![Ogden_data researcher.jpg](/sites/default/files/styles/featured_xl/public/images/featured/Ogden_data%20researcher.jpg?h=51a72048&itok=xzl6UpkE)
An AI competition, hosted by The Michael J. Fox Foundation for Parkinson’s Research (MJFF) and the Kaggle data science platform, generated nearly 25,000 new machine learning solutions for detecting freezing of gait (FOG) in people with Parkinson’s disease (PD). These findings can help inform and improve treatments for a symptom that people with Parkinson’s routinely report as bothersome. People with PD who experience freezing of gait often say they feel as if their feet are glued to the ground. But despite how disruptive the symptom can be, it remains difficult to measure. That’s why MJFF sponsored an AI contest to automatically and unobtrusively measure FOG incidents using accelerometer data collected from a wearable sensor.
Measuring these incidents allows researchers to pursue a better understanding of their causes and their different manifestations while also empowering clinical trials to measure potential improvements.
Results of the competition were published in Nature Communications in June 2024. There, the authors note that 1,379 teams from 83 countries submitted 24,862 solutions, with the winners proving precise and accurate, confirmed by having trained clinicians review video footage of tasks designed to provoke freezing of gait.
FOG is inconsistent and unpredictable, so there are benefits to a more expansive monitoring approach, such as wearables that measure movement 24/7. Using artificial intelligence to parse datasets on FOG could dramatically improve our ability to characterize it and identify patterns.
In fact, the authors note they did successfully identify previously unobserved patterns in FOG occurrences using the top solutions.
The lead authors, researchers at the Tel Aviv Sourasky Medical Center, also produced a “Behind the Paper” blog, where they provide additional color to the competition’s impact, including:
[T]he contest rapidly improved our ability to objectively quantify FOG, with the results comparable to those of experts. Moreover, the contest results pave the way for 24/7 monitoring of feedback, a possibility that promises to shed new light on a mysterious phenomenon and, hopefully in the long run, inform and improve treatments for a symptom that can be extremely bothersome and debilitating.
MJFF has targeted freezing of gait specifically because it can be so disruptive for people living with Parkinson’s, including funding multiple projects through an open application in 2023.