A Novel Vision transformer based deep learning model to predict chronic kidney disease

Chronic kidney disease (CKD) is a progressive condition where the kidneys lose their ability to filter waste and fluids from the blood, posing a significant global health challenge. As of 2017, CKD affected approximately 753 million people worldwide, contributing to 1.2 million deaths annually. Early detection is critical for timely intervention and effective treatment, but manual analysis often falls short due to limited access to screening and inadequate resources. Utilizing Machine Learning (ML) and Deep Learning (DL) methodologies offers a reliable approach for early-stage chronic kidney disease (CKD) prediction, offering clinicians precise data-driven insights to enhance decision-making. Hence, the aim of this proposed study is to address a critical issue in CKD detection by implementing the ResNet50 model, ResNet50 model using Random Forest Classifier and Vision Transformer for CKD prediction and classification where the models achieve an accuracy of 83.18%, 97.75% and 99.28% providing an effective solution for early diagnosis.