I am a MASc candidate in the Department of Electrical and Computer Engineering at Concordia University. My current research focuses on predicting multimodal brain connectomes using graph neural networks: a unified approach for neuroimaging and Alzheimer’s disease.
I completed my bachelor’s degree in Computer Engineering at the Islamic Azad University of Tehran. During my studies, I explored applications of machine learning in healthcare, which sparked my interest in computational neuroscience. I gained hands-on experience in computer science fundamentals, data science methods, and software engineering, building a solid foundation for my current research on brain connectomes.
My research focuses on developing advanced graph neural network (GNN) models for multimodal brain connectome prediction. In the first phase, I aim to design a model that takes as input a brain network derived from one neuroimaging modality—such as EEG or MEG—and accurately predicts the corresponding functional or structural brain graph from another modality, such as fMRI. This approach enables the inference of hard-to-acquire or expensive brain connectivity data from more accessible measurements. In the second phase, I will extend this framework to handle dynamic brain networks, predicting time-varying graphs from input graphs that themselves evolve over time. By modeling these spatiotemporal dynamics, the project seeks to capture the progression of brain connectivity patterns, which is critical for advancing our understanding of healthy aging and enabling the early detection of neurodegenerative diseases such as Alzheimer’s. Ultimately, this work aims to provide novel tools for brain research and potential biomarkers for clinical diagnosis.
3 Minute Thesis summary: https://www.linkedin.com/posts/leila-mousavi-8a3b9a2aa_3mt-mt180-concordiauniversity-activity-7311159662794412033-v5I9/?utm_source=share&utm_medium=member_desktop&rcm=ACoAAEqSBXsBsoYUEUfUtm4KP7ijh9fXRzJjiBM
