FDLM: Fusion Deep Learning Model for Classifying Obstructive Sleep Apnea and Type 2 Diabetes

Abstract

This research paper proposes a Fusion model, which is based on an ensemble majority vote classification, which includes several hidden-layers (BPNN, MP, AS, and SSTM) for classifying Obstructive Sleep Apnea and Diabetes. This paper aims to increase the accuracy in classifying Obstructive Sleep Apnea and Type 2 Diabetes through a Convolutional Neural Network (CNN) and Deep Belief Networks (DBN) using Shifted Filter Responses by identify deep learning features and to reduce the computational time. The experiments are carried out using datasets consisting of attributes of Obstructive Sleep Apnea and Diabetes. The experimental results indicate that the findings are improved than the previous model using the proposed Fusion Deep Learning Model.