Security Threats and Artificial Intelligence Based Countermeasures for Internet of Things Networks: A Comprehensive Survey

The Internet of Things (IoT) has emerged as a technology capable of connecting heterogeneous nodes/objects, such as people, devices, infrastructure, and makes our daily lives simpler, safer, and fruitful. Being part of a large network of heterogeneous devices, these nodes are typically resource-constrained and became the weakest link to the cyber attacker. Classical encryption techniques have been employed to ensure the data security of the IoT network. However, high-level encryption techniques cannot be employed in IoT devices due to the limitation of resources. In addition, node security is still a challenge for network engineers. Thus, we need to explore a complete solution for IoT networks that can ensure nodes and data security. The rule-based approaches and shallow and deep machine learning algorithms- branches of Artificial Intelligence (AI)- can be employed as countermeasures along with the existing network security protocols. This paper presented a comprehensive layer-wise survey on IoT security threats, and the AI-based security models to impede security threats. Finally, open challenges and future research directions are addressed for the safeguard of the IoT network.

COVID-19 Patient Outcome Prediction Using Selected Features from Emergency Department Data and Feed-Forward Neural Networks

The severity of COVID-19 varies dramatically, ranging from asymptomatic infection to severe respiratory failure and death. Currently, few prognostic markers for disease outcomes exist, impairing patient triaging and treatment. Here, we train feed-forward neural networks on electronic health records of 819 confirmed SARS-CoV-2 positive patients admitted to a two-site NHS Trust hospital in London, England. To allow early risk assessment, the models ingest data collected in the emergency department (ED) to predict subsequent admission to intensive care, need for mechanical ventilation and in-hospital mortality. We apply univariate selection and recursive feature elimination to find the minimal subset of clinical variables needed for accurate prediction. Our models achieve AUC-ROC scores of 0.78 to 0.87, outperforming standard clinical risk scores. This accuracy is reached with as few as 13% of clinical variables routinely collected within the ED, which increases the practical applicability of such algorithms. Hence, state-of-the-art neural networks can predict severe COVID-19 accurately and early from a small subset of clinical variables. © 2021, ICST Institute for Computer Sciences, Social Informatics and Telecommunications Engineering.