Biomathematics
Department of Mathematics and Statistics
Texas Tech University
Abstract pdf
The Biomath seminar may be attended virtually Friday at 11:00 AM CDT (UT-5) via this Zoom link.
Meeting ID: 938 8653 3169
Passcode: 883472
Abstract pdf
The Biomath seminar may be attended virtually Friday at 11:00 AM CDT (UT-5) via this Zoom link.
Meeting ID: 938 8653 3169
Passcode: 883472
Continuous-time deterministic compartmental mathematical models are often used to simulate and analyze the course of epidemic outbreaks in time. Whereas modelling of control measures in the context of mathematical epidemiology has focused on optimal resource allocation, an important problem is the study of transient dynamic behaviour under state and input constraints for the dynamical system. These constraints normally reflect limitations in the intervention measures due to budget, capacity of healthcare system, or public policy goals.
We address the problem of viability, or existence of solutions of the controlled dynamical system that share a given set of properties, namely solutions that respect a given upper bound on the phase variable (in our case the size of the infected human compartment) subject to the control function taking values in a given compact set. Such transient dynamic behaviour of the solutions can be studied using viability kernels, which represent the largest set of initial states of the dynamical system such that the proportion of infected individuals is sustained below a given ceiling for all future times. Our goal is to characterise the viability kernel, and we focus on a level-set approach based on Bellman's value function satisfying a dynamic programming principle. This approach allows kernels with positive Lebesgue measure to be approximated numerically even for systems of higher dimensions.
As examples we study several models for vector-borne diseases of Susceptible-Infected or Susceptible-Infected-Recovered type for the human host, and Susceptible-Infected for the mosquito vector, as well as a two-patch system with human mobility based on the Lagrangian modelling framework. The epidemic control is based on the use of mosquito repellents in clothing and the control function takes values constrained by the maximum proportion of the host population employing the repellent-treated clothes. The inspiration for this work originates from COST Action 16227 Investigation and mathematical analysis of the effects of avantgarde disease control via mosquito nano-tech-repellents (2017-2022), funded by EU Horizon 2020 Framework Programme.
The Biomath seminar may be attended virtually Friday at 11:00 AM CDT (UT-5) via this Zoom link.
Meeting ID: 938 8653 3169
Passcode: 883472
Adaptive cancer therapy is a new paradigm of treatment for non-curative disease that aims to prolong emergence of resistance, and thus treatment failure. Here we use a mathematical model to explore how incorporating treatment toxicity into the protocol of adaptive therapy can be beneficial by both extending time to treatment failure and improving the quality of life for the patient. I will then demonstrate how the assumptions used when constructing a mathematical model can impose bias in simulations, resulting in differences in treatment predictions. These differences and biases affect personalized treatment plans and are critical to understand in order to quantifying the uncertainty in model predictions of cancer therapy.
The Biomath seminar may be attended virtually Friday at 11:00 AM CDT (UTC-5) via this Zoom link.
Meeting ID: 938 8653 3169
Passcode: 883472
First, a mathematical model for the spatiotemporal distribution of a migratory bird species is derived and analyzed. The birds have specific sites for breeding and winter feeding, and usually several stopover sites along the migration route, and therefore a patch model is the natural choice. However, we also model the journeys of the birds along the flyways, and this is achieved using a continuous space model of reaction-advection type. In this way proper account is taken of flight times and in-flight mortalities which may vary from sector to sector, and this information is featured in the ordinary differential equations for the populations on the patches through the values of the time delays and the model coefficients. The seasonality of the phenomenon is accommodated by having periodic migration and birth rates.
Second, another approach to modeling bird migration is proposed, in which there is a region where birds do not move but spend time breeding. Birds leave this breeding region and enter a migration flyway which is effectively a one-way corridor starting and ending at the breeding location. Mathematically, the flyway is a curve parametrized by arc-length. Flight speed depends on position along the flyway, to take account of factors such as wind and the pausing of birds at various locations for wintering or stopovers. Per-capita mortality along the flyway is both position and age-dependent, allowing for increased risks at stopover locations due to predation, and increased risks to immature birds. We also model indirect transmission, via contact with viruses, of avian influenza in migratory and nonmigratory birds, taking into account age structure. Sufficient conditions are obtained for the local stability of the disease-free equilibrium (for a species without migration) and for the disease-free periodic solution (for a migratory species). The birds have specific sites for breeding and winter feeding, and usually several stopover sites along the migration route, and therefore a patch model is the natural choice. However, we also model the journeys of the birds along the flyways, and this is achieved using a continuous space model of reaction-advection type. In this way proper account is taken of flight times and in-flight mortalities which may vary from sector to sector, and this information is featured in the ordinary differential equations for the populations on the patches through the values of the time delays and the model coefficients. The seasonality of the phenomenon is accommodated by having periodic migration and birth rates.
The Biomath seminar may be attended virtually Friday at 11:00 AM CDT (UTC-5) via this Zoom link.
Meeting ID: 938 8653 3169
Passcode: 883472