Attaining an IoMT-based health monitoring and prediction: a hybrid hierarchical deep learning model and metaheuristic algorithm

Internet of Medical Things (IoMT) visualizes a network of medical devices and society adopting wireless communications to enable interchange of healthcare data. IoMT is utilized to gather real-time data from medical equipment and sensors. This enables possibility for continuous health monitoring and prediction. There is concern related to potential privacy and safety hazards connected with the group and transmission of sensitive health data over the network. This study proposes a hybrid hierarchical deep learning (DL) model enhanced with features and a metaheuristic algorithm to achieve health monitoring and prediction based on IoMT. The information gained from the analysis helps to identify important features for prediction. The feature selection phase applies Self-regularized Quantum Coronavirus Optimization Algorithm (SQCOA) to prioritize important features for prediction. The prediction phase includes Optimized Long Short-Term Memory (OLSTM) and Hierarchical Convolutional Spiking Neural Network (HCSNN) for feature learning and performance prediction, respectively. The proposed model is simulated by adopting MATLAB. The model attains the highest accuracy of 98%.