Automated Disease Diagnosis and Monitoring System Using Deep Learning in IoT Healthcare Applications

Dr. Mrunali Sonwalkar

Abstract: Technology has become a fundamental aspect of modern life, enabling the automation of previously labor-intensive processes and making them easier and more efficient worldwide. Artificial Intelligence (AI), the Internet of Things (IoT), blockchain, and Deep Learning (DL) have all made significant contributions to the advancement of several industries. In the medical industry, where exact diagnoses are required daily, traditional approaches are often inefficient due to errors and excessive delays. A smart healthcare system, on the other hand, can be constructed using recent advances in DL and IoT, which are useful for providing accurate diagnoses promptly. This study focuses on an automated disease diagnosis and monitoring system that employs advanced technologies, including DL and IoT. The dataset, based on patients' health parameters, is generated via a Google search. This dataset is used to test disease prediction results using DL models such as Convolutional Neural Network (CNN), Multi-Layer Perceptron (MLP), and Bidirectional Long Short-Term Memory (Bi-LSTM). The Bi-LSTM model is identified as the best model, with 99% accuracy. The Bi-LSTM is then deployed in the cloud. The health parameters of individuals are acquired using wearable sensors. The data is transmitted to the cloud, where the model analyzes it and predicts the individual's health status. If an irregularity is discovered, the cloud sends an alert message to the authorized doctor.

Keywords: Healthcare, Internet of Things, Deep Learning, Disease, Sensor, Performance Measure