KINEMATIC TRAJECTORIES WHILE WALKING WITHIN THE LOKOMAT ROBOTIC GAIT-ORTHOSIS

Abstract

Purpose/Hypothesis: To compare the kinematic patterns of subjects walking in the Lokomat robotic gait-orthosis with those demonstrated during treadmill walking, and to quantify the amount of relative movement inside the device. The field of neurorehabilitation has been adopting robotic devices to help deliver mass-practice therapy to individuals with various types of neurological disorders (Hidler et al, 2005 for review). The Lokomat (Hocoma AG, Volketswil Switzerland) is the most widely adopted robotic system used for gait training. Previous studies have quantified changes in muscle activation pattern when subjects walk in the Lokomat to those on a treadmill (Hidler and Wall, 2005). No studies to date have tracked the kinematic patterns of the legs inside the Lokomat. Number of Subjects: 6 healthy subjects (4 male, 2 female) participated in the study. Materials/Methods: A Codamotion (Charnwood Dynamics, UK) motion analysis system was used to track the kinematic patterns of the subject's pelvis and legs, and the robotic limbs of the Lokomat. Marker clusters were placed on each limb segment and tracked at 100 Hz. Subjects walked on the treadmill with and without the Lokomat at four randomly selected walking speeds (2.0, 2.4, 2.8, 3.2 km/hr). For all trials, subject and Lokomat marker positions and ground reaction forces were recorded for 30-second step sequences. A number of kinematic metrics were compared between the two modalities using a one sample t-test (alpha = 0.05), including ankle, knee, and hip range of motion, maximum and minimum joint flexion and extension angles, and the point in the gait cycle each occurred. The relative knee and hip movement between the subject's and Lokomat's joint centers in both the frontal and sagittal planes were quantified. Results: While some subjects experienced slight differences in kinematic patterns, there were no significant differences in any of the metrics computed for all subjects across walking modalities. Despite there being no differences in sagittal joint angular patterns, the side to side motion of the hip in the medial-lateral plane was significantly less in the Lokomat than on the treadmill, sometimes restricted by as much as 75%. The relative movement of subjects' legs inside the Lokomat was substantial, where the variability in the knee joint center in the sagittal plane was approximately 1 cm while deviations exceeded 2 cm for the hip joint. Conclusions: With proper setup, the Lokomat gait patterns mimic those utilized during treadmill ambulation. Differences in the amount of medial-lateral hip motion can be attributed to the reduced degrees of freedom the Lokomat places on the subject. Clinical Relevance: The finding that there is significant movement of the subject's legs inside the Lokomat is important clinically, as a variable step pattern has been shown to be a more effective gait training strategy than one which forces the same kinematic pattern across successive steps (Cai et al., 2005).