AICONS Lab

Research

Our research revolves around three pivotal and complementary themes, all linked by the overarching goal of developing novel computational and AI tools for neuroscience. This work leads to novel translational biomarkers, advances understanding of circuit disorder mechanisms, and improves outcomes through earlier and more personalized treatments.

I
Mapping of Neuronal & Vascular Brain Networks

Mapping brain networks across resolutions of measurement is essential for delineating structural and functional changes specific to aging and disease. Recent advances in imaging technology and clearing techniques enable high-fidelity, whole-brain mapping of tracers, cell distributions and processes in intact brain tissue, generating tera-scale, complex 3D datasets. We develop and validate neural networks and computational methods for precise, fast and automated whole-brain mapping of connectivity and vasculature, and for integrated multimodal analysis across spatial scales. We aim to define signatures of circuit dysfunction that could serve as novel outcome measures in neurological disorders.

Theme 1 Figure 1
Automated computational pipeline coupling MRI with cellular information from 3D histology to model brain-wide circuit alterations
Theme 1 Figure 2
Studying spatio-temporal stroke effects on connected regions across the whole brain
Theme 1 Figure 3
Machine learning algorithms for high-throughput analysis of cerebrovascular networks
II
Advanced, Multimodal Neuroimage Analyses

Neurodegenerative diseases are predicted to surpass cancer as the 2nd leading cause of death by 2040. Our work aims to understand and predict brain network re-organization in Alzheimer's disease and other neurodegenerative disorders. We develop state-of-the-art image processing and connectivity analysis pipelines with AI modules including structure and lesion segmentation, image registration, super-resolution, and quality control. We also investigate the effects of MR-guided focused ultrasound on brain morphology, connectivity, and metabolism during blood-brain barrier opening or minimally-invasive neurosurgery.

Theme 2 Figure 1
Automated pipelines for image segmentation, connectivity analyses, and patient classification
Theme 2 Figure 2
Deep neural network model development and optimization
Theme 2 Figure 3
Understanding brain (re)organization through connectomics, graph theory and gradient-based analyses
III
Predictive Modelling & Hybrid Learning

Our group applies cutting-edge AI techniques to build and validate predictive and prognostic tools, employing imaging, clinical and genomic data to predict patient outcomes and cognitive decline, identify at-risk individuals, and characterize unique patient sub-populations. We aim to develop AI networks that guide clinical decision making and save critical time in the context of stroke and brain trauma.

Theme 3 Figure 1
Predicting subject-specific longitudinal cognitive outcomes
Theme 3 Figure 2
Bayesian neural networks for guiding patient selection in acute stroke
Theme 3 Figure 3
Deep learning model benchmarking on out-of-distribution data and MRI corruptions
Funding

Our work is kindly supported by

CRC CIHR NSERC CFI NIH SRI CC AA OBI BC AS