Understanding Alzheimer disease’s structural connectivity through explainable AI

Achraf Essemlali, Etienne St-Onge, Maxime Descoteaux, Pierre-Marc Jodoin
Proceedings of the Third Conference on Medical Imaging with Deep Learning, PMLR 121:217-229, 2020.

Abstract

In the following work, we use a modified version of deep BrainNet convolutional neural network (CNN) trained on the diffusion weighted MRI (DW-MRI) tractography connectomes of patients with Alzheimer’s Disease (AD) and Mild Cognitive Impairment (MCI) to better understand the structural connectomics of that disease. We show that with a relatively simple connectomic BrainNetCNN used to classify brain images and explainable AI techniques, one can underline brain regions and their connectivity involved in AD. Results reveal that the connected regions with high structural differences between groups are those also reported in previous AD literature. Our findings support that deep learning over structural connectomes is a powerful tool to leverage the complex structure within connectomes derived from diffusion MRI tractography. To our knowledge, our contribution is the first explainable AI work applied to structural analysis of a degenerative disease.

Cite this Paper

BibTeX
@InProceedings{pmlr-v121-essemlali20a, title = {Understanding Alzheimer disease’s structural connectivity through explainable AI}, author = {Essemlali, Achraf and St-Onge, Etienne and Descoteaux, Maxime and Jodoin, Pierre-Marc}, booktitle = {Proceedings of the Third Conference on Medical Imaging with Deep Learning}, pages = {217--229}, year = {2020}, editor = {Arbel, Tal and Ben Ayed, Ismail and de Bruijne, Marleen and Descoteaux, Maxime and Lombaert, Herve and Pal, Christopher}, volume = {121}, series = {Proceedings of Machine Learning Research}, month = {06--08 Jul}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v121/essemlali20a/essemlali20a.pdf}, url = {https://proceedings.mlr.press/v121/essemlali20a.html}, abstract = {In the following work, we use a modified version of deep BrainNet convolutional neural network (CNN) trained on the diffusion weighted MRI (DW-MRI) tractography connectomes of patients with Alzheimer’s Disease (AD) and Mild Cognitive Impairment (MCI) to better understand the structural connectomics of that disease. We show that with a relatively simple connectomic BrainNetCNN used to classify brain images and explainable AI techniques, one can underline brain regions and their connectivity involved in AD. Results reveal that the connected regions with high structural differences between groups are those also reported in previous AD literature. Our findings support that deep learning over structural connectomes is a powerful tool to leverage the complex structure within connectomes derived from diffusion MRI tractography. To our knowledge, our contribution is the first explainable AI work applied to structural analysis of a degenerative disease.} }
Endnote
%0 Conference Paper %T Understanding Alzheimer disease’s structural connectivity through explainable AI %A Achraf Essemlali %A Etienne St-Onge %A Maxime Descoteaux %A Pierre-Marc Jodoin %B Proceedings of the Third Conference on Medical Imaging with Deep Learning %C Proceedings of Machine Learning Research %D 2020 %E Tal Arbel %E Ismail Ben Ayed %E Marleen de Bruijne %E Maxime Descoteaux %E Herve Lombaert %E Christopher Pal %F pmlr-v121-essemlali20a %I PMLR %P 217--229 %U https://proceedings.mlr.press/v121/essemlali20a.html %V 121 %X In the following work, we use a modified version of deep BrainNet convolutional neural network (CNN) trained on the diffusion weighted MRI (DW-MRI) tractography connectomes of patients with Alzheimer’s Disease (AD) and Mild Cognitive Impairment (MCI) to better understand the structural connectomics of that disease. We show that with a relatively simple connectomic BrainNetCNN used to classify brain images and explainable AI techniques, one can underline brain regions and their connectivity involved in AD. Results reveal that the connected regions with high structural differences between groups are those also reported in previous AD literature. Our findings support that deep learning over structural connectomes is a powerful tool to leverage the complex structure within connectomes derived from diffusion MRI tractography. To our knowledge, our contribution is the first explainable AI work applied to structural analysis of a degenerative disease.
APA
Essemlali, A., St-Onge, E., Descoteaux, M. & Jodoin, P.. (2020). Understanding Alzheimer disease’s structural connectivity through explainable AI. Proceedings of the Third Conference on Medical Imaging with Deep Learning, in Proceedings of Machine Learning Research 121:217-229 Available from https://proceedings.mlr.press/v121/essemlali20a.html.

Related Material