Background

Osteoarthritis and the pain associated with it result in gait pattern alteration, in particularly gait asymmetry when the disease is unilateral [1–2]. The quantification of such asymmetry could assist with the diagnosis and follow up. Various asymmetry indices have been proposed to compare the spatiotemporal, kinematic and kinetic parameters of lower limbs during the gait cycle. One, the Continuous Relative Phase [3] compares the joints angle and its derivatives to assess the gait asymmetry during the gait cycle. However, the indices rely on marker based gait measurement systems that are costly and generally require manual examination, calibration procedures and the precise placement of sensors/markers on the body of the patient.

Osteoarthritis and the pain associated with it result in gait pattern alteration, in particularly gait asymmetry when the disease is unilateral [1–2]. The quantification of such asymmetry could assist with the diagnosis and follow up. Various asymmetry indices have been proposed to compare the spatiotemporal, kinematic and kinetic parameters of lower limbs during the gait cycle. One, the Continuous Relative Phase [3] compares the joints angle and its derivatives to assess the gait asymmetry during the gait cycle. However, the indices rely on marker based gait measurement systems that are costly and generally require manual examination, calibration procedures and the precise placement of sensors/markers on the body of the patient.

To overcome these issues, a new asymmetry index was proposed in [4]. It uses an inexpensive, easy-to-use and markerless depth camera (Microsoft Kinect™) output. Without requiring joint localization, it directly uses depth images provided by the Kinect™. It is based on the longitudinal spatial difference between lower-limb movements averaged during several gait cycles. To evaluate the relevance of this index ILong, its sensitivity versus the position of the sensor and the sensitivity versus the number of gait cycle, fifteen healthy subjects were tested on a treadmill walking normally and with an artificially induced gait asymmetry created by placing a thick sole under one shoe. The gait movement was simultaneously recorded using two Kinects™, one placed in front of and another behind the subject, and a motion capture system.

The Continuous Relative Phase computed with the Kinect™ skeleton failed to assess gait asymmetry. With the Kinect™ placed in front of and behind the patient the proposed longitudinal index distinguished the asymmetrical gait (p<0.001). Moreover, the correlation coefficient between the index measured by Kinect™ and the ground truth of this index measured by motion capture is more than .85 when using one stride and reaches .90 when using at least five strides.

This gait asymmetry index measured with a Kinect™ is low cost, easy to use and is a promising development for clinical gait analysis for Osteoarthritis disease diagnostic and follow up.