A novel deep fusion strategy for COVID-19 prediction using multimodality approach.

Over the last two years, the novel coronavirus has become a significant threat to the health of the public, and numerous approaches are developed to determine the symptoms of COVID-19. To deal with the complex symptoms of COVID-19, a Deep Learning-assisted Multi-modal Data Analysis (DMDA) approach is introduced to determine COVID-19 symptoms by utilizing acoustic and image-based data. Furthermore, the classified events are forwarded to the proposed Dynamic Fusion Strategy (DFS) for confirming the health status of the individual. Initially, the performance of the proposed solution is evaluated on both acoustic and image-based samples and the proposed solution attains the maximum accuracy of 96.88% and 98.76%, respectively. Similarly, the DFS has achieved an overall symptom determination accuracy of 98.72% which is highly acceptable for decision-making. Moreover, the proposed solution shows high reliability with an accuracy of 95.64% even in absence of any one of the data modalities during testing.

Artificial intelligence-inspired comprehensive framework for Covid-19 outbreak control.

COVID-19 is a life-threatening contagious virus that has spread across the globe rapidly. To reduce the outbreak impact of COVID-19 virus illness, continual identification and remote surveillance of patients are essential. Medical service delivery based on the Internet of Things (IoT) technology backed up by the fog-cloud paradigm is an efficient and time-sensitive solution for remote patient surveillance. Conspicuously, a comprehensive framework based on Radio Frequency Identification Device (RFID) and body-wearable sensor technologies supported by the fog-cloud platform is proposed for the identification and management of COVID-19 patients. The J48 decision tree is used to assess the infection degree of the user based on corresponding symptoms. RFID is used to detect Temporal Proximity Interactions (TPI) among users. Using TPI quantification, Temporal Network Analysis is used to analyze and track the current stage of the COVID-19 spread. The statistical performance and accuracy of the framework are assessed by utilizing synthetically-generated data for 250,000 users. Based on the comparative analysis, the proposed framework acquired an enhanced measure of classification accuracy, and sensitivity of 96.68% and 94.65% respectively. Moreover, significant improvement has been registered for proposed fog-cloud-based data analysis in terms of Temporal Delay efficacy, Precision, and F-measure.

A cognitive framework based on deep neural network for classification of coronavirus disease.

Since December 2019, the pandemic of coronavirus (CorV) is spreading all over the world. CorV is a viral disease that results in ill effects on humans and is recognized as public health concern globally. The objective of the paper is to diagnose and prevent the spread of CorV. Spatio-temporal based fine-tuned deep learning model is used for detecting Corv disease so that the prevention measures could be taken on time. Deep learning is an emerging technique that has an extensive approach to prediction. The proposed system presents a hybrid model using chest X-ray images to early identify the CorV suspected people so that necessary action can be taken timely. The proposed work consists of various deep learning neural network algorithms for the identification of CorV patients. A decision model with enhanced accuracy has been presented for early identification of the suspected CorV patients and time-sensitive decision-making. A SQueezeNet model is used for the classification of the CorV patient. An experiment has been conducted for validation purposes to register an average accuracy of 97.8%. Moreover, the outcomes of statistical parameters are compared with numerous state-of-the-art decision-making models in the current domain for performance assessment.

Internet of things-inspired healthcare system for urine-based diabetes prediction.

Healthcare industry is the leading domain that has been revolutionized by the incorporation of Internet of Things (IoT) technology resulting in smart medical applications. Conspicuously, this study presents an effective system of home-centric Urine-based Diabetes (UbD) monitoring system. Specifically, the proposed system comprises of 4-layers for predicting and monitoring diabetes-oriented urine infection. The system layers including Diabetic Data Acquisition (DDA) layer, Diabetic Data Classification (DDC) layer, Diabetic-Mining and Extraction (DME) layer, and Diabetic Prediction and Decision Making (DPDM) layer allow an individual not exclusively to track his/her diabetes measure on regular basis but the prediction procedure is also accomplished so that prudent steps can be taken at early stages. Additionally, probabilistic measurement of UbD monitoring in terms of Level of Diabetic Infection (LoDI), which is cumulatively quantified as Diabetes Infection Measure (DIM) has been performed for predictive purposes using Recurrent Neural Network (RNN). Moreover, the existence of UbD is visualized based on the Self-Organized Mapping (SOM) procedure. To validate the proposed system, numerous experimental simulations were performed on datasets of 4 individuals. Based on the experimental simulation, enhanced results in terms of temporal delay, classification efficiency, prediction efficiency, reliability and stability were registered for the proposed system in comparison to state-of-the-art decision-making techniques.

