Security Investment in Contagious Networks.

Security of the systems is normally interdependent in such a way that security risks of one part affect other parts and threats spread through the vulnerable links in the network. So, the risks of the systems can be mitigated through investments in the security of interconnecting links. This article takes an innovative look at the problem of security investment of nodes on their vulnerable links in a given contagious network as a game-theoretic model that can be applied to a variety of applications including information systems. In the proposed game model, each node computes its corresponding risk based on the value of its assets, vulnerabilities, and threats to determine the optimum level of security investments on its external links respecting its limited budget. Furthermore, direct and indirect nonlinear influences of a node's security investment on the risks of other nodes are considered. The existence and uniqueness of the game's Nash equilibrium in the proposed game are also proved. Further analysis of the model in a practical case revealed that taking advantage of the investment effects of other players, perfectly rational players (i.e., those who use the utility function of the proposed game model) make more cost-effective decisions than selfish nonrational or semirational players.

Machine Learning for Risk Factor Identification and Cardiovascular Mortality Prediction Among Patients with Osteoporosis.

Risk prediction tools are increasingly popular aids in clinical decision-making. However, the underlying models are often trained on data from general patient cohorts and may not be representative of and suitable for use with targeted patient groups in actual clinical practice, such as in the case of osteoporosis patients who may be at elevated risk of mortality. We developed and internally validated a cardiovascular mortality risk prediction model tailored to individuals with osteoporosis using a range of machine learning models. We compared the performance of machine learning models with existing expert-based models with respect to data-driven risk factor identification, discrimination, and calibration. The proposed models were found to outperform existing cardiovascular mortality risk prediction tools for the osteoporosis population. External validation of the model is recommended.Clinical Relevance- This study presents the performance of machine learning models for cardiovascular death prediction among osteoporotic patients as well as the risk factors identified by the models to be important predictors.