## Theses, Dissertations, and Projects

To access this work you must either be on the Smith College campus OR have valid Smith login credentials.

On Campus users: To access this work if you are on campus please Select the Download button.

Off Campus users: To access this work from off campus, please select the Off-Campus button and enter your Smith username and password when prompted.

Non-Smith users: You may request this item through Interlibrary Loan at your own library.

# Using genetic algorithms to incorporate antibiotic resistance in the modeling of tuberculosis in the United States

2017-5

Gwen Spencer

Honors Thesis

Bachelor of Arts

## Department

Mathematics and Statistics

## Keywords

Infectious diseases, Tuberculosis, Mathematical modeling, Antibiotic resistance, Genetic algorithms, Drug resistance, Communicable diseases-Mathematical models, Tuberculosis-Mathematical models

## Abstract

Tuberculosis (TB) is still a major public health issue: millions of individuals are infected by the causal bacillus Mycobacterium tuberculosis (Mtb). The greatest challenge to the eradication of the disease has been the emergence of antibiotic-resistant strains, especially those resistant to first line drugs, such as isoniazid (H), rifampin (R), pyrazinamide (PYZ) and ethambutol (ETB). Resistance renders the disease more difficult and expensive to treat [14]. The past few years, although several models have been developed to explain TB prevalence [13, 16, 20], none of them fully incorporate the effects of antibiotic resistance. The aim of this project is to integrate the effects of single- and multi- drug resistant strains to develop a more realistic and predictive model of TB dynamics in the US population. For this purpose, we developed a compartmental mathematical model described by differential equations. Data from the Centers of Disease Control and Prevention (CDC) on TB incidence and prevalence for the years 1993-2014 was collected. A set of 27 parameters was fitted to the data, using a genetic algorithm to minimize an error function that cannot be differentiated. Local minima were identified and multiple sensitivity analysis tests were performed to identify which parameters the model is sensitive to.

English