AUTHOR=Nayak Saurav , Singh Arvind , Mangaraj Manaswini , Saharia Gautom Kumar 

TITLE=Predicting immune risk in treatment-naïve HIV patients using a machine learning algorithm: a decision tree algorithm based on micronutrients and inversion of the CD4/CD8 ratio

JOURNAL=Frontiers in Nutrition

VOLUME=Volume 11 - 2024

YEAR=2024

URL=https://www.frontiersin.org/journals/nutrition/articles/10.3389/fnut.2024.1443076

DOI=10.3389/fnut.2024.1443076

ISSN=2296-861X

ABSTRACT=Introduction: Micronutrients have significant functional implications for the human immune response, and the quality of food is a major factor of the severity and mortality of HIV in individuals on antiretroviral therapy. A fall in CD4 lymphocyte numbers and a spike in CD8 lymphocyte counts are the hallmarks of HIV infection, which causes the CD4/CD8 ratio to invert from a normal value of > 1.6 to < 1.0. In this study we are trying to analyse whether the nutrient status of the individual have an impact on the CD4/CD8 ratio inversion by utilising Machine Learning algorithms.Method: Fifty-five confirmed HIV positive patients who have not started with anti-retroviral therapy were included in the study after getting informed written consent from the participants.Fifty-five age and sex-matched, relatives, and attendants of patients found negative in screening test were enrolled as controls. All individual patient datapoints were analysed for developing the models with an 80-20 train-test split. The four trace elements, Zn, P, Mg and Ca were utilized by implementing Random Forest Classifier. The targets were inverted CD4/CD8 Ratio.Result: A total of 110 participants' data were included in the analysis. The algorithm thus generated had a Sensitivity of 80% and Specificity of 83%, with LR+ of 4.8 and LR-of 0.24.The utilization of ML algorithm boosts the narrow evidence that exists currently regarding the role of micronutrients especially trace elements in the causation of immune risk. Being one of the first studies of this kind that utilized ML based Decision Tree algorithm to classify immune risk in HIV patients is inherently our strength.Our study uniquely corroborates nutritional data to immune risk in treatment naïve HIV patients through the utilization of Decision Tree ML algorithms. This will subsequently be an important classification and prognostic tool in the hands of clinicians.