Exploring the Future of Automated Eye Movement Analysis

Dysfunction of the supranuclear ocular motor pathways typically causes highly localizable deficits. With sophisticated neuroimaging, it is critical to better understand structure-function relationships and precisely localize pathology within the brain.

In this episode, Lyell K. Jones Jr, MD, FAAN, speaks with Gregory P. Van Stavern, MD, author of the article “Supranuclear Disorders of Eye Movements” in the Continuum® April 2025 Neuro-ophthalmology issue.

Dr. Jones is the editor-in-chief of Continuum: Lifelong Learning in Neurology® and is a professor of neurology at Mayo Clinic in Rochester, Minnesota.

Dr. Van Stavern is the Robert C. Drews professor of ophthalmology and visual sciences at Washington University in St Louis, Missouri.

Additional Resources

Read the article: Internuclear and Supranuclear Disorders of Eye Movements

Subscribe to Continuum®: shop.lww.com/Continuum

Earn CME (available only to AAN members): continpub.com/AudioCME

Continuum® Aloud (verbatim audio-book style recordings of articles available only to Continuum® subscribers): continpub.com/Aloud

More about the American Academy of Neurology: aan.com

Social Media

facebook.com/continuumcme

@ContinuumAAN

Host: @LyellJ

Full episode transcript available here

Dr Jones: This is Dr Lyell Jones, Editor-in-Chief of Continuum. Thank you for listening to Continuum Audio. Be sure to visit the links in the episode notes for information about earning CME, subscribing to the journal, and exclusive access to interviews not featured on the podcast.

Dr Jones: This is Dr Lyell Jones, Editor-in-Chief of Continuum: Lifelong Learning in Neurology. Today I'm interviewing Dr Gregory Van Stavern, who recently authored an article on intranuclear and supranuclear disorders of eye movements for our latest Continuum issue on neuro-ophthalmology. Dr Van Stavern is the Robert C Drews professor of ophthalmology and visual sciences at Washington University in Saint Louis. Dr Van Stavern, welcome, and thank you for joining us today. Why don't you introduce yourself to our audience?

Dr Van Stavern: Hi, my name is Gregory Van Stavern. I'm a neuro-ophthalmologist located in Saint Louis, and I'm pleased to be on this show today.

Dr Jones: We appreciate you being here, and obviously, any discussion of the visual system is worthwhile. The visual system is important. It's how most of us and most of our patients navigate the world. Roughly 40% of the brain---you can correct me if I'm wrong---is in some way assigned to our visual system. But it's not just about the sensory experience, right? The afferent visual processing. We also have motor systems of control that align our vision and allow us to accurately direct our vision to visual targets of interest. The circuitry is complex, which I think is intimidating to many of us. It's much easier to see a diagram of that than to describe it on a podcast. But I think this is a good opportunity for us to talk about the ocular motor exam and how it helps us localize lesions and, and better understand diagnoses for certain disorders. So, let's get right to it, Dr Van Stavern. If you had from your article, which is outstanding, a single most important message for our listeners about recognizing or treating patients with ocular motor disorders, what would that message be?

Dr Van Stavern: Well, I think if we can basically zoom out a little to the big picture, I think it really emphasizes the continuing importance of the examination. History as well, but the examination. I was reading an article the other day that was essentially downplaying the importance of the physical examination in the modern era with modern imaging techniques and technology. But for neurology, and especially neuro-ophthalmology, the history and the examination should still drive clinical decision-making. And doing a careful assessment of the ocular motor system should be able to tell you exactly where the lesion is located, because it's very easy to order a brain MRI, but the MRI is, like Forrest Gump might say, it's like a box of chocolates. You never know what you're going to find. You may find a lot of things, but because you've done the history and the examination, you can see if whatever lesion is uncovered by the MRI is the lesion that explains what's going on with the patient. So even today, even with the most modern imaging techniques we have, it is still really important to know what you're looking for. And that's where the oculomotor examination can be very helpful.

Dr Jones: I did not have Forrest Gump on my bingo card today, Dr Van Stavern, but that's a really good analogy, right? If you order the MRI, you don't know what you're going to get. And then- and if you don't have a really well-formed question, then sometimes you get misleading information, right? 

Dr Van Stavern: Exactly.

