COVID-19 spread control policies based early dynamics forecasting using deep learning algorithm.

Many severe epidemics and pandemics have hit human civilizations throughout history. The recent Sever Actuate Respiratory disease SARS-CoV-2 known as COVID-19 became a global disease and is still growing around the globe. It has severely affected the world's economy and ways of life. It necessitates predicting the spread in advance and considering various control policies to avoid the country's complete closure. In this paper, we propose deep learning-based stacked Bi-directional long short-term memory (Stacked Bi-LSTM) network that forecasts COVID-19 more accurately for the country of South Korea. The paper's main objectives are to present a lightweight, accurate, and optimized model to predict the spread considering restriction policies such as school closure, workspace closing, and the canceling of public events. Based on the fourteen parameters (including control policies), we predict and forecast the future value of the number of positive, dead, recovered, and quarantined cases. In this paper, we use the dataset of South Korea comprised of several control policies implemented for minimizing the spread of COVID-19. We compare the performance of the stacked Bi-LSTM with the traditional time-series models and LSTM model using the performance metrics mean absolute error (MAE), mean absolute percentage error (MAPE), and root mean square error (RMSE). Moreover, we study the impact of control policies on forecasting accuracy. We further study the impact of changing the Bi-LSTM default activation functions Tanh with ReLU on forecasting accuracy. The research provides insight to policymakers to optimize the pooling of resources more optimally on the correct date and time prior to the event and to control the spread by employing various strategies in the meantime.

Attention based parameter estimation and states forecasting of COVID-19 pandemic using modified SIQRD Model.

In this work, we propose a new mathematical modeling of the spread of COVID-19 infection in an arbitrary population, by modifying the SIQRD model as m-SIQRD model, while taking into consideration the eight governmental interventions such as cancellation of events, closure of public places etc., as well as the influence of the asymptomatic cases on the states of the model. We introduce robustness and improved accuracy in predictions of these models by utilizing a novel deep learning scheme. This scheme comprises of attention based architecture, alongside with Generative Adversarial Network (GAN) based data augmentation, for robust estimation of time varying parameters of m-SIQRD model. In this regard, we also utilized a novel feature extraction methodology by employing noise removal operation by Spline interpolation and Savitzky-Golay filter, followed by Principal Component Analysis (PCA). These parameters are later directed towards two main tasks: forecasting of states to the next 15 days, and estimation of best policy encodings to control the infected and deceased number within the framework of data driven synergetic control theory. We validated the superiority of the forecasting performance of the proposed scheme over countries of South Korea and Germany and compared this performance with 7 benchmark forecasting models. We also showed the potential of this scheme to determine best policy encodings in South Korea for 15 day forecast horizon.

Vulnerability- and Diversity-Aware Anonymization of Personally Identifiable Information for Improving User Privacy and Utility of Publishing Data.

Personally identifiable information (PII) affects individual privacy because PII combinations may yield unique identifications in published data. User PII such as age, race, gender, and zip code contain private information that may assist an adversary in determining the user to whom such information relates. Each item of user PII reveals identity differently, and some types of PII are highly identity vulnerable. More vulnerable types of PII enable unique identification more easily, and their presence in published data increases privacy risks. Existing privacy models treat all types of PII equally from an identity revelation point of view, and they mainly focus on hiding user PII in a crowd of other users. Ignoring the identity vulnerability of each type of PII during anonymization is not an effective method of protecting user privacy in a fine-grained manner. This paper proposes a new anonymization scheme that considers the identity vulnerability of PII to effectively protect user privacy. Data generalization is performed adaptively based on the identity vulnerability of PII as well as diversity to anonymize data. This adaptive generalization effectively enables anonymous data, which protects user identity and private information disclosures while maximizing the utility of data for performing analyses and building classification models. Additionally, the proposed scheme has low computational overheads. The simulation results show the effectiveness of the scheme and verify the aforementioned claims.

N-screen aware multicriteria hybrid recommender system using weight based subspace clustering.

This paper presents a recommender system for N-screen services in which users have multiple devices with different capabilities. In N-screen services, a user can use various devices in different locations and time and can change a device while the service is running. N-screen aware recommendation seeks to improve the user experience with recommended content by considering the user N-screen device attributes such as screen resolution, media codec, remaining battery time, and access network and the user temporal usage pattern information that are not considered in existing recommender systems. For N-screen aware recommendation support, this work introduces a user device profile collaboration agent, manager, and N-screen control server to acquire and manage the user N-screen devices profile. Furthermore, a multicriteria hybrid framework is suggested that incorporates the N-screen devices information with user preferences and demographics. In addition, we propose an individual feature and subspace weight based clustering (IFSWC) to assign different weights to each subspace and each feature within a subspace in the hybrid framework. The proposed system improves the accuracy, precision, scalability, sparsity, and cold start issues. The simulation results demonstrate the effectiveness and prove the aforementioned statements.

A network and visual quality aware N-screen content recommender system using joint matrix factorization.

