Objective: An appropriate biomarker for spinocerebellar degeneration (SCD) is needed.

Background: Previously we reported that the average amplitude of medial-lateral of straight gait gained by triaxial accelerometers fixed in the middle of the upper back (ML) could be an appropriate biomarker. There is a vast amount of literature indicating that inertial sensors can directly analyze acceleration (e.g., the root mean square of acceleration along each of the 3 axes).

Methods: We analyzed 25 SCD patients by triaxial accelerometers. Acceleration signals were measured during 6 min walking by two sets of triaxial accelerometers that were secured with a fixation vest to the middle of the subject’s lower and upper back. The subjects were evaluated while they shuttle-walked a 30m straight line for 6 min. When standing in the anatomical position, the orientation of the three acceleration axes, X, Y, and Z, were medial/lateral (ML), vertical (VT), and anterior/posterior (AP), respectively. We extracted two gait parameters, the average and the coefficient of variation (CV) of the motion trajectory amplitude, from each acceleration component by the following methods: 1. The acceleration signal was integrated twice in the time domain and processed with high-pass filtering based on a moving window average to generate motion trajectory, namely, relative displacement. 2. The upper and lower envelopes of the trajectory signal (denoted as Y1 and Y2) were obtained by spline interpolation of its positive and negative peaks, respectively. 3. The amplitude time series was defined as (Y1 − Y2) / 2, and divided into two parts: signals corresponding to straight walking and to turning around the cone. In addition, we calculated the root mean square (RMS) of the 3 axes and looked at correlation with Scale for the assessment and rating of ataxia(SARA).

Results: Average ML amplitude in straight gait of the upper back correlated significantly with SARA (R= 0.65, p < 0.001). In RMS, the VT RMS of the upper back when turning showed the best correlation to SARA (R = -0.52, p = 0.007).

Conclusions: A quantitative evaluation of gait ataxia by accelerometers showed that the ML acceleration rate calculated using double integration was more correlated to SARA than those by root mean square.

« Back to 2018 International Congress

MDS Abstracts - https://www.mdsabstracts.org/abstract/a-comparison-of-relative-displacement-by-double-integration-with-root-mean-square-in-the-quantitative-evaluation-of-gait-ataxia-by-triaxial-accelerometers/