Dr Jones: We'll get into some technology here in a minute, because I think that's relevant for this discussion. I think most of our listeners are going to agree with us that the exam is important in neuro-ophthalmology, and neurology broadly. So, I think you have some sympathetic listeners there. Again, the point of the exam is to localize and then lead to a diagnosis that we can help patients with. When you think about neurologic disorders where the ocular motor exam helps you get to the right diagnosis, obviously disorders of eye movements, but sometimes it's a clue to a broader neurologic syndrome. And you have some nice discussions in your article about the ocular motor clues to Parkinson disease or to progressive supranuclear palsy. Tell us a little more about that. In your practice, which neurologic disorders do you find the ocular motor exam being most helpful?

Dr Van Stavern: Well, just a very brief digression. So, I started off being an ophthalmology resident, and I do two years of ophthalmology and then switch to neurology. And during neurology residency, I was debating which subspecialty to go into, and I realized that neuro-ophthalmology touches every other subspecialty in neurology. And it goes back to the fact that the visual system is so pervasive and widely distributed throughout the brain. So, if you have a neurologic disease, there is a very good chance it is going to affect vision, maybe in a minor way or a major way. That's why careful assessment of the visual system, and particularly the oculomotor system, is really helpful for many neurologic diseases. Neuromuscular disease, obviously, myasthenia gravis and certain myopathies affect the eye movements. Neurodegenerative diseases, in particular Parkinson's disease and parkinsonian conditions, often affect the eye movements. And in particular, when you're trying to differentiate, is this classic Parkinson's disease? Or is this progressive supranuclear palsy? Is it some broad spectrum multisystem atrophy? The differences between the eye movement disorders, even allowing for the fact that there's overlap, can really help point in one direction to the other, and again, prevent unnecessary testing, unnecessary treatment, and so on.

Dr Jones: Very good. And I think, to follow on a thread from that concept with patients who have movement disorders, in my practice, seeing older patients who have a little bit of restriction of vertical gaze is not that uncommon. And it's more common in patients who have idiopathic Parkinson disease. And then we use that part of the exam to help us screen patients for other neurodegenerative syndromes like progressive nuclear- supranuclear palsy. So, do you have any tips for our listeners to- how to look at, maybe, vertical gaze and say, this is maybe a normal age-related degree of change. This is something that might suggest idiopathic Parkinson disease. Or maybe something a little more progressive and sinister like progressive super nuclear palsy?

Dr Van Stavern: Well, I think part of the issue- and it's harder to do this without the visual aspect. One of my colleagues always likes to say for a neurologist, the eye movement exam begins and ends with the neurology benediction, just doing the sign of the cross and checking the eye movements. And that's a good place to start. But I think it's important to remember that all you're looking at is smooth pursuit and range of eye movements, and there's much more to the oculomotor examination than that. There’s other aspects of eye movement. Looking at saccades can be really helpful; in particular, classically, saccadic movements are selectively abnormal in PSP versus Parkinson's with progressive supranuclear palsy. Saccades, which are essentially rapid movements of the eyes---up and down, in this case---are going to be affected in downward gaze. So, the patient is going to have more difficulty initiating downward saccades, slower saccades, and less range of movement of saccades in downgaze. Whereas in Parkinson's, it's classically upward eye movements and upgaze. So, I think that's something you won't be able to see if you're just doing, looking at, you know, your classic, look at your eye movements, which are just assessing, smooth pursuit. Looking carefully at the eye movements during fixation can be helpful.

Another aspect of many parkinsonian conditions is saccadic intrusions, where there's quick movements or saccades of the eye that are interrupting fixation. Much, much more common in PSP than in Parkinson's disease. The saccadic intrusions are what we call square-wave jerks because of what they look like. Eye movement recordings are much larger amplitude in PSP and other multisystem atrophy diseases than with Parkinson's. And none of these are perfect differentiators, but the constellation of those findings, a patient with slow downwards saccades, very large amplitude, and frequent saccadic intrusions might point you more towards this being PSP rather than Parkinson's.

Dr Jones: That's a great pearl, thinking about the saccades in addition to the smooth pursuit. So, thank you for that. And you mentioned eye movement measurements. I think it's simultaneously impressive and a little scary that my phone can tell when I'm looking at it within a few degrees of visual attention. So, I imagine there are automated tools to analyze eye movement. Tell us, what's the state of the art there, and what should our listeners be aware of in terms of tools that are available and what they can and can't do?

