Electronic Thesis and Dissertation Repository

Thesis Format

Integrated Article

Degree

Doctor of Philosophy

Program

Medical Biophysics

Supervisor

Khan, Ali R.

Abstract

Short-ranged connectivity comprise the majority of connections throughout the brain, joining together nearby regions and contributing to important networks that facilitate complex function and cognition. Despite constituting the majority of white matter in the brain and their importance, studies examining short-ranged connections have thus far been limited in part due to the challenges associated with identifying and validating them. Tractography, a computational technique for reconstructing axon trajectories from diffusion magnetic resonance imaging, has been commonly used to identify and study major white connections (e.g. corticospinal tract), which are easier to identify relative to the short-ranged connections. The use of additional constraints (e.g. geometry, regions of interest) together with tractography has enabled the ability to identify short-ranged connections of interest, such as the ”U”-shaped tracts residing just below the cortical surface, and the subcortical connectome tracts found in the deep brain.

In this thesis, we aimed to quantify the reliability of such techniques for studying the short- ranged connections and applied them to examine changes to short-ranged connectivity in patients with first episode schizophrenia. First, the reliability of identifying short-ranged, ”U”- shaped tracts is examined in Chapter 1, leveraging geometric constraints for identifying the ”U”-shaped geometry together with clustering techniques to establish distinct tracts. Here, we two different clustering techniques, applying them to two datasets to study both the reliability of identifying short-ranged, ”U”-shaped tracts across different subjects and in a single individual (across different sessions). In Chapter 2, the reliability for identifying the subcortical connectome (short-ranged connections between subcortical structures) is evaluated. Connectivity of the deep brain is often hard to recapitulate due to the multiple orientations contributing to com- plex diffusion signals. Thus, we leveraged regions of interests determined through histological data to aid identification of the short-ranged connections in the compact region. Finally, Chapter 4, uses the techniques from chapter 2 in combination with quantitative measures sensitive to microstructural changes to study changes to short-ranged, ”U”-shaped tracts in the frontal lobes of patients with first-episode schizophrenia (FES). By studying the short-ranged connections in patients with FES, biomarkers associated with clinical presentation may be elucidated and may aid the current understanding to improve future treatment. Overall, the projects presented here quantify the reliability of current techniques for investigating short-ranged connectivity and provides a framework for evaluating of future techniques. Additionally, the techniques evaluated here can be used to elucidate new findings and improve treatment in clinical popula- tions.

Summary for Lay Audience

Similar to how roads and highways join nearby cities, basic human functions are made possible through the structural connections (e.g. roads and highways) connect different regions of the brain (e.g. cities). Much like the roads (joining local landmarks) and highways (joining cities far apart), the brain also has both short and long-ranged connections. Using diffusion magnetic resonance imaging (dMRI), scientists are able to study these connections non-invasively. As with any technique, especially one that is used to non-invasively study the brain, confidence of a method improves with reliability. That is, when a method is applied multiple times, similar results should be produced across different healthy individuals. The current thesis evaluates the reliability of current techniques to study the short-ranged connections of the human brain.

Once reliability of the techniques to study connections of the brain has been established, the same techniques can be applied to examine how such connections can change due to disease. For example, we can measure properties such as the integrity along the length of the connection. In a patient, we may see reduced integrity along it’s length, similar to how a pothole may be encountered along a highway between two cities. Such changes may identify important associations with experienced symptoms and help us understand the progression of disease. Furthermore, identifying these changes can help to improve the treatment administered through better understanding of how the brain changes.

Short-ranged connections found just below the surface of the brain demonstrate a unique ”U”- shaped geometry. Using a geometric constraints with tractography, reliability of identifying these connections are evaluated both across different individuals and in the same individual at different timepoints in chapter 2. Chapter 3 examines the reliability of short-ranged connections found connecting regions deep below the surface of the brain. Identification of these deep connections is complicated due to the various crossing connections muddying the signal in a small, condensed region. After establishing reliability in chapter 2, we use the same technique to identify similar connections, examining for changes along its length in patients with first-episode schizophrenia. Altogether, we quantify the reliability of current methods for investigating short-ranged connections, while also providing a framework for evaluating new techniques in the future. Furthermore, we demonstrate how these techniques can be used to improve our current understanding of changes to the brain’s connections due to schizophrenia or other diseases.

Creative Commons License

Creative Commons Attribution 4.0 License
This work is licensed under a Creative Commons Attribution 4.0 License.

Download

Share

COinS