Phase Portrait Based Index for Gait Asymmetry Assessment in Post-stroke Patients

Gait asymmetry in stroke patients results in various disadvantages, such as increased fall risks and metabolic costs. To tackle this problem, various wearable robots were introduced for daily assistance, improving gait asymmetry of stroke patients. However, their performances were limited because they couldn’t target gait asymmetry over whole gait phases. Furthermore, their gait asymmetry measurements include errors under variable speeds, which are common in stroke patients. To solve this problem, a new metric quantifying gait asymmetry in real-time is required. In this study, we propose an index that defines real-time gait asymmetry by comparing the phase portraits of the affected and unaffected limbs. Proposed asymmetry index is advantageous for improving gait asymmetry due to its real-time measurability over gait phases and robustness over variable gait speeds. We investigated whether the index is consistent with conventional gait symmetry index, Range of Motion Symmetry Index. Three stroke patients’ data were analyzed and revealed a positive correlation between the two. Also, we validated that the proposed index can be calculated over gait phases in real-time.