Remote monitoring using commercially available activity trackers and computer vision provides a holistic, low-cost assessment of Parkinson’s disease symptoms

University Of Iowa, Iowa City IA

Investigators

Abstract

Although there has been considerable interest in using remote monitoring in diseases like Parkinson’s disease, Alzheimer’s disease, and other Alzheimer’s disease-related dementias, there has been essentially no change in clinical care. Yet, remote monitoring could radically reduce barriers to care, especially in rural areas – 80% of rural areas medically underserved. The long-term goal is to use remote monitoring to improve the lives of people with neurodegenerative causes of dementia. The overall objectives of this application are to identify the value of wearable devices, automated fine-motor and speech assessments in detecting and measuring cognitive and non-cognitive symptoms of Parkinson’s disease. The central hypothesis is that low-cost wearable devices combined with computer vision and speech analysis can provide useful, holistic assessment of Parkinson’s disease. The rationale for this project is that remote monitoring can increase the effectiveness of expert care, alleviating barriers of access and distance in underserved, rural communities. The hypothesis will be tested with two specific aims: 1) Identify the value of low-cost, holistic assessment for predicting Parkinson’s disease diagnosis and cognitive impairment and 2) Identify the value of low-cost, holistic assessment for predicting one-year cognitive, motor progression in Parkinson’s disease. In the first aim, people with possible Parkinson’s disease will be provided an activity tracker and conduct home-based motor and speech assessments with the goal of building a classifier to predict the ultimate diagnosis (Parkinson’s disease versus not Parkinson’s disease) and cognitive status at baseline (normal cognitive, mild cognitive impairment, dementia). The second aim will build a cohort of people recently diagnosed with Parkinson’s disease and conduct a 4-week assessment using an activity tracking watch and home-based motor and speech assessments. The objective will be to improve prediction of cognitive and motor symptoms at 1 year. The proposed research is innovative because it focuses on using data collection using low-cost, commercially available devices to construct a holistic measure of both the motor and cognitive manifestations of Parkinson’s disease. The proposed research is significant because it is expected to provide a low-cost framework useful for non-expert providers to screen for and evaluate Parkinson’s disease and to provide a more accurate prognosis for people newly diagnosed. As a career development grant, this proposal is ideal: it builds on the applicant’s past skills in health economics by adding knowledge of neuroscience, neurodegenerative diseases and dementia, human subjects, and clinical research – all relatively novel areas for the applicant. The applicant will accomplish these career development goals through both conducting the proposed research and didactic instruction in clinical trials, neuroscience, and neuropathology. The applicant’s division and institution are dedicated to providing the needed support and resources for success and both have outstanding track records for transforming ESIs into successful, independent, NIH-funded researchers.