Integration of Machine Intelligence and Internet of Things in Smart Healthcare for Voice Pathology Detection (IoT

Abstract

     The use of Artificial Intelligence (AI) together with the Internet of Things (IoT) in smart healthcare represents a promising direction for innovation. Recent progress in Deep Learning techniques has significantly advanced the automation of diagnostic and classification systems. Furthermore, the emergence of 5G wireless networks has provided faster and more reliable data transmission, accelerating the deployment of intelligent healthcare solutions.

 The COVID-19 pandemic highlighted the critical importance of such systems. Voice disorders affect a large number of individuals worldwide, yet early detection can make them highly treatable. In this study, we introduce a smart healthcare framework for the detection of voice pathology. Voice and electroglottography (EGG) signals are recorded using IoT-enabled devices, including microphones and EGG sensors. These signals are converted into Mel-spectrograms and analyzed through pre-trained convolutional neural networks (CNNs). The extracted features are then integrated using a bidirectional LSTM (BiLSTM) network. The proposed framework is evaluated with the Saarbrücken speech database. Experimental results indicate that bimodal inputs provide better performance than unimodal ones, achieving a classification accuracy of 95.65%.