Authors :
Presenting Author: Giorgia Atanasio, MD – Multimodal Imaging and Connectome Analysis Lab, McConnell Brain Imaging Centre, Montrea
Jordan deKraker, PhD – Multimodal Imaging and Connectome Analysis Lab, McConnell Brain Imaging Centre, Montrea
Donna G. Cabalo, PhD – Multimodal Imaging and Connectome Analysis Lab, McConnell Brain Imaging Centre, Montrea
Thaera Arafat, MD – Multimodal Imaging and Connectome Analysis Lab, McConnell Brain Imaging Centre, Montrea
Ella Sahlas, PhD student – Multimodal Imaging and Connectome Analysis Lab, McConnell Brain Imaging Centre, Montrea
Judy Chen, PhD student – Multimodal Imaging and Connectome Analysis Lab, McConnell Brain Imaging Centre, Montrea
Jessica Royer, PhD – Multimodal Imaging and Connectome Analysis Lab, McConnell Brain Imaging Centre, Montrea
Raul Rodriguez-Crucez, MD PhD – Multimodal Imaging and Connectome Analysis Lab, McConnell Brain Imaging Centre, Montrea
Fatemeh Fadaie, PhD – Multimodal Imaging and Connectome Analysis Lab, McConnell Brain Imaging Centre, Montrea
Andrea Bernasconi, MD – Neuroimaging of Epilepsy Laboratory and Department of Neurology and Neurosurgery, Montreal Neurological Institute, McGill University, Montreal, Canada.
Neda Bernasconi, MD, PhD – Neuroimaging of Epilepsy Laboratory and Department of Neurology and Neurosurgery, Montreal Neurological Institute, McGill University, Montreal, Canada.
Angelo Labate, MD, PhD – Regional Epilepsy Center, University of Messina, Messina, Italy
Boris Bernhardt, PhD – Multimodal Imaging and Connectome Analysis Lab, McConnell Brain Imaging Centre, Montrea
Rationale:
The piriform cortex (PC), known as “area tempestas” , is a key epileptogenic region in animal models. In humans, it has been less investigated, partly due to segmentation challenges in MRI. We combined BigBrain 3D histology and high-field MRI at 3T and 7T to characterize PC organization along a periallocortex–allocortex axis, test cross-scale correspondence, and localize alterations in patients with temporal lobe epilepsy (TLE).Methods:
In the 3D BigBrain dataset (100 μm), the left PC was labeled based on cytoarchitecture. A geodesic axis spanning periallocortex to allocortex was defined in both datesets. Vertex-wise depth profiles (16 intracortical surfaces in BigBrain and 14 inMRI data) were sampled from histology (Merker intensity) and MRI (T1 relaxometry). From these we derived depth-profile metrics and the first microstructural gradient (G1) via diffusion embedding. Axis–feature associations used Spearman’s ρ with Moran spectral randomization. Cross-scale validation summarized G1 along the axis in each dataset (uniform binning after min–max normalization) and correlated the resulting profiles with 1D Moran correction. MRI analyses included two cohorts, 7T (13 unilateral TLE, 21 healthy controls, HC) and 3 T (38 TLE, 47 HC). We carried out surface-based vertex-wise inference within the PC mask and complementary axis-based summaries in each dataset.Results:
In BigBrain, G1 closely tracked the periallo–allocortex axis (Spearman ρ ≈ −0.81; Moran-corrected p < 10⁻⁴), indicating that this axis captures the dominant cytoarchitectonic variation within human PC and that G1 largely reflects the allo-to-periallo transition recovered by diffusion embedding. Cross-scale agreement was high: binned G1 profiles from BigBrain and 7 T HC were strongly correlated (ρ ≈ 0.95; Moran-corrected p < 2×10⁻⁴).