International Journal of Computer Theory and Engineering

Abstract

Background: Bone screws are crucial elements intreating many types of open/closed fractures and arthroplasties in different joints in the body. The life time of many plates used in the fracture treatment are dependent on the screws function. The fracture of screws at any point of their lifetime will cause failure of the treatment for that specific pathology. This may lead to increase risk of new surgeries, osteomyelitis and less commonly septic arthritis. These complications not only have a negative impact on patient’s quality of life, increase comorbidity and mortality but also increase health care cost significantly. Method: We study the lifetime of bone joint screws made up of biostable (polysulfone) and biosorbable (poly-lactide-co-glycolide) polymer composite materials. The lifetime estimations under in vitro conditions were calculated based on extremely small sample size. A computational intelligent model has been developed to estimate the lifetimes, which is superior to least square and real-coded Genetic Algorithm methods, specifically, for a small sample size of data. Retrospectively, 76 X-rays with screw fracture indication (37 polysulfon screws and 39 poly-lactide-co-glycolide screws) constitute the sample size in this study. The funding sources were provided by the office of research services at Ryerson University. Results: The proposed model is a robust method because it does not converge to a local optimum, and also it does not need the use of differential calculus facilitating the computational implementation. The findings make a significant contribution to reliability of composite implants. Conclusion: The application of this model for two types of composite materials used for bone joint screws proves that polysulfone screws lifetime is better than that of poly-lactide-co-glycolide screws. Therefore, using the polysulfone screws could decrease the health related complications such as new surgeries and osteomyelitis.