We propose a network and visual quality aware N-Screen content recommender system. N-Screen provides more ways than ever before to access multimedia content through multiple devices and heterogeneous access networks. The heterogeneity of devices and access networks present new questions of QoS (quality of service) in the realm of user experience with content. We propose, a recommender system that ensures a better visual quality on user's N-screen devices and the efficient utilization of available access network bandwidth with user preferences. The proposed system estimates the available bandwidth and visual quality on users N-Screen devices and integrates it with users preferences and contents genre information to personalize his N-Screen content. The objective is to recommend content that the user's N-Screen device and access network are capable of displaying and streaming with the user preferences that have not been supported in existing systems. Furthermore, we suggest a joint matrix factorization approach to jointly factorize the users rating matrix with the users N-Screen device similarity and program genres similarity. Finally, the experimental results show that we also enhance the prediction and recommendation accuracy, sparsity, and cold start issues.

Erratum to "a network and visual quality aware N-screen content recommender system using joint matrix factorization".

[This corrects the article DOI: 10.1155/2014/806517.].

Mobile sensors based platform of Human Physical Activities Recognition for COVID-19 spread minimization.

The development of smartphones technologies has determined the abundant and prevalent computation. An activity recognition system using mobile sensors enables continuous monitoring of human behavior and assisted living. This paper proposes the mobile sensors-based Epidemic Watch System (EWS) leveraging the AI models to recognize a new set of activities for effective social distance monitoring, probability of infection estimation, and COVID-19 spread prevention. The research focuses on user activities recognition and behavior concerning risks and effectiveness in the COVID-19 pandemic. The proposed EWS consists of a smartphone application for COVID-19 related activities sensors data collection, features extraction, classifying the activities, and providing alerts for spread presentation. We collect the novel dataset of COVID-19 associated activities such as hand washing, hand sanitizing, nose-eyes touching, and handshaking using the proposed EWS smartphone application. We evaluate several classifiers such as random forests, decision trees, support vector machine, and Long Short-Term Memory for the collected dataset and attain the highest overall classification accuracy of 97.33%. We provide the Contact Tracing of the COVID-19 infected person using GPS sensor data. The EWS activities monitoring, identification, and classification system examine the infection risk of another person from COVID-19 infected person. It determines some everyday activities between COVID-19 infected person and normal person, such as sitting together, standing together, or walking together to minimize the spread of pandemic diseases.

Dynamic adhesive force measurements under vertical and horizontal motions of interacting rough surfaces.

An instrument to measure dynamic adhesive forces between interacting rough surfaces has been developed. It consists of four parts, namely, main instrument body, vertical positioning system with both micrometer and nanometer positioning accuracies, horizontal positioning system with nanometer positioning accuracy, and custom-built high-resolution, and high dynamic bandwidth capacitive force transducer. The vertical piezoelectric actuator (PZT) controls the vertical (approaching and retracting) motion of the upper specimen, while the horizontal PZT controls the horizontal (reciprocal) motion of the lower specimen. The force transducer is placed in line with the upper specimen and vertical PZT, and directly measures the adhesive forces with a root-mean-square load resolution of 1.7 microN and a dynamic bandwidth of 1.7 kHz. The newly developed instrument enables reliable measurements of near-contact and contact adhesive forces for microscale devices under different dynamic conditions. Using the developed instrument, dynamic pull-in and pull-off force measurements were performed between an aluminum-titanium-carbide sphere and a 10 nm thick carbon film disk sample. Three different levels of contact force were investigated; where for each contact force level the vertical velocity of the upper sample was varied from 0.074 to 5.922 microms, while the lower sample was stationary. It was found that slower approaching and retracting velocities result in higher pull-in and pull-off forces. The noncontact attractive force was also measured during horizontal movement of the lower sample, and it was found that the periodic movements of the lower disk sample also affect the noncontact surface interactions.

A multicriteria adaptive opportunistic treecast routing protocol for multimedia dissemination in vehicle-to-vehicle telescreen.

This paper presents vehicle-to-vehicle telescreen (VVT) and a multicast scheme to disseminate digital signage multimedia services to vehicular ad hoc networks (VANETs). Multimedia dissemination in VANETs is challenging because of the high packet losses (PLs), delays and longer disconnection times, which degrade the network quality of service (QoS) and user quality of experience (QoE). To reduce the PLs and delays, most existing multimedia multicast schemes in VANETs primarily select routes based on longer route expiration times (RET) or lowest path delays. The RET-based schemes suffer less from PLs when there are fewer active multicastings in the network. When the number of active multicastings increases, delay-based schemes suffer less from PLs comparatively. This tradeoff implies to design an adaptive mechanism by mutually complementing the RET-based and delay-based schemes to reduce PLs and delays. In this paper, we propose a multicriteria adaptive opportunistic treecast routing protocol (MAOTRP), which adapts the route selection mechanism according to active multicastings for efficient multimedia dissemination in VVT. The MAOTRP adjusts the weights of route selection parameters, including RET and delays, by considering their contribution in improving packet delivery ratio. MAOTRP extends a tree-based multicast protocol to provide robustness through alternate routes for link failures to reduce PLs. Through several experimental evaluations, we show that the proposed dissemination scheme improves QoS and QoE, and reduces the average disconnection time.

