Summary
Arthur Dehgan is a PhD candidate in Neuro-AI with nine years of experience applying machine learning and deep learning to large-scale time-series neuroimaging data (MEG/EEG). He combines signal and image processing foundations with production-minded software skills—Python, PyTorch, NumPy/Pandas—and scalable Slurm/HPC workflows used on Compute Canada/Québec clusters. His research on terabyte-scale datasets like CamCAN produced widely cited publications and practical tools, including the open-source MEEGNet library that makes neural networks more interpretable for neuroscientists. Arthur has hands-on experience building reproducible pipelines, data visualizations for model interpretability, and Riemannian-based covariance methods for brain-signal classification. Based in Montreal, he bridges academic rigor and engineering discipline to translate Neuro-AI research into deployable machine learning solutions. A less obvious strength is his background in big-data cluster setup (Hadoop/Spark) from early CNRS work, giving him end-to-end expertise from infrastructure to models.
9 years of coding experience
Doctor of Philosophy - PhD, Neuro-AI, Doctor of Philosophy - PhD, Neuro-AI at Université de Montréal
MPSI - PSI, MPSI - PSI at CPGE Janson de Sailly
Master of Science (M.S.), High Tech Imaging, Master of Science (M.S.), High Tech Imaging at Télécom SudParis
English, German, French