CHRONIC KIDNEY DISEASE STAGE CLASSIFICATION IN HIV PATIENTS USING MACHINE LEARNING TECHNIQUES

Abstract

Chronic Kidney Disease (CKD) is a major global health concern, associated with high morbidity and mortality. Early stages of CKD are often asymptomatic, leading to delayed diagnosis, particularly in patients with HIV who are at higher risk of severe kidney impairment. Early detection is critical for timely medical intervention and slowing disease progression. With the growing availability of pathology data, machine learning techniques have become increasingly valuable for disease classification and prediction.This paper presents a machine learning-based approach for classifying CKD in HIV-infected patients. CKD stages are determined based on estimated glomerular filtration rate (eGFR) and albuminuria, following National Kidney Foundation Kidney Disease Outcomes Quality Initiative (NKF KDOQI) guidelines. A Deep Neural Network (DNN) model achieved 99% accuracy in classifying CKD stages, highlighting its potential for early and precise detection. Key aspects of CKD management in HIV patients include regular monitoring of eGFR, assessment of urine albumin-to-creatinine ratio (ACR), adjustment of antiretroviral therapy, control of comorbidities such as hypertension and diabetes, and lifestyle modifications. Multidisciplinary care involving nephrologists, infectious disease specialists, and other healthcare professionals is essential. Ongoing research continues to explore novel therapies and the complex interactions between HIV infection, antiretroviral treatment, and kidney health.