Year: 2025
Type: article
Abstract: Chronic Kidney Disease (CKD) is a progressive health condition that requires timely diagnosis and ongoing monitoring to prevent severe complications. This study explores the development of machine learning models aimed at predicting the onset and stage of CKD using clinical data. Various models, including Convolutional Neural Networks (CNN) for handling large datasets, SMOTE for class imbalance correction, and CatBoost for rapid and accurate prediction, are employed. Feature extraction techniques such as Recursive Feature Elimination (RFE) are integrated to improve model performance, alongside hybrid approaches that combine multiple algorithms for enhanced accuracy. The dataset, consisting of clinical parameters such as blood pressure, sugar levels, and kidney function indicators, undergoes a rigorous process of cleaning, normalization, and imputation of missing values.
Language: English
Subfield: Health Information Management
Field: Health Professions
Domain: Health Sciences
Open Access status: bronze
OpenAlex ID: W4410178217