Customization of Joint Articulations Using Soft Computing Methods

The wear of artificial joints, which is one of the main causes of re-implantation of the hip joint, can be minimized or entirely eliminated through a precise adjustment of artificial biobearing to individual conditions of the patient. This is possible through utilization of modern engineering tools for support of doctors in multi-aspect selection of head elements and acetabulum that work in the artificial joint. Despite a substantial number of materials used for the friction pair of both head and acetabulum, there is no perfect and universal set of biomaterials that allow for recreation of functionality of joints in all patients, while improperly selected joint articulation results in faster wear of components, migration of products of wear to soft tissue and, consequently numerous complications and necessity of repeated surgical interventions. An innovative approach that allows for customization of joint replacement, which is impossible to be achieved using conventional methods, is to utilize machine learning systems to adjust friction pair to anthropometric and goniometric characteristics of a patient. The internal elements which are used to train a fuzzy classifier ensemble are results of clinical, experimental and numerical studies that allow for prediction of the functional cycle for a patient.