Fuzzy-Centric Fog–Cloud Inspired Deep Interval Bi-LSTM Healthcare Framework for Predicting Yellow Fever Outbreak

Yellow fever is a vigorous, phlebotomic, vector-borne disease that poses a significant public health threat in regions with high mosquito density and inadequate vaccination coverage. The disease's toxic phase is lethal, making prompt identification and control measures crucial. The emergence of the latest technologies and data analytics techniques, such as edge-cloud computing, data analytics, and machine learning/deep learning, has played a pivotal role in revolutionizing remote healthcare services. Henceforth, applying the abovementioned technologies leads to improvements in the response time, service quality, and location awareness of healthcare systems. Relative to this context, we propose an intelligent fuzzy-centric fog–cloud-assisted healthcare framework to identify and control yellow fever epidemics. Initially, at the fog layer, singular value decomposition is used for data dimensionality reduction analysis and the Fuzzy-C mean clustering (FCM) algorithm is leveraged to get rigorous results. Moreover, for better results and to focus on time-series patterns, the deep interval type 2 fuzzy Bi-LSTM model is proposed at the cloud layer to generate a yellow fever severity index and visualize each yellow fever region based on self-organized maps. In addition, we propose an alert generation mechanism to facilitate real-time decision-making. Finally, results show that the proposed system yields significant efficacy, compared with other state-of-the-art methodologies.