To Confirm Efficacy of Activity-Based Parkinson’s Therapy
By Frank A. Fornari, PhD, Founder and CEO, BioMech Health
LinkedIn: Frank Fornari
LinkedIn: BioMech
AI-driven advances in clinical motion analysis helps grow the body of evidence on the efficacy of activity-based therapies for Parkinson’s disease.
While physical activity, physical therapy, and structured exercise have long been considered effective treatments for Parkinson’s disease, their efficacy has been difficult to scientifically establish due to the inability to measure changes in key functional motion indicators like gait and overall balance – which has, in turn, hampered patients’ motivation to adhere to these therapies as part of their treatment. Not to mention the relationship between physical activity and current pharmacological treatment of Parkinson’s disease, which remains a significant question mark without data to support or refute medication efficacy.
However, recent advances in clinical motion analytics, artificial intelligence (AI) and wearable medical devices are finally filling in the evidentiary blanks, with several studies suggesting the relationship between physical activity and functional performance among patients with Parkinson’s disease is significant.
An Oxford University study, Identification of motor progression in Parkinson’s disease using wearable sensors and machine learning, published in Nature Medicine, established for the first time that leveraging specially-trained machine learning (ML) algorithms to analyze data gathered using sensors worn by patients could accurately measure the progression of Parkinson’s. The technique enabled researchers to determine that the progression of motor symptoms could be detected in as little as 15 months and that wearable sensors and ML track progression better than the conventionally used clinical rating scales.
New Research Supports Boxing for Parkinson’s Impact
The Oxford study is part of a growing body of evidence showcasing how the combination of wearable sensors and AI technology is advancing the treatment of Parkinson’s and other motion disorders. For Brian Soucy, 66, an engineer who was diagnosed with the degenerative disorder in 2020, the ability to quantify the impact of his physical activity on Parkinson’s progression comes as welcome news.
Soucy participates in a Boxing for Parkinson’s program at the Center for Movement Challenges in Atlanta, which uses non-contact boxing to slow the progression and ease the symptoms of the disease. “I really noticed after the first few sessions that my attitude improved; just coming here and feeling like you’re doing something about it is a good feeling,” he says. But “it would be nice to have some data to show for it.”
To that end, Soucy and 52 others took part in a study conducted by the Center and BioMech that utilized objective clinical motion analytics to better understand the link between physical activity and exercise in Parkinson’s patients over a 2-month period. Testing was conducted using BioMech Lab sensors and leveraged AI-driven clinical devices and sensor technology to assess participants’ quantitative functional motion (gait and balance) at the beginning and end of the two-month program.
The study found a significant and measurable improvement in functional motion (gait and overall balance) among participants. More than half (56%) improved by 24.5% overall, while 66% saw an improvement in gait performance (stride length, gait speed, impact symmetry, pelvic neutral deviation, single support time symmetry and normative support time and toe-off symmetry) of nearly 9%. It also demonstrated that analytics can be used to assess the impact of changes to pharmacotherapeutic regimes and/or the frequency and dosage of structured exercise even in those participants who recorded a negative change in functional motion.
AI/ML-Driven Advances in Treatment
In addition to adding critical metrics to the battle against Parkinson’s disease, the Boxing for Parkinson’s research also demonstrated that gait and balance analyses can provide additional objective and actionable metrics with higher granularity and accuracy, thereby arming clinicians with critical data to augment traditional subjective pre- and post-treatment assessment protocols.
Research like the Boxing for Parkinson’s study that uses BioMech’s AI/ML platform, Coretex.ai, demonstrates the powerful potential of using motion as a functional biomarker and endpoint for many treatments. The ability to leverage advanced sensors and wireless, noninvasive, and self-calibrating technology to instantly capture and stream three-dimensional motion data means clinicians can now continuously monitor the progress of prescribed therapies and rapidly intervene when adjustments are necessary. As such, it reinforces the benefits of adhering to prescribed physical therapy and other exercises for Soucy and other patients with Parkinson’s and other motion disorders – and provides hope for slowing the disease’s progression.
Soucy, who has been taking Boxing for Parkinson’s classes up to four times a week for more than a year, says it’s important to him to have analytics to support assumptions that the activity is doing more than just improving his attitude.
“It’s very interesting, as an engineer, to have data where you can actually measure your balance before and after a workout,” he says.