Using Wearable Sensors to Identify Movement Biomarkers of Health and Disease
By PhD student Dheepak Arumukhom Revi and Assistant Professor Louis N. Awad
The emerging fields of telehealth and digital therapeutics are primed to change the face of healthcare. They are also well-suited to leverage wearable sensors that can monitor movement in the everyday world. Indeed, we have recently seen a rapid expansion of everyday technologies incorporating wearable sensors for movement monitoring. For example, the latest release of the IOS software found in iPhones and other products includes algorithms that measure how we move about our world. While these tools are a great first step in making movement data accessible to the general community, the algorithms that power these devices lack the specificity required by clinicians who need to make treatment decisions for individual patients.
The movement metrics that existing commercial systems can track are clinically useful, but the past 15 years of scientific research have shown us that a deeper understanding of the forces we create during movement can provide key insights into health and disease. For example, for every step we take, we need to control the propulsion and braking forces that accelerate and decelerate our body as we walk about our world. Propulsion – how hard a person pushes off the ground in order to move – has emerged as a key concept in the rehabilitation and recovery of walking after neurological injuries, such as stroke. Despite this, only a handful of clinics in the world have the equipment and expertise required to routinely evaluate propulsion metrics when assessing and treating neurological gait impairments.
One line of research in the Neuromotor Recovery Laboratory aims to use wearable sensors to identify movement-based biomarkers of neurological health and disease. Wearable sensors have the potential to collect data continuously in the background of everyday walking activities. Leveraging the low cost and high accessibility of wearable sensors, our team focused on designing and testing new approaches to using wearable sensors to estimate propulsion function, together with common clinical metrics like walking speed, in people with impaired gait. It’s working, too: estimates made by these wearable sensors strongly approximate the measurements made by expensive and sophisticated lab-based 3D-motion analysis systems. We have validated our algorithms on 32 individuals so far, and more than half of them have substantial gait impairments due to stroke.
The movement data that our algorithms can capture are not only useful for neurorehabilitation but can also advance new healthcare paradigms for aging adults and people with chronic health conditions. For example, preventative interventions can be deployed in response to the detection of gait changes that may signal a change in health status or may precede catastrophic events, such as a fall. However, in its current state, our wearable sensor system is not ready for clinical use. Our team is uniquely positioned to tackle this problem. With expertise in gait biomechanics, engineering, and clinical practice, our research benefits from the different perspectives brought by the different backgrounds of individual team members. Our interdisciplinary team allows us to quickly make relevant decisions, drawing on research data and domain expertise.
Movement is considered a window into health and disease. Our development of new movement algorithms and sensing systems for people with neurological impairments has the potential to advance the emerging fields of digital therapeutics and tele-rehabilitation. If our work is successful, we can create a future where clinicians use movement data collected from wearable sensors to prescribe more effective treatments that are uniquely tailored to each patient’s needs. This process can also be implemented as part of telehealth treatment paradigms – greatly increasing treatment efficacy and efficiency, as well as the accessibility of quality care in rural areas across the country and the world.
About the Authors
Dheepak Arumukhom Revi is a PhD student in Mechanical Engineering at Boston University. His interests are in translational technologies for people living with walking-related disability. Before pursuing his PhD, he worked as a Research Fellow at the Paulson School of Engineering and Applied Sciences at Harvard University in developing and studying exosuit technologies.
Dr. Louis Awad is the founding director of Boston University’s Neuromotor Recovery Laboratory—a cross-disciplinary research group that works to develop, study, and translate novel rehabilitation therapies and technologies for people living with neuromotor conditions resulting in walking-related disability. He is an Assistant Professor in the Department of Physical Therapy and Athletic Training at Boston University, with affiliations in the Department of Mechanical Engineering and the Center for Neurophotonics. He is also a Research Associate in the Department of Physical Medicine and Rehabilitation at Harvard Medical School and in the Paulson School of Engineering and Applied Sciences at Harvard University.
This article originally appeared on the Institute for Health System Innovation & Policy blog.