Deep Learning-Based Early Diagnosis System for Predicting Chronic Diseases

Abstract

Chronic diseases like the Chronic Kidney Disease (CKD) are silent in nature and therefore hard to diagnose in the early stages. There is a need for early detection so that intervention and treatment can be done effectively. This research proposes a new deep learning-based system which is set to be used in the early diagnosis of chronic diseases. In this case, it focuses on CKD. The system proposed uses a fuzzy DNN in the analysis of routine medical consultation data, to be able to predict and classify CKD at different stages accurately. The deep learning model outperforms the traditional methods having the accuracy rate of 99.23% showing better precision, recall, and F-measure. Whereas current diagnostic approaches depend massively on doctor intervention, this system has the potential to produce meaningful predictions without the need of doctors, thus improving the affordability of early-stage disease diagnosis. The findings show the power of AI in reshaping healthcare through accurate advance alerts that will positively impact the healthcare of chronic diseases, creating room for early medical management