In 1998, the Iraq Liberation Act, signed into law by President Bill Clinton, stated that “it should be the policy of the United States to seek to remove the Saddam Hussein regime from power in Iraq.” In the following years, the U.S. and other coalition forces would overwhelm the Iraqi army and end Hussein’s dictatorial rule.
At the time, Dr. J. Lucas McKay was an undergraduate student at Brown University studying electrical engineering: specifically, he was working with analog circuit design, which he described as an “old-fashioned” engineering discipline. McKay’s career pathways were mostly limited to making parts of weapons, and he wasn’t compelled to pursue such endeavors.
“I didn’t really want to spend the next 20 years of my life working on technologies that could be applied to weapons,” McKay said. “It was kind of tumultuous times, sorta like now, and that was a biggie.”
McKay pursued engineering largely because he was good at math, and because it seemed like he would learn a lot as an engineer. However, he admitted that, when he went into junior and senior year, he realized that he hadn’t chosen his degree in a very examined way.
At the time, data science wasn’t a job, and Google was just starting out, McKay said. The iPhone wouldn’t be invented for another few years, there was no Youtube, and it was generally a very different environment in terms of technology and the internet.
“It was a lot harder to figure out what you wanted to do by surveying everything that people did,” McKay said. “Students now have a much better idea, particularly through forums like you’re doing right now, where they can figure out what is going on.”
McKay worked on a project called BrainGate, a program focused on developing brain-computer interface technologies to restore the independence of people with neurologic disease. The project was designed to create a computer software to directly interface with the brains of people with spinal cord injuries, such as Pitt-Hopkins syndrome.
Combining clinical research and computer engineering, the project allowed McKay to apply his analog circuit design experience to medicine, and its potential to help people compelled him to get involved.
“There are definitely easier ways to apply technology to help people,” McKay said. “If you figure out how to filter water, you’re going to help more people. But in terms of healthcare applications of electrical engineering in healthcare, there was a pretty clear application.”
After obtaining his Bachelor of Science degree in Electrical Engineering, McKay completed a Masters of Science in Electrical Engineering. During his Masters, he helped build a part of a computer chip that would go in an animal model of motor cortex recordings, designed for brain-computer interface in animals.
Most entry-level engineering jobs weren’t that interesting and that most were very limiting in terms of career advancement, so McKay decided to get his Ph.D. at Georgia Tech University — and, simply because he wanted to work on more science.
A focus on Parkinson’s disease
Over the course of his academic and professional career, McKay said that he has become more and more involved with the clinical side of research. As an engineer motivated by an innate urge to fix problems, he saw a potential to unravel the puzzling aspects of medicine and disease treatment.
McKay viewed medicine as a problem solving exercise, unlike most scientists and medical professionals, who were curious about underlying biological and physical phenomena.
“If there was a pill you could take to stop Parkinsons, I would just tell people to eat the pill,” McKay added. “I couldn’t be bothered.”
McKay’s Ph.D. thesis involved neuromechanical balance control in animals, which led him to work with Parkinson’s disease (PD), a disorder associated with bad balance: he studied animals in order to eventually figure out how it could be applied to people.
While McKay was at Georgia Tech, his wife was completing her Masters of Public Health at Emory University and would later go on to work at the Centers for Disease Control. The way that she and her colleagues interacted with disease was a lot like how he always thought about them — as problems to be solved.
McKay said that he became increasingly exposed to epidemiology and interventional epidemiology, and he eventually completed a short postdoc at Emory, where he started to work with PD patients for the first time. Afterward, he joined the research faculty as an Assistant Professor at Emory and Georgia Tech’s joint Department of Biomedical Engineering.
“I was the first person to enroll someone with PD in our building over at Georgia Tech — trust me, it was a lot of light work to get that to work,” McKay laughed. “They didn’t have anywhere to park, and stuff like that. Doing clinical work at Emory is a breeze compared to Georgia Tech.”
