A Novel Method of Chronic Kidney Disease Detection Based on Deep Learning Technique

There are several causes of kidney illnesses, which are typically classified as either neoplastic or non-neoplastic. The potential for accurate management of kidney disorders is made possible by deep learning based on medical imaging. Deep learning based on imaging has been extensively used for several kidney disease clinical situations. Given the high death rate linked to it, early detection is crucial to ensuring rising treatment success rates and declining mortality. Raw CT medical image data consists of noise. In this study, we have proposed a new method to reduce the noise for pre-processing using the 2D Adaptive Log Filter Method and focusing on detecting chronic kidney disease using advanced deep learning techniques such as a 2D Binary Convolutional Neural Network Algorithm with an Inception V3 Classifier. These algorithms will make use of new technology to improve the precision and promptness of risk assessment, CKD detection, and modern therapy recommendations. The ultimate goal of specialist therapy and early intervention is to enhance patient outcomes and decrease the growth of the disease. Our study yielded very encouraging results, with our suggested method obtaining a remarkable 98% accuracy rate. The effectiveness of our technology in predicting chronic kidney disease problems from medical imaging is demonstrated by this high accuracy level.