Autonomous Smart Device for COVID-19 Detection Using Artificial Intelligence

Corona is a pandemic disease and is spreading all over the world. There is also lack of corona virus detection machines. If it is detected at very early stages without pathological intervention, then further spreading of the disease can be controlled, and many of human lives can be saved. So, the proposed biomedical device can be used for fast and accurate prediction of COVID-19 from chest x-rays. X-ray can also be taken from anywhere and sent through any communication medium. Even if error is added, it can be removed using error concealment algorithms. Automated AI-based systems will be used for prediction of normal, COVID-19, and pneumonic cases from x-ray images. It makes detection of COVID-19 infection less costly and portable. This device can be stored in less stringent conditions, making it more effective.

NTAM-LSTM models of network traffic prediction

Accurate prediction of network traffic is very important in allocating network resources. With the rapid development of network technology, network traffic becomes more complex and diverse. The traditional network traffic prediction model cannot accurately predict the current network traffic within the effective time. This paper proposes a Network Traffic Prediction Model----NTAM-LSTM, which based on Attention Mechanism with Long and Short Time Memory. Firstly, the model preprocesses the historical dataset of network traffic with multiple characteristics. Then the LSTM network is used to make initial prediction for the processed dataset. Finally, attention mechanism is introduced to get more accurate prediction results. Compared with other network traffic prediction models, NTAM-LSTM prediction model can achieve higher prediction accuracy and take shorter running time.

Role of upwind precipitation in transboundary pollution and secondary aerosol formation: A case study during the KORUS-AQ field campaign

Clouds and precipitation play critical roles in wet removal of aerosols and soluble gases in the atmosphere, and hence their accurate prediction largely influences accurate prediction of air pollutants. In this study, the impacts of clouds and precipitation on wet scavenging and long-range transboundary transport of pollutants are examined during the 2016 Korea-United States Air Quality (KORUS-AQ) field campaign using the Weather Research and Forecasting model coupled with chemistry. Two simulations in which atmospheric moisture is constrained vs. it is not are performed and evaluated against surface and airborne observations. The simulation with moisture constraints is found to better reproduce precipitation as well as surface PM2.5, whereas the areal extent and amount of precipitation are overpredicted in the simulation without moisture constraints. As a results of overpredicted clouds and precipitation and consequently overpredicted wet scavenging, PM2.5 concentration is generally underpredicted across the model domain in the simulation without moisture constraints. The effects are significant not only in the precipitating region (upwind region, southern China in this study) but also in the downwind region (South Korea) where no precipitation is observed. The difference in upwind precipitation by 77% on average between the two simulations leads to the difference in PM2.5 by ∼39% both in the upwind and downwind regions. The transboundary transport of aerosol precursors, especially nitric acid, has a considerable impact on ammonium-nitrate aerosol formation in the ammonia-rich downwind region. This study highlights that skillful prediction of atmospheric moisture can have ultimate potential to skillful prediction of aerosols across regions.