Motion analysis system

All patients underwent a 3D opto-electronic motion analysis during standing and walking trials between 2 and 10 days before surgery. Passive reflective marker clusters (four markers each) were attached to participants at T1, L1, and S2 levels of the spine, with additional reflective markers at other thoracic and lumbar spinal levels, as well as the head, manubrium of the sternum, posterior superior iliac spines, shoulders, upper and lower arms and legs, and feet. The marker position was recorded by a motion analysis system (Vicon Motion Systems, Centennial, CO).

Kinematic data were collected to characterize subjects’ locomotor behaviors while walking on the treadmill during locomotor and perceptual trials. A motion analysis system (Vicon Motion Systems, Oxford, UK) was used to collect kinematic data at 100 Hz. A quintic spline interpolation was used to fill gaps in the raw kinematic data (Woltring; Vicon Nexus Software, Oxford, UK). Subjects’ movements were tracked via passive reflective markers placed bilaterally over the hip (greater trochanter) and ankle (lateral malleoulous) and asymmetrically on the thigh and shank to distinguish the legs. The duration of treadmill trials was defined by real time kinetic detection of heel strikes. Heel strikes were identified with raw vertical kinetic data collected from the instrumented treadmill (Bertec, Columbus, OH, United States).

Kinematic data were collected to characterize subjects' locomotor movements on the treadmill and overground. A motion analysis system (Vicon Motion Systems, Oxford, UK) was used to collect kinematic data at 100 Hz. A quintic spline interpolation was used to fill gaps in the raw kinematic data (Woltring; Vicon Nexus Software, Oxford, UK). Subjects' movements were tracked via passive reflective markers placed bilaterally over the hip (greater trochanter) and ankle (lateral malleoulous) and asymmetrically on the thigh and shank to distinguish the legs. The duration of treadmill trials was defined by real time kinetic detection of heel strikes whereas the duration of overground trials was defined by elapsed time. Heel strikes were identified with raw vertical kinetic data collected from the instrumented treadmill (Bertec, Columbus, OH, United States). Given that force data was only available during treadmill trials, but not overground trials, the greatest forward excursion of the ankle was used to identify heel strikes post-processing so that the same heel strike detection could be used across treadmill and overground walking epochs as had been previously done (e.g., Torres-Oviedo and Bastian, 2010 (link), 2012 (link); Sombric et al., 2017 (link)).