Gait characterization in rare bone diseases in a real-world environment – A comparative controlled study

Abstract

BackgroundRare bone diseases (RBD) cause physical and sensory disability that affects quality of life. Mobility challenges are common for people with RBDs, and travelling to gait analysis labs can be very complex. Smartphone sensors could provide remote monitoring. Research questionThis study aimed to search for and identify variables that can be used to discriminate between people with RBD and healthy people by using built-in smartphone sensors in a real-world setting. MethodsIn total, 18 participants (healthy: n=9; RBD: n=9), controlled by age and sex, were included in this cross-sectional study. A freely available App (Phyphox) was used to gather data from built-in smartphone sensors (accelerometer & gyroscope) at 60 Hz during a 15-min walk on a level surface without turns or stops. Temporal gait parameters like cadence, mean stride time and, coefficient variance (CoVSt) and nonlinear analyses, as the largest Lyapunov exponent (LLE) & sample entropy (SE) in the three accelerometer axes were used to distinguish between the groups and describe gait patterns. ResultsThe LLE (p=0.04) and the SE of the z-axis (p=0.01), which are correlated with balance control during walking and regularity of the gait, are sufficiently sensitive to distinguish between RBD and controls. SignificanceThe use of smartphone sensors to monitor gait in people with RBD allows for the identification of subtle changes in gait patterns, which can be used to inform assessment and management strategies in larger cohorts.