Applied convolutional neural network framework for tagging healthcare systems in crowd protest environment
Healthcare systems constitute a significant portion of smart cities infrastructure. The aim of smart healthcare is two folds. The internal healthcare system has a sole focus on monitoring vital parameters of patients. The external systems provide proactive health care measures by the surveillance me...
Guardado en:
| Autores principales: | , , |
|---|---|
| Formato: | article |
| Lenguaje: | EN |
| Publicado: |
AIMS Press
2021
|
| Materias: | |
| Acceso en línea: | https://doaj.org/article/f9a1dd05239443b29fd92f226f5857ff |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| Sumario: | Healthcare systems constitute a significant portion of smart cities infrastructure. The aim of smart healthcare is two folds. The internal healthcare system has a sole focus on monitoring vital parameters of patients. The external systems provide proactive health care measures by the surveillance mechanism. This system utilizes the surveillance mechanism giving impetus to healthcare tagging requirements on the general public. The work exclusively deals with the mass gatherings and crowded places scenarios. Crowd gatherings and public places management is a vital challenge in any smart city environment. Protests and dissent are commonly observed crowd behavior. This behavior has the inherent capacity to transform into violent behavior. The paper explores a novel and deep learning-based method to provide an Internet of Things (IoT) environment-based decision support system for tagging healthcare systems for the people who are injured in crowd protests and violence. The proposed system is intelligent enough to classify protests into normal, medium and severe protest categories. The level of the protests is directly tagged to the nearest healthcare systems and generates the need for specialist healthcare professionals. The proposed system is an optimized solution for the people who are either participating in protests or stranded in such a protest environment. The proposed solution allows complete tagging of specialist healthcare professionals for all types of emergency response in specialized crowd gatherings. Experimental results are encouraging and have shown the proposed system has a fairly promising accuracy of more than eight one percent in classifying protest attributes and more than ninety percent accuracy for differentiating protests and violent actions. The numerical results are motivating enough for and it can be extended beyond proof of the concept into real time external surveillance and healthcare tagging. |
|---|