A year later, supported by the Georgia Clinical and Translational Science Alliance as a KL2 scholar, he completed a Master’s degree in Clinical Research at Emory.
“During that process, I initially thought it was going to be a waste of time, because I’d already taken so many classes,” McKay said. “But it was definitely the best thing I ever did.”
In particular, Mitch Klein’s epidemiology class was a game changer for McKay. The class taught him how to use epidemiological concepts to understand clinical problems; once he started to approach problems with that toolset, things began to “make more sense and fall into place.”
McKay also said that, as a Ph.D. student working in an insular, specialized lab, he gained a set of highly specific techniques that “may or may not be generalizable to other fields.” In contrast, clinical researchers largely use the same skillsets, which allows them to understand a variety of different diseases.
“I figured that in order to approach these problems, I’d need to learn to be a doctor,” McKay said. “But that’s not the case at all — there’s plenty of doctors who would like to do research, but do not have a way in. If you can work with them, it’s a win-win.”
Movement disorders and Basal Ganglia Pathophysiology
After completing his KL2 fellowship, he joined the Emory National Primate Research Center (ENPRC) as a National Institute of Health (NIH)-supported K25 scholar. The K25 award was designed to get people with technical expertise to work directly with people working on disease mechanisms or clinical care.
ENPRC Associate Director Dr. Thomas Wichmann was McKay’s basic sciences mentor. Wichmann was a primate physiologist, and he helped create the most established model of how the basal ganglia goes into dysfunction in PD, leading to bradykinesia.
Emory Comprehensive Parkinson’s Disease Center Director Dr. Stewart Factor was McKay’s clinical director. They worked closely on gait – a person’s manner of walking – and balance problems in PD; specifically, freezing of gait, an abnormal gait pattern that can accompany PD.
According to McKay, there is still much debate as to how freezing of gait responds to front line therapies, such as combination carbidopa-levodopa drug treatment, partially because it’s very difficult to measure objectively. One of his recent projects involved applying modern machine learning tactics to quantify motor symptoms that are usually done by expert observation.
“Expert observation is tough to learn how to do, and our fellows spend two to three years learning how to rate things based on observation,” McKay said. “That’s going to reduce access to care.”
McKay aimed to automate aspects of rating diagnostic cases by supporting clinicians with computer software analysis. Rather than leaving diagnosis completely up to a computer, it would serve as an auxiliary instrument, much like how predictive text assists with Google searches.
“I’d like the computer to look at the patients and say, ‘Oh, that’s probably PD,’” McKay said. “And then the clinician looks at the patient and says, ‘You’re probably right, computer,’ or ‘You’re completely wrong, we need to do a second level assessment.’”
Though the role of AI in medicine is still up in the air, McKay said that most clinicians would probably like it in the context of decision support and predictive analytics. However, the computer wouldn’t make diagnostics on its own.
“You’re not going to trust Google to diagnose your mom with something — nobody actually thinks that’s going to be the future,” McKay added.
Impacts and goals
McKay’s future clinical goals involve falls, which are a significant complication involved in PD patients that greatly impacts quality of life. He hopes to change his center so that, if a patient comes for treatment, they’ll be half as likely to fall in the next 12 months compared to other centers.
“You need to have a lot of both surveillance and clinical practice guidelines,” McKay said. “So that’s a really long term project.”
He also hopes to extend pre-existing clinical research methods, specifically by urging people to use big data and epidemiology methods for discovery science. Instead of considering a sample of 12 patients, he wants researchers to scale up and about how to study every single PD patient in the United States.
In terms of the broader community, McKay said that he “rubs off” on his colleagues that there are established clinical research methodologies that researchers should use, even in early stage observational studies.
“A lot of people I work with come from animal work, where you get 100 Wistar rats and they’re all the same,” McKay said. “But patients are not like that — each one of them is absolutely special.”
