The process of designing bio-inspired knee joint has been a challenging issue due to the complicated kinematics and dynamics of the human knee joint. This paper addresses this issue by presenting a design methodology that has been used to model the human knee joint from Magnetic Resonance Imaging (MRI) scans, and curve fitting method to approximate the condylar profiles. The study was to extract the knee profiles from MRI scans and create a 3D model of a condylar knee joint for robotic applications. A medical imaging software was used to assist the process of converting these scans into a 3D model. This model was then imported into a 3D Computer Aided Design (CAD) software and various profiles from the model were extracted to derive the articular surfaces of the condylar knee. The condylar profiles that were extracted to create the bio-inspired knee model based on articular surfaces, were also analysed in MATLAB using polynomial equations. The results of the polynomial equation are a good fit of the condylar profile extracted from the 3D model, which can benefit the design of a prosthetic joint. The generated 3D model of the knee from the MRI scans can also be used to assist with the sizing and movement of a life-like knee implant.