Enhancing Swine Flu Management: A Framework Integrating Deep learning and Fog Computing

Abstract

Swine flu, a prevalent viral infection with global significance, poses a serious public health concern, especially in countries like India where the number of cases continues to rise annually. Traditional disease detection methods are time-consuming and labour-intensive, highlighting the need for innovative approaches leveraging technology. This paper proposes a comprehensive framework that integrates artificial neural networks (ANNs) for diagnosis and fog-centric Internet of Things (IoT) for real-time monitoring and control of swine flu outbreaks. The first part of the framework focuses on leveraging ANNs for pig influenza diagnosis. ANNs are utilized to analyze clinical and laboratory data, providing an efficient and accurate means of identifying positive cases. The model employs Hybrid-ANN to select pertinent attributes, optimizing the training process and enhancing diagnostic accuracy. In parallel, the paper introduces a fog-centric IoT-based smart healthcare support system tailored for swine flu epidemic management. By harnessing fog computing, health data processing is expedited, enabling timely decision-making. A hybrid classifier is employed to classify swine flu patients at early stages, generating alerts for health officials and patients' guardians. Experimental evaluations demonstrate the efficacy of the proposed framework. Results indicate improved network bandwidth reliability, operational efficiency, and reduced response times compared to traditional cloud-only models. By integrating AI-driven diagnosis with fog computing-enabled monitoring, this research contributes to early detection and proactive management of swine flu outbreaks, ultimately enhancing public health preparedness and response capabilities.