I-Corps: Translation Potential of an Intelligent Foot Motion Recognition Technology to Prevent Freezing of Gait and Falls

This I-Corps project focuses on developing a commanding insole, which utilizes intelligent foot motion recognition technology collected by onboard sensors to prevent freezing of gait and falls. Freezing of gait is the most perplexing and disabling symptom of Parkinson's disease and often leads to falls. As the fastest-growing neurodegenerative disease, Parkinson's disease affects nearly 1 million Americans and 10 million people worldwide. Eighty-five percent of patients suffer from freezing of gait, which increases fall risks and reduces the quality of life and the likelihood of independent living. Commanding insoles can be worn in various home settings to enable processing of large amounts of data in real time. The solution can be potentially expanded to include orthotics, diabetic pressure ulcer prevention, and sports performance monitoring.

This I-Corps project utilizes experiential learning coupled with a first-hand investigation of the industry ecosystem to assess the translation potential of the technology. This solution is based on developing three innovative components: an insole hardware; an onboard, tiny, machine learning algorithm; and a mobile app. The tiny machine learning algorithm and app visualization techniques including, but not limited to, heat maps and bar plots, make more accurate and reliable decisions. Existing monitoring devices have restricted interconnectivity and comfort and cannot identify new or unseen gait freezing events in living conditions beyond their original training context. The training observations constantly gathered by the same patient in the place of inference keeps the classifier accurate, even with the limited size of the dataset. The automatic score scheme correlates highly ranked features with a personalized rehabilitation plan through toe/heel-tapping exercises to mitigate future freezing of gait or fall risks. The new understanding of multi-model sensing and tiny machine learning will advance home monitoring and healthcare. The mobile application manages real-time performance for end-users transparently and in a straightforward manner. Parkinson's patients have validated the ability to monitor disease progression and evaluate fall risk.

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

Award Number: 2521315
Principal Investigator: Ya Wang
Funds Obligated: $50,000
State: TX