ABSTRACT
Having started since late 2019, COVID-19 has spread through far many nations around the globe. Not being known profoundly, the novel virus of the Coronaviruses family has already caused more than half a million deaths and put the lives of many more people in danger. Policymakers have implemented preventive measures to curb the outbreak of the virus, and health practitioners along with epidemiologists have pointed out many social and hygienic factors associated with the virus incidence and mortality. However, a clearer vision of how the various factors cited hitherto can affect total death in different communities is yet to be analyzed. This study has put this issue forward. Applying artificial intelligence techniques, the relationship between COVID-19 death toll and determinants mentioned as strongly influential in earlier studies was investigated. In the first stage, employing Best-Worst Method, the weight of the primer contributing factor, effectiveness of strategies, was estimated. Then, using an integrated Best-Worst Method-local linear neuro-fuzzy-adaptive neuro-fuzzy inference system approach, the relationship between COVID-19 mortality rate and all factors namely effectiveness of strategies, age pyramid, health system status, and community health status was elucidated more specifically.
Subject(s)
Artificial Intelligence , COVID-19 , Humans , United States/epidemiology , Neural Networks, Computer , Fuzzy LogicABSTRACT
This paper proposes an Evolving Local Linear Neuro-Fuzzy Model for modeling and identification of nonlinear time-variant systems which change their nature and character over time. The proposed approach evolves through time to follow the structural changes in the time-variant dynamic systems. The evolution process is managed by a distance-based extended hierarchical binary tree algorithm, which decides whether the proposed evolving model should be adapted to the system variations or evolution is necessary. To represent an interesting but challenging example of the systems with changing dynamics, the proposed evolving model is applied to model car-following process in a traffic flow, as an online identification problem. Results of simulations demonstrate effectiveness of the proposed approach in modeling of the time-variant systems.
ABSTRACT
In this paper, we present and validate a data-driven method to lossy tube-load modeling of arterial tree in humans. In the proposed method, the lossy tube-load model is fitted to central aortic and peripheral blood pressure (BP) waves in the time domain. For this purpose, we employ a time-domain lossy tube-load model in which the wave propagation constant is formulated to two terms: one responsible for the alteration of wave amplitude and the other for the transport delay. Using the experimental BP data collected from 17 cardiac surgery patients, we showed that the time-domain lossy tube-load model is able to accurately represent the relation between central aortic versus upper-limb and lower-limb BP waves. In addition, the comparison of lossy versus lossless tube-load models revealed that (1) the former outperformed the latter in general with the root-mean-squared errors (RMSE) of 3.1 mm Hg versus 3.5 mm Hg, respectively (p-value < 0.05), and (2) the efficacy of the former over the latter was more clearly observed in case the normalized difference in the mean central aortic versus peripheral BP was large; when the difference was >5% of the underlying mean BP, lossy and lossless models showed the RMSE of 2.7 mm Hg and 3.7 mm Hg, respectively (p-value < 0.05).
