Comparative Study of Four Instrumented Mobility Analysis Tests on Neurological Disease Patients

Wearable inertial sensor systems are widely used for mobility analysis of neurological disease patients. Different assessment methodologies, including - Timed Up and Go, Sit and Stand, walking tests and different sensor configurations are used in practice. Sensor signal processing complexities of competing methods have not been thoroughly investigated. Apparently, computational robustness and noise insensitivity are the key parameters for instrumented mobility monitoring, when automated patient assessment is targeted. This paper describes results of comparisons of 4 different mobility assessment methods conducted on 35 multiple sclerosis patients with variable clinical disability scores. The results demonstrate high variability in inertial sensor signal patterns, even for patients with rather weak disability symptoms. Efficiency of widely used instrumented mobility analysis methodologies is discussed, concluding with the authors' proposals.