Researchers using ordinary differential equations to model phenomena face two main challenges among others: implementing the appropriate model and optimizing the parameters of the selected model. The latter often proves difficult or computationally expensive. Here, we implement Particle Swarm Optimization, which draws inspiration from the optimizing behavior of insect swarms in nature, as it is a simple and efficient method for fitting models to data. We demonstrate its efficacy by showing that it outstrips evolutionary computing methods previously used to analyze an epidemic model.
The abundance of type and quantity of available data in the healthcare field has led many to utilize machine learning approaches to keep up with this influx of data. Data pertaining to COVID-19 is an area of recent interest. The widespread influence of the virus across the United States creates an obvious need to identify groups of individuals that are at an increased risk of mortality from the virus. We propose a so-called clustered random forest approach to predict COVID-19 patient mortality. We use this approach to examine the hidden heterogeneity of patient frailty by examining demographic information for COVID-19 patients. We find that our clustered random forest approach attains predictive performance comparable to other published methods. We also find that follow-up analysis with neural network modeling and k-means clustering provide insight into the type and magnitude of mortality risks associated with COVID-19.
We consider the problem of using time-series data to inform a corresponding deterministic model and introduce the concept of genetic algorithms (GA) as a tool for parameter estimation, providing instructions for an implementation of the method that does not require access to special toolboxes or software. We give as an example a model for cholera, a disease for which there is much mechanistic uncertainty in the literature. We use GA to find parameter sets using available time-series data from the introduction of cholera in Haiti and we discuss the value of comparing multiple parameter sets with similar performances in describing the data.
We extend existing impulsive differential equation models for integrated pest management (IPM) by including stage structure for both predator and prey as well as by adding stochastic elements in the birth rate of the prey. Based on our model, we propose an approach that incorporates various competing stochastic components. This approach enables us to select a model with optimally determined weights for maximum accuracy and precision in parameter estimation. This is significant in the case of integrated pest management because the proposed model accommodates varying unknown environmental and climatic conditions, which affect the resources needed for pest eradication.