Machine Learning-Backed Planning of Rapid COVID-19 Tests With Autonomous Vehicles With Zero-Day Considerations

The COVID-19 pandemic has stretched public health resources to the limits, and the only realistic way to keep the infection rates low is effective testing to prevent community transmission. In this research study, we propose an innovative method to empower autonomous vehicle-driven mobile assessment facilities to support early detection of the cases contracted with the virus, and enable early detection of sources for hot spots. We describe a self organizing feature map (SOFM) approach to the allocation of the mobile assessment centers, and also use the same method to determine the travel route of the autonomous vehicles, and provide critical decision support to the supply chain manager. Our results reveal that the optimal number of neurons under varying test times can be obtained by 5 different zero-day coordinates of initially contracted cases and worst-case scenario to find out the contracted cases in 17 days and 27 days under two different test time scenarios.

A Novel Machine Learning-Assisted Policy Recommendation Method on COVID-19 Vaccination Campaign

As the most serious global infectious disease in the past 100 years, it has caused severe loss of life and property to countries and their people worldwide in the past year. As the most powerful tool in the fight against the epidemic, how to quickly promote the COVID-19 vaccine administration plays a vital role in gradually establishing an immune barrier in the population as soon as possible and blocking the COVID-19 epidemic. In this paper, we provide a machine learning-based policy recommendation method on the vaccination campaign of COVID-19 by minimizing three different cost factors: the duration of the pandemic, the budget of the COVID-19 battle as well as the death toll. To generate a more efficient vaccination policy, we construct an Age-stratified Susceptible-Infected-Recovered (ASSIR) model. We validate our method based on the real-world dataset of India by comparing our simulated results with the government's vaccination plan from machine learning prediction. Our approach shows a 13% decrease in disease control time and government budget. At the same time, we find out that vaccination based on each province's population leads to a 12.4% decrease in the death toll than on infection cases. The model developed in this study has practical implications for COVID-19 vaccination campaigns and the infection control of other infectious diseases. © 2021 IEEE.

A Novel Statistical and Neural Network Combined Approach for the Cloud Spot Market

The instance price in the Amazon EC2 spot model is often much lower than in the on-demand counterpart. However, this price reduction comes with a decrease in the availability guarantees. To our knowledge, there is no work that accurately captures the short-term trade-off between spot price and availability, and does long-term analysis for spot price tendencies in favor of user decision making. In this work, we propose a utility-based strategy, that balances cost and availability of spot instances and is targeted to short-term analysis;and a LSTM neural network framework for long term spot price tendency analysis. Our experiments show that, for r4.2xlarge, 90% of spot bid suggestions ensured at least 5.73 hours of availability, with a bid price of approximately 38% of the on-demand price. The LSTM experiments predicted spot price tendencies for several instance types with low error. Our LSTM framework predicted an average value of 0.19 USD/hour for the r5.2xlarge instance type, which is about 37% of the on-demand price. Finally, we used our combined mechanism on an application that compares thousands of SARS-CoV-2 sequences and show that our approach is able to provide good choices of instances, with low bids and very good availability. IEEE

AI-driven autonomous vehicles as COVID-19 assessment centers: A novel crowdsensing-enabled strategy

Internet of Things(IoT) facilitates key technologies that rely on sensing, communication and processing in daily routines. As an IoT-enabled paradigm, mobile crowdsensing (MCS) can offer more possibilities for data collection to support various IoT applications and services. As an extension, MCS can be used for data gathering amid COVID-19 pandemic crisis. Bridging Artificial Intelligence and IoT can achieve not only maintaining low infection rates of COVID-19 but can also facilitate an effective rapid testing strategy to reduce community spread. In this research, an intelligent strategy to deploy autonomous vehicle-based mobile testing facilities is proposed to enable early detection of infected cases based upon MCS data acquired through smart devices via wireless communications such as Wifi, LTE and 5G. To this end, a Self Organizing Feature Map is designed to manage MCS-based data for planning of the autonomous mobile assessment centers. Pre-identified zero-day locations and worst-case scenario are considered to determine the best combination for MCS participation rate and budget limitations. Numerical results demonstrate that once 30% of MCS participants are recruited, it becomes possible to cover the pre-identified zero-day locations and enable detection of infected cases under the worst case scenario to determine the AV routes more efficiently than other options for a certain number of neurons in SOFM. The worst-case scenario demonstrates that 30% participant rate ensures detection of infected cases in 27 days for 81 stops even infected cases are outside of the autonomous vehicle testing coverage. © 2021 Elsevier B.V.