Date of Award
Spring 2026
Language
English
Embargo Period
4-26-2028
Document Type
Dissertation
Degree Name
Doctor of Philosophy (PhD)
College/School/Department
Department of Epidemiology and Biostatistics
Program
Biostatistics
First Advisor
Edward Valachovic
Committee Members
Erika Martin, Edward Valachovic, Rachel Hart-Malloy, Petko Bogdanov
Keywords
Anomaly detection, disease surveillance
Subject Categories
Applied Statistics | Biostatistics | Epidemiology | Longitudinal Data Analysis and Time Series | Vital and Health Statistics
Abstract
Determining the time and location of potential disease outbreaks is a critical component of disease surveillance. As the syphilis epidemic in the United States evolves, identifying high-risk areas and populations enables public health agencies to mitigate its impact. Although anomaly detection methods have been applied to routinely collected surveillance data, rapid advances across disparate fields have outpaced their implementation for outbreak detection. This dissertation 1) compares the performance of several existing anomaly detection methods using real-time spatiotemporal syphilis data 2) develops a new anomaly detection method, TGSD-STARMA, and 3) demonstrates the ability of TGSD-STARMA to detect potential syphilis outbreaks in real-time.
License
This work is licensed under the University at Albany Standard Author Agreement.
Recommended Citation
Angles, John S., "Anomaly Detection in Real-Time Spatiotemporal Count Data: A Novel Approach with Application to Case-Based Syphilis Surveillance and Outbreak Detection" (2026). Electronic Theses & Dissertations (2024 - present). 423.
https://scholarsarchive.library.albany.edu/etd/423
Included in
Applied Statistics Commons, Biostatistics Commons, Epidemiology Commons, Longitudinal Data Analysis and Time Series Commons, Vital and Health Statistics Commons