Stability and Transformation Prediction of HIV Viral Load Using the Application of Machine Learning and Principal Component Algorithms.

Abstract:

The current antiretroviral therapy (ART) medication gives hope to people living with human immunodeficiency virus (HIV) and those with acquired immunodeficiency syndrome (AIDS). This great development contributes to and improves the lives of HIV-treated patients. Lacking education and awareness among pupils in Nigeria and Africa as a whole about HIV is a major setback to this problem. In this study, only ART patients visiting the Federal Teaching Hospital in Gombe State, Nigeria, were followed. Artificial Intelligence (AI) single models of supervised Machine Learning (ML) algorithms and principal component analysis (PCA) algorithms were adopted. The performance of the model by AI algorithms showcases the stability of ART-treated patients. Consequently, the dimensionality plot in principal components showcases the transformation of the distance covered by the model's performance. However, the supervised (ML) single models harnessed in this study include a multilayer perceptron (MLP) and a neuro-fuzzy (NF). The PCA consists of factor loading, Initial eigenvalues, and the rotated component matrix. The NF DC explaining more than 89% of the data (0.902418 and 0.895384) outperforms MLP, in both training and testing, proving that longer duration on ART-drug can effectively stabilize the suppression of viral load. The factor rotation converged in seven iterations to retain 3 components with eigenvalues above 1, revealing that the new transformed PCA component did a good job of explaining more than 58% of the distance covered by the performance of the model using the first three variables.