Theses Doctoral

Characterization and modeling of the human left atrium using optical coherence tomography

Lye, Theresa Huang

With current needs to better understand the interaction between atrial tissue microstructure and atrial fibrillation dynamics, micrometer scale imaging with optical coherence tomography has significant potential to provide further insight on arrhythmia mechanisms and improve treatment guidance. However, optical coherence tomography imaging of cardiac tissue in humans is largely unexplored, and the ability of optical coherence tomography to identify the structural substrate of atrial fibrillation has not yet been investigated. Therefore, the objective of this thesis was to develop an optical coherence tomography imaging atlas of the human heart, study the utility of optical coherence tomography in providing useful features of human left atrial tissues, and develop a framework for optical coherence tomography-informed cardiac modeling that could be used to probe dynamics between electrophysiology and tissue structure.
Human left atrial tissues were comprehensively imaged by optical coherence tomography for the first time, providing an imaging atlas that can guide identification of left atrial tissue features from optical coherence tomography imaging. Optical coherence tomography image features corresponding to myofiber and collagen fiber orientation, adipose tissue, endocardial thickness and composition, and venous media were established. Varying collagen fiber distributions in the myocardial sleeves were identified within the pulmonary veins. A scheme for mapping optical coherence tomography data of dissected left atrial tissues to a three-dimensional, anatomical model of the human left atrium was also developed, enabling the mapping of distributions of imaged adipose tissue and fiber orientation to the whole left atrial geometry. These results inform future applications of structural substrate mapping in the human left atrium using optical coherence tomography-integrated catheters, as well as potential directions of ex vivo optical coherence tomography atrial imaging studies.
Additionally, we developed a workflow for creating optical mapping models of atrial tissue as informed by optical coherence tomography. Tissue geometry, fiber orientation, ablation lesion geometry, and heterogeneous tissue types were extracted from optical coherence tomography images and incorporated into tissue-specific meshes. Electrophysiological propagation was simulated and combined with photon scattering simulations to evaluate the influence of tissue-specific structure on electrical and optical mapping signals. Through tissue-specific modeling of myofiber orientation, ablation lesions, and heterogeneous tissue types, the influence of myofiber orientation on transmural activation, the relationship between fluorescent signals and lesion geometry, and the blurring of optical mapping signals in the presence of heterogeneous tissue types were investigated.
By providing a comprehensive optical coherence tomography image database of the human left atrium and a workflow for developing optical coherence tomography-informed cardiac tissue models, this work establishes the foundation for utilizing optical coherence tomography to improve the structural substrate characterization of atrial fibrillation. Future developments include analysis of optical coherence tomography imaged tissue structure with respect to clinical presentation, development of automated processing to better leverage the large amount of imaging data, enhancements and validation of the modeling scheme, and in vivo evaluation of the left atrial structural substrate through optical coherence tomography-integrated catheters


  • thumnail for Lye_columbia_0054D_15173.pdf Lye_columbia_0054D_15173.pdf application/pdf 7.87 MB Download File

More About This Work

Academic Units
Electrical Engineering
Thesis Advisors
Hendon, Christine P.
Ph.D., Columbia University
Published Here
April 24, 2019