Related Experiment Video
Updated: May 16, 2026
The Multiple Sclerosis Performance Test (MSPT): An iPad-Based Disability Assessment Tool
Published on: June 30, 2014
Assessing Disability Level and Fatigue in Multiple Sclerosis With Smart Sock Sensor Technology
Julie F Stowell1, Victory A Ladipo2, Russell Jeter3
1Virginia C. Crawford Research Institute, Shepherd Center. Atlanta, GA; Georgia State University, Byrdine F. Lewis College of Nursing and Health Professions, Atlanta, GA.
Objective:
To examine whether smart sock-derived gait metrics can distinguish disability levels and detect fatigue-related gait changes during prolonged walking in people with multiple sclerosis (MS).
Design:
Cross-sectional study.
Setting:
Outpatient rehabilitation program.
Participants:
Thirty-two participants (N=32) with MS (Patient-Determined Disease Steps [PDDS] scores 0-6) completed 3 standardized walking tests while wearing smart socks. Gait metrics were derived via a validated analytic pipeline and included cadence, speed, step and stride length, gait cycle, stance, swing, and single- and double-support times. Disability groups were defined as mild-moderate (PDDS scores 0-4) and severe (PDDS 5-6).
Interventions:
Not applicable.
Main Outcome Measures:
Timed 25-foot walk test (T25FWT, 3 trials), 6-minute walk test (6MWT, 1 trial), and timed up and go (TUG, 3 trials) were used to assess gait metrics, and fatigue was assessed using pre/post Visual Analog Scale for Fatigue scores.
Results:
Across tests, participants with severe disability exhibited significantly slower cadence (-24% to -35%) and speed (-45% to -50%) with prolonged temporal phases, including gait cycle (+34% to 45%), stance time (+40% to 51%), and double-support (+45% to 64%) compared with the mild-moderate group (P≤.05). Correlations between PDDS and spatiotemporal metrics were strongest for temporal variables during the 6MWT (ρ≈0.60 to 0.68). Discrimination of disability status was highest for timing metrics (area under the receiver operating characteristic curve [ROC-AUCs] 0.84-0.87) compared with spatial metrics (ROC-AUCs≤0.70). Least Absolute Shrinkage and Selection Operator regression retained cadence as a stable predictor for the timed up and go, with strong cross-validated discrimination (ROC-AUC≈0.84). The 6MWT also revealed fatigue-related declines, with gait cycle (7.5%±14.4%) and double-support time (49.1%±53.5%) increasing; perceived fatigue rose by 197%±236% (P≤.05).
Conclusions:
Smart sock-derived spatiotemporal metrics differentiated disability levels and captured motor fatigability, with temporal coordination as the dominant impairment marker. These findings support the clinical utility of smart sock wearables as scalable tools for remote monitoring and precision rehabilitation in MS care.
