About
I'm currently a M.Sc. graduate student at McGill University under the supervision of Prof. Tal Arbel. I'm researching how to leverage deep learning models to predict the future outcomes of patients diagnosed with Multiple Sclerosis. I really like programming and using open-source software tools as part of my research.
You can find my complete CV here.
Research Interests
- Deep Learning
- Representational Learning
- Computer Vision
- Neuroimage Analysis
Education
-
M.Sc. (Thesis) in Electrical and Computer Engineering
McGill University, Montreal, Canada
-
B.Eng (Software Engineering), Minor in Computer Science
McGill University, Montreal, Canada
Experience
-
Machine Learning Developer (R&D)
Sama, Canada
-
Research Assistant
McGill University, Canada
-
Software Developer
Matrox Electronics System Ltd., Canada
Publications
Cohort Bias Adaptation in Aggregated Datasets for Lesion Segmentation
B. Nichyporuk. J. Cardinell, J. Szeto, R. Mehta, D.L. Arnold, S.Tsaftaris and T. Arbel, "Cohort Bias Adaptation in Federated Datasets for Lesion Segmentation", in Proceedings of the MICCAI 2021 Workshop: 3rd MICCAI Workshop on Domain Adaptation and Representation Transfer (DART), held in conjunction with the 24th International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI 2021), held virtually (Strasbourg, France), September 2021. BEST PAPER AWARD
Accounting for Variance in Machine Learning Benchmarks
X. Bouthillier, P. Delaunay, M. Bronzi, A. Trofimov, B. Nichyporuk, J. Szeto, N. Mohammadi Sepahvand, E. Raff, K. Madan, V. Voleti, S. Ebrahimi Kahou, V. Michalski, T. Arbel, C. Pal, G. Varoquaux, P. Vincent, “Accounting for Variance in Machine Learning Benchmarks”, in Proceedings of Machine Learning and Systems 3, (MLSys 2021).
Optimizing Operating Points for High Performance Lesion Detection and Segmentation Using Lesion Size Reweighting
B. Nichyporuk, J. Szeto, D.L. Arnold and T. Arbel, "Optimizing Operating Points for High Performance Lesion Detection and Segmentation Using Lesion Size Reweighting", the 4th Conference on Medical Imaging With Deep Learning (MIDL 2021), held virtually (Lubeck, Germany), July 7-9, 2021. (short paper)
2021, made with in pure Bootstrap 4, inspired by Academic Template for Hugo