A Multi-Modal Wearable Sensor for Early Detection of Cognitive Decline and Remote Monitoring of Cognitive-Motor Decline Over Time

ABSTRACT Recent demographic changes have led to the emergence of the so-called “dementia epidemic”. Dementia causes great stress to medical, social, and informal care and is currently affecting approximately 55 million persons worldwide. This number is projected to increase to 75.6 million by 2030 and 135.5 million by 2050. Alzheimer’s disease (AD) is a degenerative brain disease and the most common form of dementia. During the last 15 years, more than 400 different AD clinical trials have been conducted for managing AD symptoms, with a combined failure rate of 99.6%. A lack of robust tools for measurement and monitoring of cognitive ability and function is considered a key contributor to the very high failure rate. Additionally, early diagnosis of cognitive decline and measuring its subtle progression over time enables early intervention to prevent further cognitive and functional decline. Finally, access to conventional cognitive screening tools is often limited for those with high socioeconomic deprivation and individuals living in remote areas. Thus, there is an urgent need for a robust solution to identify individuals in the earliest stages of cognitive decline and measure digital biomarkers associated with subtle changes in cognitive-motor performance over time. In this Fast-Track SBIR study, we propose to develop PAMSys+, a multi-modal wearable sensor for remote monitoring of cognitive-motor decline based on measuring dual-task (DT) physical activities (e.g., walking while talking) and digital speech biomarkers during activities of daily living. PAMSys+ includes 1) bi-directional microphones (located inwards to and outward from the chest) with synchronized audio signal data collection, and 2) an accelerometer to collect chest motion data. It combines the capabilities of audio and motion monitoring to create a first-in-kind wearable sensor for remote monitoring of DT physical activities using a patented method developed by our team (U.S. Patent App. No. 17/807,104). PAMSys+ also analyzes audio signals to measure digital speech biomarkers that are shown to be correlated with cognitive function (e.g., speech duration and rate). In Phase I, we will develop a prototype of PAMSys+ and demonstrate its proof of concept in 50 older adults (age 60+) including individuals with and without cognitive impairments (25 per group). Additionally, we will assess the acceptability of PAMSys+ by surveying the participants and a focused expert panel (n=10) using the technology acceptance model (TAM). In Phase II, we will complete the development of PAMSys+ and will study its validity for early detection of AD and also remote monitoring of cognitive-motor decline over time by recruiting 100 adults (age 60+) including 50 individuals with amnestic mild cognitive impairment (aMCI) or mild to moderate AD and 50 age-matched cognitive intact individuals. BioSensics is a 2020 Tibbetts Award winner with a proven track record of rapidly transitioning applied governmental research funding into real commercial results. The proposed project is well aligned with the long-term objectives of BioSensics and is a critical component of our broader plans for development of novel technologies to improve the lives of older adults.