Discovering the World of ‘Doge’: The Meme That Transformed into Money

The Daily Money: All about 'Doge.' Good morning! It’s Daniel de Visé with your Daily Money. So, what is "Doge"? Dogecoin, the meme cryptocurrency often associated with Elon Musk, soared in value after President-elect Donald Trump greenlit the tech billionaire's ideas for a new executive department with an evocative acronym. Trump announced that Musk, the world's richest person
HomeDiseaseCognitiveRevolutionary AI Video Test for Early Parkinson's Progression Tracking

Revolutionary AI Video Test for Early Parkinson’s Progression Tracking

An innovative technique utilizing artificial intelligence could transform the way Parkinson’s disease is managed.

A video analysis method created at the University of Florida, leveraging artificial intelligence, aims to enhance the monitoring of Parkinson’s disease progression in patients, potentially improving their care and quality of life.

The system, developed by Diego Guarin, Ph.D., an assistant professor of applied physiology and kinesiology at the UF College of Health and Human Performance, utilizes machine learning to assess video recordings of patients performing the finger-tapping test, a common diagnostic test for Parkinson’s disease that involves rapidly tapping the thumb and index finger ten times.

“By analyzing these videos, we can identify even subtle changes in hand movements that are indicative of Parkinson’s disease but may be challenging for clinicians to detect visually,” stated Guarin, affiliated with the Norman Fixel Institute for Neurological Diseases at UF Health. “This technology allows patients to record themselves taking the test, which is then analyzed by the software to provide insights to the clinician regarding the patient’s movements, aiding in decision-making.”

Parkinson’s disease is a neurological condition affecting movement and can manifest as slow movement, tremors, stiffness, and difficulties with balance and coordination. Symptoms typically progress gradually over time. Diagnosis of Parkinson’s disease does not rely on a specific lab or imaging test but rather on a series of assessments and activities performed by the patient to help clinicians evaluate the severity of the condition.

One of the most commonly used scales to monitor the progression of Parkinson’s disease is the Movement Disorder Society-Unified Parkinson’s Disease Rating Scale. Guarin highlighted that despite its reliability, the scale is limited to a 5-point system, making it challenging to capture subtle changes in progression and subject to individual interpretations.

The research team, including UF neurologists Joshua Wong, M.D.; Nicolaus McFarland, M.D., Ph.D.; and Adolfo Ramirez-Zamora, M.D., developed a more objective approach to quantify motor symptoms in Parkinson’s patients by leveraging machine learning algorithms to analyze videos, enabling the detection of nuanced changes in the disease over time.

“Through the use of a camera and computer, we can observe the same characteristics that clinicians aim to identify,” Guarin explained. “With the support of AI, this examination process becomes more efficient and less time-intensive for all parties involved.”

The automated system not only uncovered previously unnoticed movement details using precise camera data but also highlighted aspects such as the speed of finger movements and variations in movement properties with each tap.

“In Parkinson’s disease, we observed a delay in the movement when opening compared to healthy individuals,” Guarin noted. “This new information, nearly impossible to measure without video and computer analysis, showcases how technology can aid in characterizing the impact of Parkinson’s disease on movement and offer new markers to evaluate therapy effectiveness.”

To refine the system, initially designed by Guarin for analyzing facial features related to conditions other than Parkinson’s disease, the team utilized UF’s HiPerGator, one of the world’s most powerful AI supercomputers, for training some of its models.

“HiPerGator allowed us to design a machine learning model that simplifies video data into a movement score,” Guarin elaborated. “Through HiPerGator, we trained, tested, and enhanced various models with extensive video data, enabling these models to be run on smartphones now.”

Michael S. Okun, M.D., director of the Norman Fixel Institute and medical advisor for the Parkinson’s Foundation, emphasized that automated video-based assessments could significantly impact clinical trials and patient care.

“The finger-tapping test plays a pivotal role in diagnosing and measuring the progression of Parkinson’s disease,” Okun noted. “Currently, expert interpretation is required for evaluating the results, but what’s groundbreaking is Diego and three Parkinson’s neurologists at the Fixel Institute utilizing AI to standardize disease progression assessment.”

Aside from making this technology accessible to neurologists and healthcare providers, Guarin is collaborating with UFIT to develop a mobile app allowing individuals to monitor their disease progression at home over time.