Leverage mechanical alterations during walking at self-selected speed in patients with Parkinson's disease

MINIMIZED FEATURE SELECTION FOR DETECTION OF PARKINSON’S DISEASE USING NEURO-FUZZY SYSTEM

This study presents a methodology for detecting Parkinson’s disease using a neuro-fuzzy system (NFS) with feature selection. From all the 22 features, the five most accurate minimized features were selected using neural networks with weighted fuzzy membership functions (NEWFMs), which supported the nonoverlapping region method (NORM). NORM eliminates the worst features and can select the minimized features constituting each interpretable fuzzy membership. As an input to the NEWFMs, all 22 features indicated a performance sensitivity, specificity and accuracy of 87.43%, 96.43% and 88.72%, respectively. In addition, at least five features of the NEWFMs showed performance sensitivity, specificity and accuracy of 95.24%, 85.42% and 92.82%, respectively.

Mechanical advantage and joint function of the lower limb during hopping at different frequencies

Mechanical output at a joint level could be influenced by its leverage characteristics and by its functional behaviour and both could change to accommodate the demands of a given locomotor task. In this study, the mechanical power generated at the knee and ankle joints and their functional indexes (i.e. damper, strut, spring and motor like-function) were calculated by using 3D kinematic and kinetic data during hopping at 2, 2.5, 3 and 3.5 Hz. The effective mechanical advantage (i.e. the ratio between internal and external moment arm) of the knee (EMA_K) and ankle (EMA_A) and joint stiffness were calculated as well. Joint stiffness increased with frequency whereas positive and negative joint power decreased with it, the ankle power values being always larger (20-50%) than those at the knee. EMA_A reached its highest value (0.4) during the propulsive phase at 3 Hz whereas no significant changes in EMA_K were observed as a function of frequency in both the absorption and propulsive phases. Knee joint-functional index shifted from a spring to a strut-like function with increasing frequency (from 56 to 8% and from 4 to 51%, respectively) while the ankle operated mainly as a spring (from 90 to 53%), its damper and motor-like indexes being negligible at all frequencies (<5%). Therefore, in hopping, the knee works to dissipate mechanical energy (the combination of its damper and strut indexes increase from 23 to 72% at these frequencies) and the primary source of mechanical power is attributable to the elastic function of the ankle.