The design of a cost-effective mechanical prosthetic knee

Abstract

The limited functionality of a typical low-cost knee prosthesis results in poor gait performance and subsequently leads to reduced quality of life due to adverse health outcomes. Several concepts and methods are used in premium knees to improve functional performance (e.g., stance-phase locking and swing-phase damping), but due to high costs and specialised maintenance from international suppliers, few patients can be helped with these devices. Therefore, the problem in South Africa is restricted access to cost-effective prosthetic knees with functional gait performance, exacerbated by the lack of expertise in the local design and manufacture of these prostheses. This study involved the design, manufacture, and testing of a cost-effective mechanical prosthetic knee. A narrative review of the literature on prosthetic knee types, control mechanisms, materials, design tools, design frameworks, and testing methodologies informed the design approach. A voluntary control four-bar mechanism knee was selected as a suitable candidate for the purposes of this study. A MATLAB based link-segment mathematical model was created to evaluate candidate geometry for stability, range of motion, and ground clearance. The geometry was then used as a base to design a functional prototype that can be used for gait analysis. Modular components, standard fasteners, and readily available materials were used to fabricate the prototype. It was manufactured using a CNC machine and a precision lathe. The prototype performance was evaluated through a controlled human gait trial using 3D motion-capture analysis. The participant completed walking trials with the mechanical prototype and with a microprocessor-controlled knee, which served as a performance baseline. Gait symmetry was assessed using cross-correlation (r) for waveform parameters such as joint kinematics and ground reaction force, while the Symmetry Index was used to indicate quality when comparing spatio-temporal parameters such as cadence, swing time, and stance time. The average maximum r for the prototype was 0.948 with a signed lag of -4%, while the corresponding values for the microprocessor knee was 0.964 and -5%, respectively. The SIs for crucial metrics were evaluated for the prototype and the microprocessor knee respectively: stride time (1.34% and 1.83%), speed (1.68% and 2.19%), step length (10.14% and 10.85%), step time (14.94% and 17.82%), steps/minute (15.08% and 17.8%), stride length (0.32% and 0.34%), stance time (18.1% and 16.28%) and swing time (26.36% and 21.87%). The prototype demonstrated consistent and physiologically appropriate gait parameters and gait symmetry metrics that closely match those of the microprocessor knee. Although the microprocessor knee still performed better overall, the prototype demonstrated comparable functional outcomes at radically lower cost(≈ R22 000 vs > R500 000). However, the scope of the testing was limited to level ground walking at self-selected speeds with one participant. Therefore, more research is required to investigate the operational limitations of the prototype. Nevertheless, this study achieved it's aim to design, manufacture and evaluate a cost effective mechanical knee. The results were encouraging and contributed to the local knowledge base for the design of transfemoral prosthetic solutions for lower resource settings.