UTILIZING AN INTEGRATED STATISTICAL APPROACH TO MODEL CHOLESTEROL LEVELS IN INDIVIDUALS WITH DYSLIPIDEMIA AND TYPE 2 DIABETES MELLITUS

Abstract

Background: The total cholesterol level signifies the quantity of cholesterol in the blood. This includes both low-density lipoprotein (LDL) and high-density lipoprotein (HDL) cholesterol. Elevated blood cholesterol levels lead to the formation of plaques on the walls of arteries. These plaques have the potential to obstruct arteries, and if they rupture, they can trigger the formation of blood clots. The combination of plaques and clots can impede the flow of blood. Researchers have undertaken a study to simulate cholesterol levels in individuals with dyslipidemia and type 2 diabetes mellitus. Objective: The objective of this paper is to utilize the developed methodology to model the factor associated with the total cholesterol status in patients with dyslipidemia and type 2 diabetes mellitus. This undertaking has the potential to improve the prediction of total cholesterol levels among the analyzed patients by integrating comprehensive supplementary data from a statistical standpoint. Material and Methods: Hospital Universiti Sains Malaysia (Hospital USM) was the source of the sample. This complex computational study design employed statistical modeling techniques to evaluate data descriptions of numerous variables, including total cholesterol, triglyceride levels, low-density lipoprotein (LDL) levels, high-density lipoprotein (HDL) levels, and alkaline phosphatase (ALP) levels. The developed method was implemented and evaluated using the R-Studio software and syntax. Using a multilayer feed-forward neural network model that includes the bootstrapping method and response surface methodology, the accuracy of the constructed method was determined. Results: The proposed technique demonstrated that the developed method is preferable. It proved that the accuracy of the analysis can be evaluated with higher certainty when data is divided into training and testing datasets, and that the integrated model technique offers complete knowledge about the true characteristics, enabling more accurate prediction. The neural network's mean square error, which is only approximately 0.021, exhibits the achievable precision. Conclusion: In this study, the proposed model demonstrates the method's capability for the improvement of research methodology. The outcome suggests that the methodology established for this investigation is capable of producing favourable results. The study's final analysis demonstrates that the model technique created for research is preferable.