Two New Indices to Assess Gait Disturbances Applied to Anterior Cruciate Ligament Reconstructed Knees

The purpose of this study was to develop two new gait indices - the Normal Gait Index (NGI) and the Abnormal Gait Index (AGI) - that include a symmetry measurement, to conclude about the recovery degree of patients subjected to the reconstruction of the anterior cruciate ligament (ACL). This methodology comprises the use of a computational intelligence technique to generate each patient's specific reference vertical ground reaction forces (VGRF). Data from ACL reconstructed patients walking at five different speeds were collected using instrumented shoes. The developed gait indices include a symmetry index (SI) and a gait error (GE) that takes into account the differences between sixteen parameters from the patients and the respective references' VGRF patterns. The reference VGRF were produced using the Extreme Learning Machine algorithm. Moreover, it was suggested a graphical representation of the GE and SI as well as a normal gait region (NGR) for each patient, providing useful data to evaluate the rehabilitation processes. The results showed that the proposed methodology can be used to evaluate the recovery degree of patients subjected to the ACL reconstruction, and it is preferable to the use of a standard measurement. Therefore, the proposed methodology can be used for gait pathology diagnosis, assessing normal (NGI) and abnormal (AGI) gait patterns. Moreover, it provides useful information to the doctors, showing the limb and stride durations with more deviations, enabling the adjustment of rehabilitation procedures.