Back to search

Georgia Institute of Technology

Lung Cancer Risk Estimation from Radon Exposure: A Multiscale Analysis Using Ecological and Machine Learning Approaches

Abstract

dc:description.abstract

Lung cancer is one of the major death causes in both United States and globally, and radon exposure is known for the second-leading cause of lung cancer incidence. The overall objective of this study is to provide a comprehensive multiscale analysis of radon exposure and its association with lung cancer risk, integrating ecological, modeling, and risk estimation methodologies. The research is structured into three primary sections. The first section presents an ecological study across counties in the SEER registry, focusing on lung cancer incidence in relation to environmental radon exposure and other relevant factors. While this analysis highlights significant associations, it also reveals data limitations inherent in the county-level scale, such as geographic variability and the granularity of exposure data. To address these limitations, the second section shifts focus to estimating and modeling residential radon exposure at the ZCTA level in Pennsylvania. Pennsylvania was chosen due to its notably high levels of both smoking and radon compared to the national average. Using machine learning techniques, residential radon levels are modeled to provide a more detailed spatial understanding of exposure at a finer scale. This approach allows for better resolution in assessing radon-related health risks and aids in overcoming some of the challenges posed by broader geographic scales. The third section involves the review of risk estimation using the BEIR VI models, incorporating innovative elements such as adopting migration data or highlighting the importance of adopting granular data. This novel approach addresses the impact of population movement on long-term radon exposure and subsequent lung cancer risk, enhancing the understanding of risk dynamics across different spatial scales. Also, in this part, some of the questions that can be raised when applying the risk models as BEIR VI, such as which floor level should be used as the reference, or which approaches can be done when the smoking data is not available. The dissertation contributes to the field by providing insights into how spatial granularity and population dynamics can influence radon risk estimates and providing methodological approaches for risk analysis of residential radon exposure. This work underscores the importance of integrating ecological analysis, fine-scale radon modeling, and migration data to improve risk estimation frameworks, offering detailed exploration of lung cancer risk attributable to radon and providing recommendations for enhancing current risk models such as BEIR VI.

Degree

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

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Lee, Heechan
Advisor dc:contributor.advisor
  • Dewji, Shaheen
Committee members dc:contributor.committeemember
  • Asensio, Omar Isaac
  • Wang, C.-K.
  • Agasthya, Greeshma
  • Biegalski, Steven
  • Hanson, Heidi

Subjects

dc:subject × 6

Rights

Language dc:language.iso
en_US

Identifiers

dc:identifier.*
Handle dc:identifier.uri
https://hdl.handle.net/1853/77808

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

Lee, Heechan. Lung Cancer Risk Estimation from Radon Exposure: A Multiscale Analysis Using Ecological and Machine Learning Approaches. Doctoral thesis, Georgia Institute of Technology, 2025. https://hdl.handle.net/1853/77808