Publication Date: 2023/04/29
Abstract: Developing effective survival analysis models would help guide the decision-making in managing major health challenges. Model development can be achieved through various approaches. Diabetes is a health challenge in Nigeria that has attracted the interest of researchers thus much research has been carried out as regards its management necessitating the development of models. This study carried out a machine learning analysis on diabetes data collected from Central Hospital, Warri, Delta State implementing Cox-PH Model due to the role both play in survival analysis. A dataset of 100 diabetic patients' records was collected. The dataset was used for training multiple machine learning algorithms, namely, SupportVector (SVM), Knearestneighbors (KNN) classifier, etc., and the proposed model (Cox-PH Hybrid or CPH-SML). The performance evaluation of the machine learning algorithms and the proposed model gave accuracy levels as follows: KNN47%, SVM; 74%, and Cox-PH Hybrid-96%. The concordance index was used to evaluate the proposed model and it had an index of 0.7204, on several covariates such as Age, Gender, Education, Marital Status, history of smoking, SBP, DBP, etc. From this study's analysis of the diabetic data, it was able to conclude that the variables associated with diabetes mortality are; the age of the patient and diabetes types. The patients' hazard ratio reduces when they are young compared to when they are old. The patient's hazard ratio is also dependent on the diabetes type. Thus, early diagnosis and proper health management of diabetics can prolong the age of diabetic patients.
Keywords: Survival Model, Machine Learning, Cox Proportional Hazard, Diabetes.
DOI: https://doi.org/10.5281/zenodo.7937239
PDF: https://ijirst.demo4.arinfotech.co/assets/upload/files/IJISRT23APR586_(2).pdf
REFERENCES