A team led by\u00a0University of Massachusetts Amherst\u00a0researchers\u00a0has\u00a0developed\u00a0a wearable wrist device\u00a0powered by a\u00a0machine-learning algorithm that\u00a0can continually track changes in\u00a0arm\u00a0movement impairment\u00a0caused by strokes.\u00a0Monitoring\u00a0changes\u00a0throughout\u00a0the\u00a0entire\u00a0rehabilitation process\u00a0will allow\u00a0clinicians\u00a0to\u00a0make real-time adjustments to therapy programs for tailored interventions and personalized care\u00a0instead of the one-size-fits-all current standard.\u00a0<\/p>\n
“We are the first group to actually show that, using wearable data, we can extract information about patients’ motor severity, which clinicians can actually use to determine whether their intervention is effective or not,” says\u00a0Sunghoon Ivan Lee, associate professor in the Manning College of Information and Computer Sciences at UMass Amherst\u00a0and corresponding author on the paper describing this new technology. The research was completed alongside colleagues from\u00a0Washington University in St. Louis, Shirley Ryan AbilityLab and Harvard Medical School\/Mass General Brigham.\u00a0<\/p>\n
Annually, more than 795,000\u00a0Americans\u00a0experience\u00a0strokes,\u00a0with\u00a0upper-limb mobility issues affecting\u00a0up to\u00a077% of patients\u00a0immediately\u00a0following\u00a0the event. About\u00a040% of patients continue to have chronic issues,\u00a0posing a major limitation\u00a0to\u00a0independent living.\u00a0<\/p>\n
While physical therapy\u00a0improves these mobility issues,\u00a0it is not without\u00a0its limitations. Currently, progress is measured by a clinician’s observational\u00a0assessment, which takes about 30 minutes.\u00a0Because this is a time-consuming process, assessment\u00a0typically\u00a0occurs\u00a0only\u00a0pre- and post-rehabilitation.\u00a0<\/p>\n
“That means during that therapy process, neither\u00a0the\u00a0patient nor the therapist has\u00a0a clear idea of how patients are responding to the treatments that they’re receiving,” says Lee.\u00a0“Currently,\u00a0clinicians aren’t able to see if patients are responding to the prescribed exercises,\u00a0and patients have no way of knowing how they are progressing.”\u00a0<\/p>\n
With a wearable monitor, recovery data\u00a0collection can be ongoing\u00a0and occur\u00a0outside of therapy sessions,\u00a0allowing the therapist to make\u00a0timely\u00a0adjustments to treatment strategies for truly personalized rehabilitation.\u00a0<\/p>\n
Also, tracking movement\u00a0in a patient’s real-life environment may be more indicative of true performance, as opposed to movement that is artificially produced in a clinic.\u00a0The\u00a0wrist-wearable also captures movement at all times of day,\u00a0versus\u00a0at\u00a0just one snapshot\u00a0of time.\u00a0And finally, patients may find that tracking their own recovery progress increases their engagement in the therapy practices and keeps them motivated.\u00a0<\/p>\n
Lee is optimistic that\u00a0this increased transparency will\u00a0translate to improved therapy outcomes.\u00a0<\/p>\n
An\u00a0accelerometer\u00a0sensor in the\u00a0device captures\u00a0upper-limb movement,\u00a0which is then interpreted by a machine-learning algorithm developed by Lee and\u00a0his graduate student\u00a0and the lead author on the paper,\u00a0Ryan Wang.\u00a0<\/p>\n
“[Movement and impairment severity]\u00a0are related because the less severe you are, the more likely you’re going to move a lot, but they’re not exactly the same,” says Lee. “Increasing the use of the limbs-yes, we can encourage the person to make use of the limb more. But patients cannot\u00a0make instant\u00a0changes to\u00a0motor severity through\u00a0short-term\u00a0behavior change.”\u00a0<\/p>\n
Their\u00a0model, described in\u00a0Science Translational Medicine [LINK],\u00a0was trained on\u00a0accelerometer\u00a0data\u00a0and clinician assessment scores of\u00a0subacute (one week to six months\u00a0after a stroke) stroke patients and healthy individuals.\u00a0They found that their\u00a0algorithm was 40-50% more\u00a0accurate-meaning a better reflection of the patient’s true condition-compared to clinician evaluation.\u00a0<\/p>\n
In addition to the clinical application of using the device to inform personalized interventions, Lee’s\u00a0work\u00a0demonstrated\u00a0that the device has a strong research application.\u00a0In recreating\u00a0a previous\u00a0study using their own digital biomarker<\/a> instead of clinician observations, the researchers\u00a0found that they generated statistically significant results with 50% fewer\u00a0participants.\u00a0<\/p>\n “We can get a clear idea of the effectiveness of the intervention using a lower sample size and far fewer resources,” says Lee. This\u00a0can\u00a0expedite\u00a0the research process and reduce costs.”\u00a0<\/p>\n This work was supported by\u00a0the\u00a0National Institutes of Health.\u00a0With a provisional patent filed,\u00a0Lee\u00a0is pursuing\u00a0commercialization\u00a0through the startup\u00a0Lumid\u00a0Health\u00a0with support from\u00a0UMass Amherst\u00a0Institute for Applied Life Sciences’\u00a0Translational Seed\u00a0Award,\u00a0the\u00a0UMass\u00a0Office of Technology Commercialization & Ventures’\u00a0(OTCV)\u00a0Technology Development Fund\u00a0and participation in\u00a0the\u00a0NSF I-Corps Training Program.\u00a0<\/p>\n To further develop the technology,\u00a0Lee’s research partners\u00a0are currently recruiting stroke patients\u00a0for\u00a0a study held\u00a0at the Spaulding Rehabilitation Hospital in Boston.\u00a0Those\u00a0interested in participating\u00a0can\u00a0see the\u00a0inclusion criteria and apply here<\/a>.\u00a0<\/p>\n Source:<\/p>\n University of Massachusetts Amherst<\/a><\/p>\n Journal reference:<\/p>\n