Can Computer Models Help Diagnose Alzheimer’s Disease Earlier?
Professor Johannes Weickenmeier and his research team aid in the fight against neurodegenerative diseases
Alzheimer’s disease can be diagnosed with complete certainty only after a patient dies. Then it becomes possible to examine the brain itself for the telltale plaques and neurofibrillary tangles without causing more harm. But a new application of an old science may provide a window into the sealed system of the brain.
Reading over material from a talk he’d attended, Stevens mechanical engineering professor Johannes Weickenmeier was struck by images in two different papers. One, a side view of the brain, showed how the toxic proteins associated with Alzheimer’s disease were distributed at different stages of the disease’s progression. The other quantified how much toxic protein was in the brain at any given time.
“I can reproduce that result with a computer model,” Weickenmeier thought.
A mechanical engineer, Weickenmeier had been modeling muscle and skin to simulate facial expressions at the time as a post-doc at Stanford University. But the principles he worked with as an engineer would also apply to the brain and allow him to reproduce how toxic proteins spread through the pathways available for them to travel.
“Mechanics influences biology and biology influences mechanics, if you think of any living material,” he says.
The idea sparked by those images forms the core of Weickenmeier’s current work, exploring the mechanical principles of the central nervous system. A member of the Center for Neuromechanics at Stevens, Weickenmeier has been awarded a grant from the National Institutes of Health and a Frank Semcer, Sr. ’65 Fellowship, which provides resources to employ doctoral student assistants. His Weickenmeier Lab at Stevens is devoted to developing experimental and computational tools to study biological systems like the brain.
The brain has been studied primarily from a neuroscience perspective — for instance, how memory works and how consciousness forms. Biochemistry, or how the brain regulates itself via chemicals, is another broad field of study. “The thing that has been under-studied is the role of mechanics in the brain,” Weickenmeier says. “A lot of neurodegenerative processes that are biochemical in nature ultimately manifest as organ-level shape changes.”
An aging brain will shrink even in a healthy adult. As a person grows older, brain tissue volume decreases and is replaced by fluid, the cortex thins, the connections between neurons begin to break down and the spaces between the folds of the brain increase.
Neurodegenerative diseases speed up these changes, but each disease also follows a unique pattern both in the time it progresses and the physical changes it causes. A brain affected by Alzheimer’s disease will look different than a brain affected by Parkinson’s disease, and that will look different than a brain affected by amyotrophic lateral sclerosis (ALS). Weickenmeier and his team are developing computer models that match the progression of those diseases and others to study how the brain’s shape changes and deforms.
The models would be helpful in multiple ways. They could help diagnose neurodegenerative diseases earlier and identify exactly what type of dementia a person is experiencing, so the most effective therapies could be applied.
Knowing whether a disease will progress quickly or slowly and how it will progress would help a patient plan their resources and time, whether it’s arranging to live closer to family, obtaining equipment for future motor issues or taking a much-desired trip.
On the research side, the models would provide tools to measure the impact of treatment and efficacy of drugs. “If you give someone a medication, you want to quantify whether you’re actually changing the disease progression,” Weickenmeier says.
A model that captures changes in the brain could let researchers add in a drug to confirm that the medication slows down the deterioration, and that’s the next big step, Weickenmeier says. “Diagnosis was what I was hoping for earlier, but in the long run we want to be able to monitor disease-changing treatments.”
Even more speculatively, if effective models can show a disease progressing, they could also be used to trace that progression in reverse, potentially showing when a brain tips from healthy to diseased — a key piece of information when so many neurodegenerative diseases progress without physical warning signs over many years.
“You have processes that happen at the smallest scales and over time they manifest as an entire brain shrinking, and the orders of magnitude that are between these processes are what I find fascinating,” Weickenmeier said. “Can we explain, can we close that humongous gap between those two scales with clever modeling? That’s the potential of our approach.”
For now, the models are in their early stages but progressing. After initially showing that a computer model could reproduce the patterns of an ailing Alzheimer’s disease-affected brain, Weickenmeier’s team is narrowing the models down to smaller and more specific problems, like differentiating between a naturally aging brain and one affected by disease.
The team is working on understanding the relationship between the brain’s structure and classic engineering properties like stiffness and viscosity, and how they’re affected by anatomy and the microscopic structures of the brain. Finally, the team is looking at the two proteins, amyloid beta and tau, that are mostly associated with Alzheimer’s and other neurodegenerative diseases, and how they interact with each other.
Weickenmeier is excited about the ways in which computer modeling could supplement existing imaging and diagnostic techniques for identifying and treating neurodegenerative diseases.
“There is a growing community of people who are recognizing the value of mechanical engineering as one of the puzzle pieces in understanding degenerative disease,” he says. “We’re providing a new way to analyze existing data that shines additional light on the disease progression, on things that are changing, and that will also change how we think about diseases and how to diagnose them and monitor their progression.”