Search for: All records

Heart rate, a commonly accessible health data from most wearables, carries rich information of a person’s well-being, yet remains of limited deep health applications, due to the lack of groundtruth of health events and their impact on heart rate patterns. Specifically, standard health analytics usually are designed based on well-modeled health conditions thus known data patterns and rich training data. To bridge the gap, we propose HeartInsightify, an exploratory framework that facilitates the process of deriving health-relevant measurable indicators from longitudinal heart rate data, without any of the above knowledge. HeartInsightify focuses on comparative and qualitative study, using model-free statistical methods such as conformal prediction, to study similarities, perform clustering and detect outliers, and build multi-resolutional data summaries, allowing human experts to efficiently examine and verify their health relevance. We conduct extensive experiments to evaluate HeartInsightify using individuals’ free-living heart rate data collected through Fitbit over 6 years. We illustrate the process of analyzing heart rate data for its health relevance and demonstrate the effectiveness of HeartInsightify. We envision that HeartInsightify lays the groundwork for personalized health analytics with continuous monitoring data from wearables. 

Abstract The increasing prevalence of wearable devices enables low-cost, long-term collection of health relevant data such as heart rate, exercise, and sleep signals. Currently these data are used to monitor short term changes with limited interpretation of their relevance to health. These data provide an untapped resource to monitor daily and long-term activity patterns. Changes and trends identified from such data can provide insights and guidance to the management of many chronic conditions that change over time. In this study we conducted a machine learning based analysis of longitudinal heart rate data collected over multiple years from Fitbit devices. We built a multi-resolutional pipeline for time series analysis, using model-free clustering methods inspired by statistical conformal prediction framework. With this method, we were able to detect health relevant events, their interesting patterns (e.g., daily routines, seasonal differences, and anomalies), and correlations to acute and chronic changes in health conditions. We present the results, lessons, and insights learned, and how to address the challenge of lack of labels. The study confirms the value of long-term heart rate data for health monitoring and surveillance, as complementary to extensive yet intermittent examinations by health care providers. 

Abstract Over 80% of older adults want to live independently in their own homes and communities, maintaining quality of life, autonomy, and dignity as they age. We are using community engaged research methods to aid in developing in-home cost-conscious remote sensing technologies to support older adults age in place. To understand their needs, we engaged older adults in discussions on home-based sensing technologies. We used visuals and demonstrations to facilitate discussions, showing participants our sensor prototypes and a vignette describing the challenges an older adult and the family face managing a chronic condition. Participants voiced their interest in monitoring for select health conditions and situations when either they or the person(s) they care for are home alone. Discussants raised concerns about personal security/privacy, loss of independence, ethics of data collection and sharing, and being overwhelmed by collected data. Discussions have provided valuable feedback to help us develop a sensor system that is flexible enough to accommodate individuals in different life stages and comfort levels, with different home environments, levels of expendable income, and support structures. As a result, we have developed a system that uses nonvisual, non-wearable sensing that measures respiration and heart rates, and indoor location tracking to monitor the health and wellbeing of users. During this session, we will provide detailed results from our community discussions, and discuss the continuing role for community engagement as we move forward with sensor development and testing. 

Abstract As Americans live longer, a dynamic opportunity has arisen to provide enhanced resources to sustain their well-being. Cost-conscious, convenient in-home sensing will assist with chronic disease management, and become part of a long-term plan to support our aging population and shrinking healthcare workforce. The purpose of this study was to obtain input from older adults about (i) their comfort level and willingness to adopt different sensor technologies, and (ii) opinions on data sharing, security, and privacy to guide our sensor development. Over 4 different survey timeframes (2018-2022), adults aged 60 and older (N=112) completed our survey either in-person (n=77) or via a REDCap online survey (n=35) (53% female; 30% age >80; 78% college graduates; 19% living alone). Though there were significant differences (p< 0.05) in demographics based upon recruitment source, no differences in attitudes towards sensor use were found by age, gender, education, or marital status. Opinions and preferences for sensor type/number/install location, and data sharing preferences significantly differed (p< 0.05) by home living arrangements (independent, 55+ or continuous care communities). Similar to national surveys, changes in technology use were observed pre- versus post COVID. Respondents living in 55+ and continuous-care housing were more comfortable with having sensors installed in their homes than those in community dwelling independent housing. This study highlights the need to include end users throughout the lifecycle of product development and provides insights into preferences by older adults for sensor use and data sharing.