Hybrid LSTM-RNN and Lion Optimization Algorithm for IoT-based Proactive Healthcare Data Management

In the realm of healthcare, the integration of Internet of Things (IoT) technology has revolutionized patient monitoring, allowing for real-time data collection and analysis. This abstract introduces a novel approach to patient monitoring utilizing a hybrid Long Short-Term Memory-Recurrent Neural Network (LSTM-RNN) architecture combined with Lion Optimization for enhanced feature selection. The proposed system aims to predict and monitor patient health conditions, particularly focusing on cardiac health. By harnessing IoT devices for continuous data acquisition, vital signs such as heart rate, blood pressure, and oxygen saturation levels are collected in real-time. The hybrid LSTM-RNN model is employed to analyze this data, leveraging its ability to capture temporal dependencies and patterns in sequential data. Furthermore, Lion Optimization, inspired by the hunting behavior of lions, is utilized for feature selection to enhance the predictive accuracy of the model. This optimization technique intelligently selects the most relevant features from the collected data, thereby improving the efficiency and effectiveness of the predictive model. The combination of IoT-based patient monitoring, hybrid LSTM-RNN architecture, and Lion Optimization offers a accuracy of 99.99 %, precision of $\mathbf{9 7 \%}$, recall of $\mathbf{9 6 . 0 6 \%}$, and f1 measure of $\mathbf{9 6 . 5 3 \%}$. By providing early detection and prediction of health issues, this approach facilitates timely interventions and personalized treatment plans, ultimately enhancing patient outcomes and quality of life.