Dr Van Stavern: Well, I could tell you, I mean, I see neuro-ophthalmic patients with eye movement disorders every day and we do not have any automated tools for eye movement. We have a ton of imaging techniques for imaging the optic nerve and the retina in different ways, but we don't routinely employ eye movement recording devices. The only time we usually do that is in somebody where we suspect they have a central or peripheral vestibular disease and we send them for vestibular testing, for eye movement recordings. There is interest in using- I know, again, sort of another digression, but if you're looking at the HINTS technique, which is described in the chapter to differentiate central from peripheral disease, which is a very easy, useful way to differentiate central from peripheral or peripheral vestibular disease. And again, in the acute setting, is this a stroke or not a stroke? Is it the brain or is it the inner ear? Part of the problem is that if you're deploying this widespread, the people who are doing it may not be sufficiently good enough at doing the test to differentiate, is a positive or negative test?

And that's where some people have started introducing this into the emergency room, these eye movement recording devices, to give the- using, potentially, AI and algorithms to help the emergency room physicians say, all right, this looks like a stroke, we need to admit the patient, get an MRI and so on, versus, this is vestibular neuritis or an inner ear problem, treat them symptomatically, follow up as an outpatient. That has not yet been widely employed. It's a similar way that a lot of institutions are having fundus photography and OCT devices placed in the emergency room to aid the emergency room physician for patients who present with acute vision issues. So, I think that could be the future. It probably would be something that would be AI-assisted or AI-driven. But I can tell you at least at our institution and most of the ones I know of, it is not routinely employed yet.

Dr Jones: So maybe on the horizon, AI kind of facilitated tools for eye movement disorder interpretation, but it's not ready for prime time yet. Is that a fair summary?

Dr Van Stavern: In my opinion, yes.

Dr Jones: Good to know. This has struck me every time I've read about ocular motor anatomy and ocular motor disorders, whether they're supranuclear or intranuclear disorders. The anatomy is complex, the circuitry is very complicated. Which means I learn it and then I forget it and then I relearn it. But some of the anatomy isn't even fully understood yet. This is a very complex real estate in the brainstem. Why do you think the neurophysiology and neuroanatomy is not fully clarified yet? And is there anything on the horizon that might clarify some of this anatomy?

Dr Van Stavern: The very first time I encountered this topic as an ophthalmology resident and later as a neurology resident, I just couldn't understand how anyone could really understand all of the circuitry involved. And there is a lot of circuitry that is involved in us simply having clear, single binocular vision with the afferent and efferent system working in concert. Even in arch. In my chapter, when you look at the anatomy and physiology of the smooth pursuit system or the vertical gaze pathways, there's a lot of, I'll admit it, there's a lot of hand waving and we don't completely understand it. I think a lot of it has to do with, in the old days, a lot of the anatomy was based on lesions, you know, lesion this area either experimentally or clinically. And that's how you would determine, this is what this region of the brain is responsible for.

Although we've gotten more sophisticated with better imaging, with functional connectivity MRI and so on, all of those have limitations. And that's why I still don't think we completely understand all the way this information is integrated and synthesized, and, to get even more big level and esoteric, how this makes its way into our conscious mind. And that has to do with self-awareness and consciousness, which is a whole other kettle of fish. It's just really complicated. I think when I'm at least talking to other neurologists and residents, I try to keep it as simple as possible from a clinical standpoint. If you see someone with an eye movement problem, try to see if you can localize it to which level you're dealing with. Is it a muscle problem? Is it neuromuscular junction? Is it nerve? Is it nucleus? Is it supranuclear? If you can put it at even one of those two levels, you have eliminated huge territories of neurologic real estate, and that will definitely help you target and tailor your workup. So, again, you're not costing the patient in the healthcare system hundreds of thousands of dollars.

Dr Jones: Great points in there. And I think, you know, if we can't get it down to the rostral interstitial nucleus of the medial longitudinal fasciculus, if we can get it to the brainstem, I think that's obviously- that's helpful in its own right. And I imagine, Dr Van Stavern, managing patients with persistent ocular motor disorders is a challenge. We take foveation for granted, right, when we can create these single cortical images. And I imagine it's important for daily function and difficult for patients who lose that ability to maintain their ocular alignment. What are some of the clinical tools that you use in your practice that our listeners should be aware of to help patients that have a persistent supranuclear disorder of ocular movement?

