Tal Arbel

Timeline

1. 1992–2000

   Degrees at McGill

   She studies electrical engineering at McGill University as an undergraduate and completes a bachelor's degree in 1992, a master's in 1995 and a PhD in 2000. Her PhD work on object recognition with entropy maps earns the D.W. Ambridge Prize for the best dissertation in physical sciences and engineering at McGill.

2. 2000–2001

   Joining the McGill faculty

   After her PhD she works at the Montreal Neurological Hospital on computer vision methods for neurology and neurosurgery, and McGill appoints her research associate in 2000 and assistant professor in 2001. The Probabilistic Vision Group, which she leads in the Centre for Intelligent Machines, is established in 2001. Its algorithms for lesion detection and segmentation in brain scans of patients with MS are used in the clinical trial analysis of most new MS drugs.

3. 2019–2020

   A CIFAR AI Chair and MELBA

   She receives the 2019 McGill Engineering Christophe Pierre Research Award. In 2020, she is awarded a Canada CIFAR AI Chair as an associate faculty member of Mila and starts the Journal on Machine Learning for Biomedical Image Analysis (MELBA), for which she serves as Editor-in-Chief.

4. 2023

   Fairness and uncertainty estimates

   In 2023, she coauthors a study with Raghav Mehta and Changjian Shui on how popular fairness methods affect uncertainty quantification in medical image analysis. It tests methods such as data balancing and distributionally robust optimization on three clinically relevant tasks.

5. 2023

   Uncertainty-aware causal models

   In 2023, she coauthors a paper with Joshua Durso-Finley and colleagues that uses Bayesian deep learning to estimate the posterior distribution over factual and counterfactual outcomes for several treatments. The model predicts new and enlarging T2 lesion counts from brain MRI of MS patients in randomized controlled trials, with an uncertainty estimate for each treatment option.

6. May 2024

   Canadian Academy of Engineering fellowship

   In May 2024, the Canadian Academy of Engineering announces the election of fifty-three new Fellows, and Arbel is among the five McGill professors inducted into the Academy.

7. 2025

   PRISM and a renewed chair

   In 2025, she coauthors PRISM, a framework that uses Stable Diffusion to generate high-resolution, language-guided counterfactual medical images. She also gives a keynote talk at the Conference on Medical Imaging with Deep Learning (MIDL 2025) in Salt Lake City. Her Canada CIFAR AI Chair is renewed in 2025.

8. As of October 2026

   Professor at McGill and Mila

   She is a full professor in McGill's Department of Electrical and Computer Engineering, where she directs the Probabilistic Vision Group and Medical Imaging Lab in the Centre for Intelligent Machines. She is also a core academic member of Mila, a Canada CIFAR AI Chair, and Editor-in-Chief and co-founder of the MELBA journal. Her research develops probabilistic deep learning for medical image analysis, with a focus on neurological diseases, and her team builds causal-temporal and 3D spatio-temporal generative models, multimodal foundation models and agentic AI frameworks.