Emory University and Georgia Tech are leading an AI-driven approach to study freezing of gait (FOG), a common symptom of Parkinson’s disease that severely impacts mobility and independence. This innovative research utilizes cutting-edge motion capture technology alongside “ClassiFOG,” an advanced artificial intelligence (AI) tool developed to accurately detect and measure FOG episodes, enhancing treatment precision and clinical trial effectiveness.
Most symptoms of Parkinson’s disease stem from the loss of dopamine-producing cells, which leads to motor impairments such as tremor and rigidity. However, FOG does not always respond to dopamine medication, suggesting additional mechanisms are involved. Early studies indicate FOG could result from the interaction of several factors, including depletion of the neurotransmitter norepinephrine, neuroinflammation and amyloid protein accumulation.
The collaborative study explores whether exposure to specific sound and light frequencies, an emerging non-invasive technique to improve motor symptoms, can reduce amyloid buildup in the brain, To advance this research, scientists at the Emory Brain Health Center’s Motion Capture Lab, a state-of-the-art facility designed to track and analyze human movement, use ClassiFOG AI software that processes high-precision motion capture data to automatically detect and assess FOG episodes.
The Motion Capture Lab is equipped with 14 high-speed Vicon cameras, specialized force plates and infrared reflective markers, enabling researchers to collect highly detailed, real-time, three-dimensional movement data with sub-millimeter accuracy.
ClassiFOG analyzes this data, pinpointing when and how FOG occurs, measuring its severity and tracking its frequency over time. This combination of advanced motion capture technology and artificial intelligence has the potential to offer unprecedented insights into the mechanisms of FOG and its potential treatments.
“Our goal is not just to understand the biological basis of FOG but also to explore potential preventive measures. By combining AI-driven analysis with sound and light therapies, we hope to mitigate the factors contributing to FOG and improve mobility and quality of life for our patients,” says principal investigator Stewart Factor, DO, professor of neurology at Emory University School of Medicine and director of the Emory Movement Disorders Program.
The study is a six-month prospective, randomized sham-controlled trial, actively recruiting two groups of 12 participants each. It involves rigorous testing, including lumbar punctures to measure spinal fluid amyloid levels, to evaluate the impact of the therapies and detailed ClassiFOG analyses of freezing episodes.
“ClassiFOG’s AI-powered precision and consistency allow us to track the severity and frequency of FOG episodes with a level of detail that simply isn’t possible through human observation alone,” says co-principal investigator Lucas Mckay, PhD, associate professor of biomedical informatics and neurology at Emory University School of Medicine and co-director of the Emory Brain Health Center Motion Capture Lab. “This enables us to detect subtle changes that might otherwise go unnoticed, crucial for evaluating treatment approaches and improving clinical trial outcomes.”
Mckay’s research centers on movement disorders, particularly Parkinson's disease, and he manages one of the country’s largest full-body behavioral data repositories, bridging engineering and clinical research to advance neuroengineering and rehabilitation.
FOG presents a serious public health challenge, affecting up to 26% of people with early-stage Parkinson’s disease and more than 60% of patients after a decade of living with the disease. Characterized by sudden episodic pauses in movement, often described by patients as feeling their feet are "glued" to the floor, FOG significantly affects those with Parkinson’s disease, frequently leading to falls, social isolation and reduced quality of life.