Predictive Modelling for Chronic Kidney Disease: A Hybrid Approach Using Exploratory Data Analysis, Machine Learning and Explainable AI

Predictive Modelling for Chronic Kidney Disease: A Hybrid Approach Using Exploratory Data Analysis, Machine Learning and Explainable AI

연구 분야: Artificial Intelligence

학회: 2025 8th International Conference on Electronics, Materials Engineering & Nano-Technology (IEMENTech)

Chronic Kidney Disease (CKD) is a growing health crisis in India, affecting nearly 17% of the population. Due to its often asymptomatic nature, CKD is frequently diagnosed only in advanced stages, leading to expensive treatments and poor patient outcomes. To address this, we explore the potential of AI and machine learning to enhance early detection and predict CKD progression, enabling timely interventions. This paper presents a Machine Learning (ML) model with Exploratory Data Analysis (EDA) for CKD prediction, emphasizing both accuracy and interpretability, which is critical for clinical adoption. We integrate Explainable AI (XAI) techniques, specifically LIME, to provide healthcare professionals with clear insights into model predictions, promoting trust and transparency. Our approach combines predictive power with interpretability, offering a practical solution for early CKD detection and improving clinical decision-making. We discuss the challenges and future directions of integrating AI in healthcare, with the aim of reducing CKD-related healthcare costs and improving patient outcomes in India.