Enhancing medical response efficiency in real-time large crowd environments via smart coverage and deep learning for stable ecological health monitoring.

Festivals and city-wide mass events are prevalent in human societies worldwide, drawing large crowds. Such events range from concerts with a dozen attendees to large-scale actions with thousands of viewers. It is the highest priority for each organizer of such an occasion to be capable of upholding a higher standard of safety and minimizing the danger of events, especially medical emergencies. Therefore, establishing sufficient safety measures is significant. There is a requirement for event organizers and emergency response personnel to identify developing, potentially critical crowd situations at an early stage during city-wide mass assemblies. In general, the localization of the global positioning system (GPS) and proximity-based tracking is employed to capture intricate crowd dynamics throughout an event. Recently, technology has been used in numerous diverse ways to achieve these large crowds. For example, computer vision-based models are employed to observe the flexibility and behaviour of crowds. In this manuscript, a model for Medical Response Efficiency in Real-Time Large Crowd Environments via Smart Coverage and Hiking Optimisation (MRELC-SCHO) is presented, aiming to maintain stable ecological health. The primary objective of this paper is to propose an effective method for enhancing medical response efficiency in large crowd environments by utilizing advanced optimization algorithms. Initially, the MRELC-SCHO model utilizes min-max normalization to transform the input data into a structured format. Furthermore, the Chimp Optimisation Algorithm (CHOA) model is employed for the feature selection (FS) process to select the most significant features from the dataset. Additionally, the MRELC-SCHO technique utilizes the bidirectional long short-term memory with an auto-encoder (BiLSTM-AE) method for classification. Finally, the parameter selection for the BiLSTM-AE model is performed by using the Hiking Optimisation Algorithm (HOA) model. The experimentation of the MRELC-SCHO approach is accomplished under the Ecological Health dataset. The comparison analysis of the MRELC-SCHO approach revealed a superior accuracy value of 98.56% compared to existing models.