This study focused on the development of a predictive model for assessing the survival of diabetes mellitus (DM). The study identified the variables monitored during the treatment of diabetes mellitus patients, formulated, simulated and validated the predictive model for the survival time of diabetes mellitus patients. Following the identification of relevant variables, data was collected from 29 patients alongside their survival time. The predictive model for assessing the survival time of DM was formulated using the support vector machines (SVM) and multi-layer perceptron (MLP) classifiers. The models were simulated using the 10-fold cross validation technique via the WEKA® simulation software. The model were then validated and compared based on the mean absolute error (MAE) and the mean square error (MSE) rates. The results showed that 21 factors were assessed among the collected data required for assessing survival time and the MLP showed the better ability to assess the survival time with error rates of 0.000 however had a longer model build time of 0.77 seconds compared to the build time of 0.02 seconds using SVM. The study concluded that information about the assessment of the survival time of patients with DM can provide decision support aimed at providing alternative or continuous treatment to DM patients.
Diabetes Mellitus, Predictive Model, Supervised Machine Learning, SVM, MLP
Cite this paper
Amoo A. O., Oyegoke T. O., Balogun J. A., Bamidele S. A., Idowu P. A. (2020) Survival Model for Diabetes Mellitus Patient Receiving Treatm. International Journal of Computers, 5, 1-13
Copyright © 2020 Author(s) retain the copyright of this article.
This article is published under the terms of the Creative Commons Attribution License 4.0