Differential effects of Fas cross-linking on phospholipase D activation and related lipid metabolism in Fas-resistant A20 cells.

A20 murine lymphoma cells undergoing Fas-mediated apoptosis showed increase in the activity of phospholipase D (PLD), which is involved in proliferative or mitogenic cellular responses. Using A20 cell lines that were resistant to Fas-induced apoptosis, we investigated the differential effects of Fas cross-linking on PLD activity and sphingolipid metabolism. The basal PLD activities in all of the selected three Fas-resistant clones (#5, #8, and #11) were about 2~4 folds higher than that of wild type A20 cells. Among the PLD isoforms, PLD2 expression was increased in all of the selected Fas-resistant clones. The Fas downstream signaling events triggered by Fas cross-linking, including the activations of PLD, phosphatidylcholine-specific phospholipase C (PC-PLC), sphingomyelinase (SMase), the increase in diacylglycerol (DAG) and protein phosphorylation levels, and the translocation of protein kinase C to membrane were not changed in both of Fas-resistant clone #5 and #8. In contrast, Fas cross-linking stimulated the activity of PLD, PC-PLC, and SMase, translocation of PKC, and protein phosphorylation in Fas-resistant clone #11, similar to that of wild type cells. We also found that clone #11 had a different Fas sequence encoding Fas B which has been known to inhibit Fas-induced apoptosis. These findings suggest that increased PLD2 expression resulting in increased basal PLD activity and the blockade of Fas downstream signaling cascades may be involved to limit apoptosis induced by Fas cross-linking.

Modulation of cyclooxygenase-2 expression by phosphatidylcholine specific phospholipase C and D in macrophages stimulated with lipopolysaccharide.

Lipopolysaccharide (LPS) enhances the expression of cyclooxygenase 2 (COX-2) in macrophages, and stimulates production of prostaglandins that cause endothelial dysfunction in septic shock. In an effort to identify strategies for reducing LPS-inducible expression of COX-2, inhibitors of the phospholipases involved in LPS dependent over-expression of COX-2 were studied. LPS enhances expression of COX-2 mRNA and protein by activating sequentially phosphatidylcholine-specific phospholipase C (PC-PLC), protein kinase C (PKC) and phosphatidylcholine-specific phospholipase D (PC-PLD). This stimulates production of phosphatidic acid (PA), which increases expression of COX-2 mRNA and protein. Inhibition of PC-PLC by D609 (tricyclodecanoyl xanthogenate), and of PC-PLD activity by 1-butanol, reduced LPS-dependent over-production of PA and suppressed the increase of COX-2 mRNA and protein. Activation of PKC, normally seen in LPS-treated cells, was mimicked with phorbol myristic acid (PMA), and this also increased PA production and enhanced COX-2 expression. Propranolol inhibition of phosphatidic acid phosphohydrolase (PPH) increased PA accumulation and enhanced LPS-dependent COX-2 protein synthesis. These results suggest that inhibitors of PC-PLC, PKC and PC-PLD, or activators of PPH could be useful in the management of LPS-induced overproduction of prostaglandins and of vascular dysfunction in septic shock.

Role of phospholipase D2 in anti-apoptotic signaling through increased expressions of Bcl-2 and Bcl-xL.

We have previously reported that Fas-resistant A20 cells (FasR) have phospholipase D (PLD) activity upregulated by endogenous PLD2 overexpression. In the present study, we investigated how overexpressed PLD2 in FasR could generate survival signals by regulating the protein levels of anti-apoptotic Bcl-2 and Bcl-xL. To confirm the effect of PLD2 on Bcl-2 protein levels, we transfected PLD2 into wild-type murine B lymphoma A20 cells. The transfected cells showed markedly the increases in Bcl-2 and Bcl-xL protein levels, and became resistant to Fas-induced apoptosis, similar to FasR. Treatment of wild-type A20 cells with phosphatidic acid (PA), the metabolic end product of PLD2 derived from phosphatidylcholin, markedly increased levels of anti-apoptotic Bcl-2 and Bcl-xL proteins. Moreover, PA-induced expressions of Bcl-2 and Bcl-xL were enhanced by propranolol, an inhibitor of PA phospholydrolase (PAP), whereas completely blocked by mepacrine, an inhibitor of phospholipase A(2) (PLA(2)), suggesting that PLA(2) metabolite of PA is responsible for the increases in Bcl-2 and Bcl-xL protein levels. We further confirmed the involvement of arachidonic acid (AA) in PA-induced survival signals by showing that 1,2-dipalmitoyl-sn-glycero-3-phosphate (DPPA), PA without AA, was unable to increase Bcl-2 and Bcl-xL proteins. Moreover, PA notably increased cyclooxygenase (COX)-2 protein expression, and PA-induced expression of both Bcl-2 and Bcl-xL was inhibited by NS-398, a specific inhibitor of COX-2. Taken together, these findings demonstrate that PA generated by PLD2 plays an important role in cell survival during Fas-mediated apoptosis through the increased Bcl-2 and Bcl-xL protein levels which resulted from PLA(2) and AA-COX2 pathway.