Aleksandra Vulović*1,2, Tijana Geroski1,2, and Nenad Filipović1,2
1Faculty of Engineering University of Kragujevac, Sestre Janjić 6, 34000 Kragujevac, Serbia
2Bioengineering Research and Development Center (BioIRC), Prvoslava Stojanovića 6, 34000 Kragujevac, Serbia
aleksandra.vulovic [at] kg.ac.rs
Abstract
Experimental studies have shown that lower shear stress values lead to better femoral bone – hip implant connection. Numerical simulations have provided option to reduce the number of experimental studies through analysis of different hip implant surface topographies. However, this approach takes time as there are different model parameters that should be considered in order to understand how they affect the obtained shear stress values. The use of classification algorithms is an approach that could reduce the time required for simulation by providing information about models with biggest potential. Eleven model parameters related to model and surface topography were considered in combination with four classification algorithms – Support Vector Machines (SVM), K – Nearest Neighbor (KNN), Decision Tree (DT), and Random Forest (RF). The considered parameters were: Number of half-cylinders lengthwise (>0); Number of half-cylinder rows (≥0); Half cylinders added or removed from the surface (0 – removed; 1 – added); Distance between half-cylinders lengthwise (≥0); Distance between half-cylinders widthwise (≥0); Number of different radius values (1 or 2); Radius 1 value (>0); Radius 2 value (≥0); Distance from the edge where loading is located (≥0); Distance from the other edge of the model (≥0); Model includes trabecular bone (0 – not included; 1 – included). The aim was to apply previously mentioned algorithms to obtain information if the maximum shear stress value was above or below user-defined threshold. The obtained results show that this approach can be useful to obtain preliminary information about models that should be numerically analyzed.
Keywords: classification, finite element analysis, hip implant, surface topographies
Acknowledgement: This research is supported by the European Union’s Horizon 2020 research and innovation programme under grant agreement No. 952603 – SGABU. This article reflects only the author’s The Commission is not responsible for any use that may be made of the information it contains. Authors also acknowledge the funding by the Ministry of Science, Technological Development and Innovation of the Republic of Serbia, contract number [451-03-47/2023-01/200107 (Faculty of Engineering, University of Kragujevac)].
