Mathematical Modelling of Diabetes Mellitus and Associated Risk Factors in Saudi Arabia
Entissar Almutairi (Brunel Uneversity London, United Kingdom (Great Britain)); Maysam F Abbod (Brunel University, United Kingdom (Great Britain)); Take Itagaki (Brunel University London, United Kingdom (Great Britain))
Mathematical modelling has been successfully applied to the healthcare domain and epidemiological chronic diseases. Specifically, diabetes mellitus, which is classified as an epidemic due to its high rates of prevalence around the world. In this paper, an attempt is made to model diabetes mellitus in Saudi Arabia along with the most relevant risk factors, namely smoking, obesity and physical inactivity for adults aged 25 years and over. The aim of this study is based on developing different mathematical models for the purpose of studying the trends in incidence rates of diabetes over 15 years (1999 -2013) and get predictions for the future level of the disease up to 2025, as this should support health policy planning and identifying the necessary costs of controlling diabetes. Different models were developed namely, Multiple Linear Regression, Adaptive Neuro-Fuzzy Interference System, Artificial Neural Networks, Support Vector Regression and Bayesian Linear Regression. A combination of these models is performed to improve the prediction accuracy using combination methods such as Average, Weighted Average and Majority Voting. Moreover, the combined model was validated by comparing the prediction of prevalence estimates by WHO, IDF and Saudi General Authority for Statistics. Improved accuracy was achieved in comparison to these studies.
Conference: UKSim-AMSS 22nd International Conference on Computer Modelling and Simulation, UKSim2020
Published: Mar 25, 2020