THE BIOMECHANICAL ASSESSMENT OF GAIT FOR STROKE PATIENTS AND THE IMPLICATIONS IN DIAGNOSIS AND REHABILITATION
Background Stroke commonly occurs in the middle-aged and elderly population, and the diagnosis of early stroke remains challenging. Patients who have suffered a stroke have different gait patterns compared to healthy people. Advanced techniques of motion analysis have been routinely used in the clinical assessment of cerebral palsy. However, so far, little research has been done on the direct diagnosis of early stroke patients using motion analysis. Objectives The aim of this study was to investigate whether patients with stroke have a different gait compared to healthy people and what the features of the gait are after stroke. Methods Thirteen patients with hemiplegia caused by both types of stroke (ischaemic or haemorrhagic) were recruited for this study (all male; age 68.77±5.02 years; height 169.9±5.7 cm; body weight 81.61±8.74 kg) plus 20 healthy subjects (12 male and 8 female; age 65.05±7.63 years; height 166.3±6.1 cm; body weight 75.35±12.74 kg). All subjects’ gait data were collected using Vicon Nexus® to obtain the gait parameters, kinetic, and kinematic parameters of the hip, knee, and ankle joints in three planes of both limbs. Participants were asked to walk along a 10 m walkway at a comfortable speed. There were ten trials of walking. From the recorded trials, three good ones were analysed using a Vicon Plug-in-Gait model to obtain the gait parameters, e.g. walking speed, cadence, stride length, and joint parameters, e.g. joint angle, force, moments, and power. The gait data were further analysed by using neural network models and principal component method in terms of diagnosis and prediction of strokes. Result The temporal-spatial variables of the stroke subjects were compared with the healthy subjects, and it was found that there was a significant difference between the groups. The step length, speed, and cadence were lower in stroke subjects compared to the healthy group. The stroke patients group showed a significantly decreased gait speed (mean and SD: 0.85± 0.33 m/s), cadence ( 96.71±16.14 step/min), and step length (0.509±017 m) compared to the healthy group, where the gait speed was 1.2±0.11 m/s, cadence 112± 8.33 step/min, and step length 0.648±0.43 m. Moreover, it was observed that patients with stroke have significant differences regarding their ankle, hip, and knee joints’ kinematic and kinetic in all planes (sagittal x, coronal y, and transverse z) compared with the healthy group. Using the neural network models, the stroke gait could be identified with accuracies as high as 95%-100%, although the results were from a small sample size. The principal component method could partially be helpful to identify the stroke gait. Discussion and Conclusion Stroke patients showed increase ankle plantar flexion and decreased dorsiflexion on the affected side. In the terminal double stance phase, they had increased hip flexion and hip abduction in order to clear their foot from the floor. Also, stroke patients had a lack of knee flexion, which causes a stiff knee gait. These results suggest that the biomechanical behaviour of the hip and ankle joint during the gait cycle in patients with stroke is closely related. These findings could help physiotherapists to understand gait problems after stroke and provide a justification for applying their treatment to the correct areas. The neural network can potentially be used to diagnose the strokes and further modelling with larger sample size should be carried out in the future.