Convolutional-recurrent neural networks approximate diffusion tractography from T1-weighted MRI and associated anatomical context

Leon Yichen Cai, Ho Hin Lee, Nancy Rose Newlin, Cailey Irene Kerley, Praitayini Kanakaraj, Qi Yang, Graham Walter Johnson, Daniel Moyer, Kurt Gregory Schilling, Francois Rheault, Bennett A. Landman
Medical Imaging with Deep Learning, PMLR 227:1124-1143, 2024.

Abstract

Diffusion MRI (dMRI) streamline tractography is the gold-standard for in vivo estimation of white matter (WM) pathways in the brain. However, the high angular resolution dMRI acquisitions capable of fitting the microstructural models needed for tractography are often time-consuming and not routinely collected clinically, restricting the scope of tractography analyses. To address this limitation, we build on recent advances in deep learning which have demonstrated that streamline propagation can be learned from dMRI directly without traditional model fitting. Specifically, we propose learning the streamline propagator from T1w MRI to facilitate arbitrary tractography analyses when dMRI is unavailable. To do so, we present a novel convolutional-recurrent neural network (CoRNN) trained in a teacher-student framework that leverages T1w MRI, associated anatomical context, and streamline memory from data acquired for the Human Connectome Project. We characterize our approach under two common tractography paradigms, WM bundle analysis and structural connectomics, and find approximately a 5-15% difference between measures computed from streamlines generated with our approach and those generated using traditional dMRI tractography. When placed in the literature, these results suggest that the accuracy of WM measures computed from T1w MRI with our method is on the level of scan-rescan dMRI variability and raise an important question: is tractography truly a microstructural phenomenon, or has dMRI merely facilitated its discovery and implementation?

Cite this Paper

BibTeX
@InProceedings{pmlr-v227-cai24a, title = {Convolutional-recurrent neural networks approximate diffusion tractography from T1-weighted MRI and associated anatomical context}, author = {Cai, Leon Yichen and Lee, Ho Hin and Newlin, Nancy Rose and Kerley, Cailey Irene and Kanakaraj, Praitayini and Yang, Qi and Johnson, Graham Walter and Moyer, Daniel and Schilling, Kurt Gregory and Rheault, Francois and Landman, Bennett A.}, booktitle = {Medical Imaging with Deep Learning}, pages = {1124--1143}, year = {2024}, editor = {Oguz, Ipek and Noble, Jack and Li, Xiaoxiao and Styner, Martin and Baumgartner, Christian and Rusu, Mirabela and Heinmann, Tobias and Kontos, Despina and Landman, Bennett and Dawant, Benoit}, volume = {227}, series = {Proceedings of Machine Learning Research}, month = {10--12 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v227/cai24a/cai24a.pdf}, url = {https://proceedings.mlr.press/v227/cai24a.html}, abstract = {Diffusion MRI (dMRI) streamline tractography is the gold-standard for in vivo estimation of white matter (WM) pathways in the brain. However, the high angular resolution dMRI acquisitions capable of fitting the microstructural models needed for tractography are often time-consuming and not routinely collected clinically, restricting the scope of tractography analyses. To address this limitation, we build on recent advances in deep learning which have demonstrated that streamline propagation can be learned from dMRI directly without traditional model fitting. Specifically, we propose learning the streamline propagator from T1w MRI to facilitate arbitrary tractography analyses when dMRI is unavailable. To do so, we present a novel convolutional-recurrent neural network (CoRNN) trained in a teacher-student framework that leverages T1w MRI, associated anatomical context, and streamline memory from data acquired for the Human Connectome Project. We characterize our approach under two common tractography paradigms, WM bundle analysis and structural connectomics, and find approximately a 5-15% difference between measures computed from streamlines generated with our approach and those generated using traditional dMRI tractography. When placed in the literature, these results suggest that the accuracy of WM measures computed from T1w MRI with our method is on the level of scan-rescan dMRI variability and raise an important question: is tractography truly a microstructural phenomenon, or has dMRI merely facilitated its discovery and implementation?} }
Endnote
%0 Conference Paper %T Convolutional-recurrent neural networks approximate diffusion tractography from T1-weighted MRI and associated anatomical context %A Leon Yichen Cai %A Ho Hin Lee %A Nancy Rose Newlin %A Cailey Irene Kerley %A Praitayini Kanakaraj %A Qi Yang %A Graham Walter Johnson %A Daniel Moyer %A Kurt Gregory Schilling %A Francois Rheault %A Bennett A. Landman %B Medical Imaging with Deep Learning %C Proceedings of Machine Learning Research %D 2024 %E Ipek Oguz %E Jack Noble %E Xiaoxiao Li %E Martin Styner %E Christian Baumgartner %E Mirabela Rusu %E Tobias Heinmann %E Despina Kontos %E Bennett Landman %E Benoit Dawant %F pmlr-v227-cai24a %I PMLR %P 1124--1143 %U https://proceedings.mlr.press/v227/cai24a.html %V 227 %X Diffusion MRI (dMRI) streamline tractography is the gold-standard for in vivo estimation of white matter (WM) pathways in the brain. However, the high angular resolution dMRI acquisitions capable of fitting the microstructural models needed for tractography are often time-consuming and not routinely collected clinically, restricting the scope of tractography analyses. To address this limitation, we build on recent advances in deep learning which have demonstrated that streamline propagation can be learned from dMRI directly without traditional model fitting. Specifically, we propose learning the streamline propagator from T1w MRI to facilitate arbitrary tractography analyses when dMRI is unavailable. To do so, we present a novel convolutional-recurrent neural network (CoRNN) trained in a teacher-student framework that leverages T1w MRI, associated anatomical context, and streamline memory from data acquired for the Human Connectome Project. We characterize our approach under two common tractography paradigms, WM bundle analysis and structural connectomics, and find approximately a 5-15% difference between measures computed from streamlines generated with our approach and those generated using traditional dMRI tractography. When placed in the literature, these results suggest that the accuracy of WM measures computed from T1w MRI with our method is on the level of scan-rescan dMRI variability and raise an important question: is tractography truly a microstructural phenomenon, or has dMRI merely facilitated its discovery and implementation?
APA
Cai, L.Y., Lee, H.H., Newlin, N.R., Kerley, C.I., Kanakaraj, P., Yang, Q., Johnson, G.W., Moyer, D., Schilling, K.G., Rheault, F. & Landman, B.A.. (2024). Convolutional-recurrent neural networks approximate diffusion tractography from T1-weighted MRI and associated anatomical context. Medical Imaging with Deep Learning, in Proceedings of Machine Learning Research 227:1124-1143 Available from https://proceedings.mlr.press/v227/cai24a.html.

Related Material