Applying machine learning to predict patient-specific current CD₄ cell count in order to determine the progression of human immunodeficiency virus (HIV) infection

Abstract

This work shows the application of machine learning to predict current CD4 cell count of an HIV-positive patient using genome sequences, viral load and time. A regression model predicting actual CD4 cell counts and a classification model predicting if a patient’s CD4 cell count is less than 200 was built using a support vector machine and neural network. The most accurate regression and classification model took as input the viral load, time, and genome and produced a correlation of co-efficient of 0.9 and an accuracy of 95%, respectively, proving that a CD4 cell count measure may be accurately predicted using machine learning on genotype, viral load and time.

Keywords: Human immunodeficiency virus (HIV), antigens, CD4, computational biology, artificial intelligence, data mining, pattern recognition.

African Journal of Biotechnology Vol. 12(23), pp. 3724-3730