Advanced boundary-enhanced instance segmentation and spatial-temporal transformer models for automated schizophrenic investigation

Simple item page
dc.contributor.authorImarhiagbe, Osasumwen Raphael
dc.contributor.authorUniversity of Lethbridge. Faculty of Arts and Science
dc.contributor.supervisorZhang, John Z.
dc.date.accessioned2025-10-15T20:53:28Z
dc.date.available2025-10-15T20:53:28Z
dc.date.issued2025
dc.degree.levelMasters
dc.description.abstractAccurate segmentation and detection in neuroimaging is essential for advancing clinical understanding and the diagnosis of schizophrenia. This thesis introduces Boundary-Refined Attention Network (BoRefAttnNet), a novel boundary-refined 3D U-Net variant specifically designed for precise segmentation of subcortical brain structures from structural magnetic resonance imaging (sMRI). BoRefAttnNet incorporates multi-scale boundary attention modules that explicitly highlight anatomically critical edges while suppressing background noise, significantly improving segmentation accuracy for small or complex anatomical structures. Evaluations using FastSurfer-processed sMRI data from the publicly available Centre for Biomedical Research Excellence (COBRE) dataset demonstrate that BoRefAttnNet significantly outperforms conventional 3D U-Net baselines in accurately delineating key subcortical structures, including the hippocampus, amygdala, and basal ganglia. Building upon this enhanced segmentation capability, we further experiment with a Dynamic Spatial-Temporal Transformer Model (DySTTM) to detect schizophrenia by integrating structural and functional MRI (fMRI) modalities. The DySTTM leverages spatial attention to capture anatomical interdependencies from segmented sMRI data and temporal attention to model dynamic brain connectivity patterns from resting-state fMRI. Experimental results indicate that the integration of these multimodal imaging features using DySTTM provides superior diagnostic accuracy and interpretability compared to established models such as 3D ResNet and XGBoost classifiers.
dc.description.sponsorshipUniversity of Lethbridge, Alberta Innovates, Digital Research Alliance of Canada, Alberta Machine Intelligence Institute.
dc.embargoNo
dc.identifier.urihttps://hdl.handle.net/10133/7170
dc.language.isoen
dc.publisherLethbridge, Alta. : University of Lethbridge, Dept. of Mathematics and Computer Science
dc.publisher.departmentDepartment of Mathematics and Computer Science
dc.publisher.facultyArts and Science
dc.relation.ispartofseriesThesis (University of Lethbridge. Faculty of Arts and Science)
dc.subjectboundary-aware segmentation
dc.subjectattention mechanisms
dc.subjectdeep learning
dc.subjectmultimodal fusion
dc.subjectdynamic spatial-temporal modeling
dc.subjectmedical image segmentation
dc.subjectcomputational psychiatry
dc.subject.lcshDissertations, Academic
dc.subject.lcshDeep learning (Machine learning)
dc.subject.lcshSchizophrenia--Imaging
dc.subject.lcshBrain--Imaging
dc.subject.lcshBrain--Magnetic resonance imaging
dc.subject.lcshImage segmentation--Research
dc.titleAdvanced boundary-enhanced instance segmentation and spatial-temporal transformer models for automated schizophrenic investigation
dc.typeThesis
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
IMARHIAGBE_OSASUMWEN_MSC_2025.pdf
Size:
1.53 MB
Format:
Adobe Portable Document Format
Description:
Download
License bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
license.txt
Size:
3.33 KB
Format:
Item-specific license agreed upon to submission
Description:
Download
Collections
Arts and Science, Faculty of
University of Lethbridge Theses