ABSTRACT
In the area of healthcare applications such as classification, illness prediction, etc., there is a growing emphasis placed on the categorization of medical data. In addition to learning, neural systems also have other advantageous traits including poor or absent data management, such as the capacity to separate noise, vulnerability, or imprecision. Hence the significance of feature selection is that it decreases the classifier capacity to the measurements that are considered generally pertinent in precise classification. The primary objective of this work is to arrange the medical data and investigate the viability of using distinctive input features and classifiers to find the medical datasets. This work proposed deep neural network (DNN) for classification. From the result outcome, it is observed that the proposed DNN classifier produces higher accuracy, sensitivity, and specificity rates than machine learning (ML)-based classification algorithms with respect to chronic kidney disease (CKD) dataset.
