A Deep Learning-based Architecture for Diabetes Detection, Prediction, and Classification

This study examines the importance of Deep Learning (DL) in the Internet of Medical Things (IoMT) in providing impactful results in the diagnosis, classification, prediction, and categorization of stages of diabetes. A DL model was used to classify diabetic retinopathy data, based on a Multi-Layer Feed-Forward Neural Network (MLFNN). The Pima Diabetes Dataset (PDD) was used to train and test the proposed model. To increase accuracy, this study considered different activation functions and strategies to deal with lost information. The proposed Multilayer Feed-Forward Neural Network (MLFNN) model was compared with conventional Machine Learning (ML) approaches, specifically Random Forest (RF) and Naive Bayes (NB), outperforming them with a significant increase in classification accuracy.