sets with it, just shows the explosion in the last 10 years. In my presentation, I mentioned briefly that the FDA has a tracked list now of AI-approved devices. You see this huge inflection point at 2015, where suddenly, it just shoots up. And 95% of those [approvals] are in radiology, which, for a disease like multiple sclerosis, radiology and imaging is one of the core components. There’s been so much forward motion there. I think we have to be careful. It’s we’re nowhere near what a lot of people think of in terms of AI, this kind of Terminator artificial general intelligence, where the machines are going to take over. But we now have a set of tools that can help clinicians and researchers transform what they can do.
How can AI be more heavily applied to multiple sclerosis going forward?
That’s what’s very interesting. We’ve talked before a little bit about AI, specifically about some image segmentation tools and looking at those biomarkers, but what’s been really interesting is the breadth of areas that AI has been able to help with. It’s not just the classic looking at a picture and making measurements. That’s just the tip of the iceberg. What we’ve also seen is AI under the hood. It’s not necessarily the big, exciting stuff, but it’s making really important changes. For example, the reconstruction on MRI is being done now in many cases with AI, and these AI tools have been incorporated by the manufacturer. It’s already being translated, it’s not just this theoretical research. There are caveats, obviously, we need to be careful about interpreting these images, but they can speed up the images by 6-fold sometimes, and that’s huge because MRI is still the most expensive component of diagnostic workups and following these patients over time.
If we could do MRI more frequently, it would help us monitor better, treatment monitoring, and help us understand what’s happening with patients. In terms of availability of care, MRI is expensive, it’s not as easily available in all communities. This is a kind of leveler that helps make it possible to access MRI. That’s one very important area. Another is translating the academic kind of research we do. MRI is a very complicated modality, and we have different types. We have the standard type we do clinically, but then there’s also there’s also a whole set of MRIs sequences that are purely research. We had some colleagues in Amsterdam, for example, just developed a technique to synthesize double inversion recovery images from traditional images. In MS, there’s a whole set of pathology we don’t see called cortical lesions. We can see those now with some of these AI techniques.
Again, there’s caveats, we have to be careful, and we need to go step-by-step, but those steps are happening. In terms of prognosis, it’s been also very helpful. It’s helped translate biomarkers that couldn’t be easily translated before. There are groups that have looked on this central vein sign biomarker, and there’s AI tools now that have been developed by our colleagues to look at that. A lot of exciting things happening in the field.
An up-and-coming phenomenon, artificial intelligence (AI) technologies have the potential to transform many aspects of patient care, as well as administrative processes within provider, payer, and pharmaceutical organizations. There are several types of AI, including machine learning, which learned from experience without being explicitly programmed, and deep learning, which learns from raw/nearly raw data without the need for feature engineering.
Researchers at the Buffalo Neuroimaging Analysis Center (BNAC) have been at the forefront of developing and validating AI-like algorithms to improve patient care, specifically for those with multiple sclerosis (MS). At the 2023 Americas Committee for Treatment and Research in Multiple Sclerosis (ACTRIMS) Forum, February 23-25, in San Diego, California, Michael Dwyer, PhD, presented a talk on the use of AI and MRI for MS care, and how each can be applied. In his presentation, Dwyer discussed several areas of potential, including MR acquisition, image segmentation, diagnosis, prognosis, and others.
At the forum, Dwyer, director of IT and Neuroinformatics Development at BNAC, spoke on the increased exposure to AI and whether these techniques should be a part of medical schooling going forward. Additionally, he discussed the many ways AI can help both patients with MS and their clinicians in terms of tracking long-term disease progression and identifying erroneous patterns.
NeurologyLive®: Should learning AI be incorporated more into the neurology education space?
Michael Dwyer, PhD: That’s a good question, I don’t know that I have a short answer for it. think they should be aware. So I would answer that two ways. First, I’m going to give a very boring answer, because I think there’s no replacement for basics, the basic statistics, basic familiarity with how to do hypothesis testing. AI is a wonderful, powerful tool, but it is also so powerful that it can fool us very easily.