Video-assisted smart health monitoring for affliction determination based on fog analytics.

Satisfying the expectations of quality living is essential for smart healthcare. Therefore, the determination of health afflictions in real-time has been considered as one of the most necessary parts of medical or assistive-care domain. In this article, a novel fog analytic-assisted deep learning-enabled physical stance-based irregularity recognition framework is presented to enhance personal living satisfaction of an individual. To increase the utility of the proposed framework for assistive-care, an attempt has been made to record predicted activity scores on cloud by following the continuous time series policy to provide future health references to authorized medical specialist. Furthermore, a smart two-phased decision generation mechanism is proposed to intimate medical specialist and caretakers about the current physical status of an individual in real-time. The generation of the alert is directly proportional to the predicted physical irregularity and the scale of health severity. The experimental results highlight the advantages of fog analytics that helps to increase the recognition rate up to 46.45% for 40 FPS and 45.72% for 30 FPS against cloud-based monitoring solutions. The calculated outcomes justify the superiority of the proposed fog analytics monitoring solution over the conventional cloud-based monitoring solutions by achieving high activity prediction accuracy and less latency rate in decision making.

Cognitive Intelligence Assisted Fog-Cloud Architecture for Generalized Anxiety Disorder (GAD) Prediction.

Generalized Anxiety Disorder (GAD) is a psychological disorder caused by high stress from daily life activities. It causes severe health issues, such as sore muscles, low concentration, fatigue, and sleep deprivation. The less availability of predictive solutions specifically for individuals suffering from GAD can become an imperative reason for health and psychological adversity. The proposed solution aims to monitor health, behavioral and environmental parameters of the individual to predict health adversity caused by GAD. Initially, Weighted-Naïve Bayes (W-NB) classifier is utilized to predict irregular health events by classifying the captured data at the fog layer. The proposed two-phased decision-making process helps to optimize the distribution of required medical services by determining the scale of vulnerability. Furthermore, the utility of the framework is increased by calculating health vulnerability index using Adaptive Neuro-Fuzzy Inference System-Genetic Algorithm (ANFIS-GA) on the cloud. The presented work addresses the concerns in terms of efficient monitoring of anomalies followed by time sensitive two-phased alert generation procedure. To approve the performance of irregular event identification and health severity prediction, the framework has been conveyed in a living room for 30 days in which almost 15 individuals by the age of 68 to 78 years have been continuously monitored. The calculated outcomes represent the monitoring efficiency of the proposed framework over the policies of manual monitoring.

Temporal Informative Analysis in Smart-ICU Monitoring: M-HealthCare Perspective.

The rapid introduction of Internet of Things (IoT) Technology has boosted the service deliverance aspects of health sector in terms of m-health, and remote patient monitoring. IoT Technology is not only capable of sensing the acute details of sensitive events from wider perspectives, but it also provides a means to deliver services in time sensitive and efficient manner. Henceforth, IoT Technology has been efficiently adopted in different fields of the healthcare domain. In this paper, a framework for IoT based patient monitoring in Intensive Care Unit (ICU) is presented to enhance the deliverance of curative services. Though ICUs remained a center of attraction for high quality care among researchers, still number of studies have depicted the vulnerability to a patient's life during ICU stay. The work presented in this study addresses such concerns in terms of efficient monitoring of various events (and anomalies) with temporal associations, followed by time sensitive alert generation procedure. In order to validate the system, it was deployed in 3 ICU room facilities for 30 days in which nearly 81 patients were monitored during their ICU stay. The results obtained after implementation depicts that IoT equipped ICUs are more efficient in monitoring sensitive events as compared to manual monitoring and traditional Tele-ICU monitoring. Moreover, the adopted methodology for alert generation with information presentation further enhances the utility of the system.