Thesis Format
Monograph
Degree
Master of Science
Program
Neuroscience
Supervisor
Morton, J. Bruce
Abstract
Neuromelanin-sensitive MRI (NM-MRI) provides a non-invasive means of characterizing neuromelanin-rich nuclei in the midbrain. However, accurate delineation of the substantia nigra remains challenging. Here, we evaluated two alternative segmentation approaches--graph-cut (GC) and atlas-based segmentation--to gold-standard manual segmentation (MS) using ultra-high resolution 7T NM-MRI data. The GC approach employed a semi-automated region-growing algorithm guided by user-defined seeds, while the atlas-based method involved warping a previously developed SN atlas into each participant’s native space. We evaluated each method against MS using a comprehensive set of segmentation evaluation metrics. GC segmentation produced SN volumes statistically indistinguishable from MS and demonstrated moderate boundary alignment, suggesting its viability as a labor-saving alternative. In contrast, the atlas-based method substantially overestimated SN volume and showed poor boundary accuracy, underscoring the need for refined preprocessing pipelines. Overall, GC segmentation emerges as a promising tool for reliable, large-scale NM-MRI analyses, reducing reliance on time-intensive manual delineations.
Summary for Lay Audience
When we look inside the human brain using special types of scans, we can find tiny areas that are important for controlling movement and other functions. One such area, called the substantia nigra, contains cells with a dark pigment called neuromelanin. This pigment is linked to dopamine, a chemical messenger in the brain that helps regulate movement, learning, and emotions. Changes in these neuromelanin-containing cells are associated with conditions like Parkinson’s disease. If we can pinpoint these cells more precisely and measure their size and health, we can better understand brain disorders and possibly develop new treatments.
Today, advanced MRI scanners, especially those operating at very high magnetic fields (7 Tesla), let us see these small regions more clearly than ever before. However, outlining the exact shape and size of the substantia nigra from these scans is still challenging and time-consuming, especially when experts must manually draw the boundaries by hand. This manual work can be slow, prone to errors, and hard to repeat consistently.
Our study tested two computer-assisted methods to define the boundaries of the substantia nigra from ultra-detailed neuromelanin-sensitive MRI images, comparing them to the traditional manual approach. The first method, called a “graph-cut” algorithm, uses a computer program to grow a region of interest from a small starting point selected by the researcher. The second method uses a pre-made “atlas” or map of the brain and tries to fit it onto each person’s scan.
We found that the graph-cut method worked very well—nearly as good as the carefully done manual outlines—and took much less effort. The atlas-based method, on the other hand, tended to overshoot and include too much tissue that wasn’t part of the substantia nigra.
By using the graph-cut approach, future researchers and doctors may save time and achieve more accurate results, making it easier to study the brain’s structure in healthy people and those with brain disorders. Ultimately, this could lead to better tools for early detection, improved tracking of disease progression, and more targeted therapies for conditions like Parkinson’s.
Recommended Citation
Fallah, Aria A., "Evaluation of substantia nigra volume quantification approaches using neuromelanin-sensitive MRI (NM-MRI)" (2025). Electronic Thesis and Dissertation Repository. 10838.
https://ir.lib.uwo.ca/etd/10838
Creative Commons License
This work is licensed under a Creative Commons Attribution-Noncommercial 4.0 License