We’ve seen a lot of AI techniques that seemed promising, and then fizzled out because the statistical foundations weren’t necessarily there, we didn’t test them properly, or we trained them on one dataset, and then it doesn’t translate or work on the other one. I think that it should be part of a more holistic statistical and general kind of research methods framework. For clinicians and the general public, they don’t need to know how to do deep learning. You don’t need to know how to sit there and program something in pytorch. What I think they need to know how to do right now—because of the explosion that we were talking about—is to separate the baby from the bathwater, and how to recognize what’s a reliable AI tool? What can I trust, what can’t I trust.
The editors of Radiology, and the newer journal, Radiology Artificial Intelligence, have released guidelines for clinicians called the CLAIM guidelines and for publishers to have checklists for AI to be properly unethically used in these areas. That kind of thing is very important for people to understand. What’s good AI? What’s bad AI? How much you should use? And how you should use it? Take ChatGPT, that something everybody is so excited about. It is an amazing tool if you’re trying to write a document. If you look at clinicians, and the real world of clinical day to day, they spend a lot of their time filling out templates. Where they fill out the forms, ChatGPT type technology can probably help make those templates much better. But you have to use it in a way where an expert is reviewing everything that its saying, and you can’t rely on it. That’s the key. We see the negative where students use it to write their term paper. But it’s just a tool that can be used badly or it can be used to great value. We need to balance those things.
Are there ways in which AI can be used specifically to monitor disease progression?
There’s a couple of areas where it can help with that. A lot of people think about AI as replicating what humans. We train a model to do something faster, or maybe more reliably, but not fundamentally different. That’s called supervised learning. We tell it what we want it to learn. Unsupervised learning is where we tell an AI tool to look at data and see if it can find patterns. There have been some really interesting advances in that with clustering, for example. Arman Eshaghi and his group in the UK, for example, were able to identify latent clusters of MS, different types of disease pathology and say, “this, this may have a different progression going forward.” If we can identify those kinds of subtypes early, we can potentially help intervene earlier and know whether people have different responses to different treatments.
Another area is being able to synthesize data from a lot of different places. Humans are good at analysis and reduction, we’re not always great at putting together lots of data points in the same way. These AI tools can be a very helpful assistant to integrate, genomics, connectomics, other serum markers, and imaging, all together to make predictions based on lots of data points, as opposed to the kind of clinical algorithms where we just look at 2 or 3 things. That’s another potential way that that we can [use AI], and we are already starting to see a shift in that.
I should mention too, deep learning has been a big buzzword for a while. What sets this whole deep learning thing apart as opposed to from traditional machine learning, it learns on raw data. With traditional machine learning, it would still learn the rules, but you had to decide what it was going to look at, you had to take an image and say, “I’m going to measure the thalamus, the cortex, the amount of lesions or, I’m going to take a clinical assessment, I’m going to take the EDSS into these 4 scores or something, or these specific sub scores.” With deep learning, you feed much rawer data. For us, it’s working for raw MRI. Instead of extracting those features, we just tell it to look who’s progressing and who’s not and try to find predictors from the images. There are people doing gait analyses, people doing wearables, and it’s all from the raw data so we don’t have to have somebody sit there and be a gatekeeper of the information and say, “These are the features we should pull out.”
It’s very powerful there because it can pick up on things we miss. It can pick up on subtleties that we may not see: maybe that a part of the thalamus is important, but the other isn’t, and when we just go and extract the thalamus, we lose that. It’s potentially powerful, and it’s a very exciting field. We’re going to see a lot more going forward. I get we need to be careful. We need to go with our eyes open. We need to go step by step and validate these tools carefully. But there’s tremendous value here.
This Article was a 2 parts post seen on “https://www.neurologylive.com/view/neurovoices-michael-dwyer-phd-appropriately-incorporating-artificial-intelligence-chatgpt-into-multiple-sclerosis-care” with the title : NeuroVoices: Michael Dwyer, PhD, on Appropriately Incorporating Artificial Intelligence, ChatGPT into Multiple Sclerosis Care
by Marco Meglio on March 8,2023