Back to search

Georgia Institute of Technology

Efficient Parallel Computation of Particle-based Biological Systems: Applications to Simulations of Bird Flocks and Optimizing Models of Cardiac Tissue

Abstract

dc:description.abstract

Graphics processing units (GPUs) are ubiquitous in modern computers, and are characterized by their large-scale parallel computing capability. Many computing problems that require updating state variables for a large number of mostly independent agents are well-suited to simulation using GPUs, as the steps that can be parallelized often represent the majority of the computation but are individually simple. This thesis focuses on two particle-based systems which have been implemented using this approach. In both cases, these simulations use the Web Graphics Library available in all modern web browsers to utilize any available graphics hardware without requiring any manual software installation,configuration, or compilation by the end-user. The first program uses particle swarm optimization, an optimization technique inspired by the behavior of groups of animals, to identify parameterizations of cardiac models to match experimental data. This tool, called CardioFit, allows the identification of parameters in a matter of seconds on most computers and is capable of finding low-error solutions even for challenging datasets, such as data recorded from human hearts exhibiting Brugada syndrome. By design, CardioFit is easy to use through a user interface without requiring any code to be written or modified by the end-user, while still allowing control over optimization parameters for advanced users. The second program simulates flocks of up to tens of thousands of birds in near-real time, with interactive settings to choose between a variety of flocking models, change their parameters, and observe the behavior through a live visualization. This flocking simulation reveals that large flocks of birds place greater constraints on existing flocking models to obtain desired behaviors such as the ability to maintain the flock through turns due to the need for long-range information transfer.

Degree

thesis:*
Level thesis:degree_level
Doctoral
Department dc:contributor.department
Computational Science and Engineering
Grantor dc:publisher
Georgia Institute of Technology
Year dc:date.issued
2025

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Comstock, Maxfield Roth
Advisor dc:contributor.advisor
  • Cherry, Elizabeth
Committee members dc:contributor.committeemember
  • Fenton, Flavio H
  • Goldman, Daniel I
  • Xu, Helen
  • Rodriguez-Padilla, Jairo

Subjects

dc:subject × 3

Rights

Language dc:language.iso
en_US

Identifiers

dc:identifier.*
Handle dc:identifier.uri
https://hdl.handle.net/1853/78722

Chain of custody

source
Harvested from
Georgia Tech
Base URL
repository.gatech.edu/server/oai/request
Last updated
2026-11-19
Source record
OAI-PMH GetRecord
citation

Comstock, Maxfield Roth. Efficient Parallel Computation of Particle-based Biological Systems: Applications to Simulations of Bird Flocks and Optimizing Models of Cardiac Tissue. Doctoral thesis, Georgia Institute of Technology, 2025. https://hdl.handle.net/1853/78722