Dr Van Stavern: Well, I think you tailor your treatment to the symptoms, and if it's directly due to underlying condition, obviously you treat the underlying condition. If they have sixth nerve palsy because of a skull base tumor, obviously you treat the skull base tumor. But from a practical standpoint, I think it depends on what the symptom is, what's causing it, and how much it's affecting their quality of life. And everyone is really different. Some patients have higher levels of tolerance for blurred vision and double vision. For things- for patients who have double vision, depending upon the underlying cause we can sometimes use prisms and glasses. Prisms are simply- a lot of people just think prism is this, like, mystical word that means a lot. It’s simply just an optical device that bends light. So, it essentially bends light to allow the eyes- basically, the image to fall on the fovea in both eyes. And whether the prisms help or not is partly dependent upon how large the misalignment is. If somebody has a large degree of misalignment, you're not going to fix that with prism. The amount of prism you'd need to bend the light enough to land on the fovea in both eyes would cause so much blur and distortion that it would essentially be a glorified patch. So, for small ranges of misalignment, prisms are often very helpful, that we can paste over glasses or grind into glasses. For larger degrees of misalignment that- let's say it is due to some skull base tumor or brain stem lesion that is not going to get better, then eye muscle surgery is a very effective option. We usually like to give people a long enough period of time to make sure there's no change before proceeding with eye muscle surgery.

Dr Jones: Very helpful. So, prisms will help to a limited extent with misalignment, and then surgery is always an option if it's persistent. That's a good pearl for, I think, our listeners to take away.

Dr Van Stavern: And even in those circumstances, even prisms and eye muscle surgery, the goal is primarily to cause single binocular vision and primary gaze at near. Even in those cases, even with the best results, patients are still going to have double vision, eccentric gaze. For most people, that's not a big issue, but we have had a few patients… I had a couple of patients who were truck drivers who were really bothered by the fact that when they look to the left, let's say because it's a 4th nerve palsy on the right, they have double vision. I had a patient who was a golfer who was really, really unhappy with that. Most people are okay with that, but it all depends upon the individual patient and what they use their vision for.

Dr Jones: That's a great point. There's not enough neurologists in the world. I know for a fact there are not enough neuro-ophthalmologists in the world, right? There's just not many people that have that dual expertise. You mentioned that you started with ophthalmology and then did neurology training. What do you think the pipeline looks like for neuro-ophthalmology? Do you see growing interest in this among trainees, or unchanged? What are your thoughts about that?

Dr Van Stavern: No, that's a continuing discussion we're having within our own field about how to attract more residents into neuro-ophthalmology. And there's been a huge shift. In the past, this was primarily ophthalmology-driven. Most neuro-ophthalmologists were trained in ophthalmology initially before doing a fellowship. The last twenty years, it switched. Now there's an almost 50/50 division between neurologists and ophthalmologists, as more neurologists have become more interested. This is probably a topic more for the ophthalmology equivalent of Continuum. One of the perceptions is this is not a surgical subspecialty, so a lot of ophthalmology residents are disincentivized to pursue it. So, we have tried to change that. You can do neuro-ophthalmology and do eye muscle surgery or general ophthalmology. I think it really depends upon whether you have exposure to a neuro-ophthalmologist during your neurology residency. If you do not have any exposure to neuro-ophthalmology, this field will always seem mysterious, a huge black box, something intimidating, and something that is not appealing to a neurologist. I and most of my colleagues make sure to include neurology residents in our clinic so they at least have exposure to it.

Dr Jones: That's a great point. If you never see it, it's hard to envision yourself in that practice. So, a little bit of a self-fulfilling prophecy. If you don't have neuro-ophthalmologists, it's hard to expose that practice to trainees.

Dr Van Stavern: And we're also trying; I mean, we make sure to include medical students, bring them to our meetings, present research to try to get them interested in this field at a very early stage.

Dr Jones: Dr Van Stavern, great discussion, very helpful. I want to thank you for joining us today. I want to thank you for not just a great podcast, but also just a wonderful article on ocular motor disorders, supranuclear and intranuclear. I learned a lot, and hopefully our listeners did too.

Dr Van Stavern: Well, thanks. I really appreciate doing this. And I love Continuum. I learn something new every time I get another issue.

Dr Jones: Well, thanks for reading it. And I'll tell you as the editor of Continuum, I learn a lot reading these articles. So, it's really a joy to get to read, up to the minute, cutting-edge clinical content for neurology. Again, we've been speaking with Dr Gregory Van Stavern, author of a fantastic article on intranuclear and supranuclear disorders of eye movements in Continuum's most recent issue on neuro-ophthalmology. Please check it out, and thank you to our listeners for joining today.

Dr Monteith: This is Dr Teshamae Monteith, Associate Editor of Continuum Audio. If you've enjoyed this episode, you'll love the journal, which is full of in-depth and clinically relevant information important for neurology practitioners. Use the link in the episode notes to learn more and subscribe. AAN members, you can get CME for listening to this interview by completing the evaluation at continpub.com/audioCME. Thank you for listening to Continuum Audio.