NSF Award
1549761
The broader impact/commercial potential of this Small Business Technology Transfer (STTR) Phase I project is to mitigate fall risk among Parkinson's patients and the elderly, which will potentially save families and patients $34 billion annually in fall-related injuries and rehabilitation. Falling is common among individuals age 65 and over, one in three people fall at least once in a calendar year and Parkinson's patients are twice as likely to fall as their counterparts. It is expected that the proposed technology will provide a holistic view of a patient's health. The real-time data detected by the integrated sensors offers information that consumers and caretakers can use to plan health strategies at home and with their physicians. The proposed technology is expected to fit seamlessly into the lives of consumers so that they benefit from the power of technology without the difficulty utilizing it. Several fall detection systems are currently on the market; however, the two main issues with these systems are compliance and detection. No existing fall prevention devices determine when a patient's risk of falling is elevated. The proposed technology has the potential to enter the personal emergency response systems (PERS) market, which is estimated to grow to $1.86 billion by 2017.<br/><br/>The proposed project addresses consumers' need to monitor and be proactive about their chronic health symptoms, particularly as they relate to falls. The goal of the proposed project is to develop and commercialize a device that predicts when a fall is likely to occur and to provide actionable feedback. We will use machine learning to achieve the proposed research objective by analyzing data collected from sensors embedded in a back brace to develop algorithms that will predict symptom onset and alert Parkinson's patients and caretakers to increased fall risk. The algorithms developed to analyze the data collected from the sensors on the proposed technology are the intellectual merit of this project. The machine learning algorithms will be used to find a correlation between sensor readings and symptoms the individual is experiencing. The anticipated results are that the correlations found in the data will lead to a better understanding of the individual's symptoms, disease progression, and which sensor readings indicate increased fall risk. Understanding the mechanisms that cause individuals with Parkinson's to fall will develop better alert systems and improve fall prevention. The results from data collection and analysis could also lead to better detection of early warning signs of Parkinson's progression.
