Multisensor Data Can Assess MS-Related Limb Dysfunction
Extracted features from three sensors correlate with physician-rated, patient-reported disability status
WEDNESDAY, March 4, 2020 (HealthDay News) -- Features extracted from a multisensor correlate strongly with physician- and patient-reported disability outcomes in multiple sclerosis (MS), according to a study published online Feb. 26 in the Annals of Clinical and Translational Neurology.
Alireza Akhbardeh, Ph.D., from the Johns Hopkins University School of Medicine in Baltimore, and colleagues assessed baseline data for 117 participants with MS and 30 healthy controls to create a novel neurological vital sign and reliably capture MS-related upper and lower limb dysfunction. Finger and foot taps were completed by participants wearing the MYO band with accelerator, gyroscope and surface electromyogram sensors. To extract spatiotemporal features from raw sensor data, signal processing was performed. Extracted features were compared to physician- and patient-reported disability outcomes.
The researchers found that for the final selected features, intraclass correlation coefficients varied from 0.80 to 0.87. Time-based features were able to differentiate cases from controls. There was a strong correlation noted for the most informative combination of all extracted features from all three sensors with physician Expanded Disability Status Scale; equally strong associations were seen with patient-reported outcomes (World Health Organization Disability Assessment Schedule). After adjustment for age and sex, the associations persisted.
"The use of multisensors allows for use of complementary data-types that can be employed for a more comprehensive view of the movement," a coauthor said in a statement. "The types of sensors we used are widely available in different hardware products."
One author disclosed financial ties to the biopharmaceutical industry.