RenalNet: An End-to-End Hybrid Deep Learning and Ensemble Model for Accurate Kidney Disorder Classification

Kidney disorders are a major public health issue affecting millions worldwide. Renal calculi and cancer are prevalent, causing chronic kidney disease(CKD). Artificial intelligence can assist in the precise diagnosis of kidney diseases.We developed an end-to-end system using a custom CNN model to extract informative features from kidney images. Multiple ML classifiers provide complementary perspectives. A High-Performance Filter(HPF) selects the top classifiers for an ensemble model. The interactive web interface enables real-time prediction. Our approach achieves excellent performance metrics with an accuracy of 99.98%. Overall, We demonstrate an automated system combining deep learning and ensembling for the accurate characterization of diverse kidney abnormalities, which can aid in the early detection and treatment of CKD.