Impact of white matter hyperintensities on tractography bundle segmentation in aging population
Poster No:
1275
Submission Type:
Abstract Submission
Authors:
Stanislas Thoumyre1,2, François Rheault1,3, Laurent Petit2,4
Institutions:
1Sherbrooke Connectivity Imaging Lab (SCIL), Université de Sherbrooke, Sherbrooke, Canada, 2Groupe d'Imagerie Neurofonctionnelle (GIN), UMR 5293, Université de Bordeaux, Bordeaux, France, 3IRP OpTeam, Université de Sherbrooke, Sherbrooke, Canada, 4IRP OpTeam, CNRS Biologie,, Bordeaux, France
First Author:
Stanislas Thoumyre
Sherbrooke Connectivity Imaging Lab (SCIL), Université de Sherbrooke|Groupe d'Imagerie Neurofonctionnelle (GIN), UMR 5293, Université de Bordeaux
Sherbrooke, Canada|Bordeaux, France
Sherbrooke Connectivity Imaging Lab (SCIL), Université de Sherbrooke|Groupe d'Imagerie Neurofonctionnelle (GIN), UMR 5293, Université de Bordeaux
Sherbrooke, Canada|Bordeaux, France
Co-Author(s):
François Rheault
Sherbrooke Connectivity Imaging Lab (SCIL), Université de Sherbrooke|IRP OpTeam, Université de Sherbrooke
Sherbrooke, Canada|Sherbrooke, Canada
Sherbrooke Connectivity Imaging Lab (SCIL), Université de Sherbrooke|IRP OpTeam, Université de Sherbrooke
Sherbrooke, Canada|Sherbrooke, Canada
Laurent Petit
Groupe d'Imagerie Neurofonctionnelle (GIN), UMR 5293, Université de Bordeaux|IRP OpTeam, CNRS Biologie,
Bordeaux, France|Bordeaux, France
Groupe d'Imagerie Neurofonctionnelle (GIN), UMR 5293, Université de Bordeaux|IRP OpTeam, CNRS Biologie,
Bordeaux, France|Bordeaux, France
Introduction:
Probabilistic tractography profiling provides a thorough analysis of white matter lesions by leveraging the entire white matter volume (Kuchling et al., 2018). However, limitations in tractography algorithms can affect reliability (Rheault et al., 2020; Schilling et al., 2018). White matter segmentation is particularly challenging, especially in aging studies where tissue contrast-based segmentation (e.g., FAST) struggles with intensity changes in white matter hyperintensities (WMH), complicating the examination of white matter structure (Matijevic & Ryan, 2021). Still, WMH is considered a viable tissue for tractography (Humphreys et al., 2021) and tractography is effective within these lesions (Theaud et al., 2017). The current study aims to show the significance of correcting white matter masks in the presence of WMH in an aging population to improve bundle segmentation reliability.
Methods:
The study included 26 cognitively normal participants (16 men, 10 women; mean age: 71.03 ± 5.1 years) from ADNI 2/GO (Beckett et al., 2015). Each individual presents 2 acquisitions (with 20 subjects having a third acquisition) with ~3 months between each acquisition. Diffusion images were acquired using a 3 Tesla MRI scanner with: T1 MPRAGE sequence (1mm iso), FLAIR sequence (0.9x0.9x5mm), diffusion spin-echo echo-planar sequence (64 dir, bval=1000 s/mm², 2mm iso). Lesions were segmented from T1 and FLAIR using the SHIVA-WMH tool (Tsuchida et al., 2023). Data processing was performed with the tractoflow pipeline (Theaud et al., 2020) executed 3 times with 3 different segmentation methods (fig. 1.A): FAST, recon-all-clinical and recon-all-clinical plus lesion mask integration. Probabilistic tractography was conducted with a local tracking model. 47 bundles were segmented using BundleSeg (St-Onge et al., 2023), and for each bundle, streamline count and volume were quantified using Tractometry-flow. Comparisons across time points and segmentation methods were based on streamline counts, bundle volume and Dice coefficients for bundle volume overlap.
Results:
FAST-based segmentation includes minimal or no WMH (Fig 1A) but is more permissive at the gyral level. Recon-all-clinical enables lesion correction with a more restrictive segmentation and requires adding a lesion mask for complete correction. Without correction, significant tracking errors can occur (Fig 1B), affecting peripheral bundles near the anterior and posterior horns of the lateral ventricles, with a substantial portion of white matter impacted (Fig 1C). FAST segmentation generally shows higher bundle volumes and streamline counts, with greater variability. Reproducibility remains consistent, with a Dice coefficient of ~0.78 across methods (Fig 2B) among bundles. High lesion-contact bundles like the IFOF show reconstruction defects with FAST (Fig 2A), reducing reproducibility to ~0.72 (Fig 2B). Corrected segmentation maintains a dice of ~0.78, effectively improving accuracy and reproducibility.
·Figure 1: Impact of white matter segmentation methods on the tractogram.
·Figure 2 : Impact of segmentation methods on the bundle segmentation.
Conclusions:
Our results highlight the need to correct WM masks for probabilistic tractography to study WMH in an aging population. Segmentation using tissue contrast (FAST) has its limitations, with high inter-subject variability and reconstruction defects when lesions are present in the bundles. Segmentation based on an atlas is therefore more appropriate for an aging population but all segmentation tools are imperfect, so it is essential to verify if yours fits your population. Also, it is often better to segment lesions separately, then add them to the white matter mask. Our goal was to reduce algorithmic variability and segmentation pitfalls; we can now be more confident in our interpretations of the microstructural changes due to WMH. The next steps are to conduct a longer-term longitudinal study and quantify the impact of white matter hyperintensities on brain connectivity within bundles, in order to draw up a lesion profile.
Lifespan Development:
Aging
Modeling and Analysis Methods:
Diffusion MRI Modeling and Analysis 1
Segmentation and Parcellation 2
Neuroanatomy, Physiology, Metabolism and Neurotransmission:
White Matter Anatomy, Fiber Pathways and Connectivity
Keywords:
Aging
MRI
Segmentation
Tractography
White Matter
WHITE MATTER IMAGING - DTI, HARDI, DSI, ETC
1|2Indicates the priority used for review
Abstract Information
By submitting your proposal, you grant permission for the Organization for Human Brain Mapping (OHBM) to distribute your work in any format, including video, audio print and electronic text through OHBM OnDemand, social media channels, the OHBM website, or other electronic publications and media.
The Open Science Special Interest Group (OSSIG) is introducing a reproducibility challenge for OHBM 2025. This new initiative aims to enhance the reproducibility of scientific results and foster collaborations between labs. Teams will consist of a “source” party and a “reproducing” party, and will be evaluated on the success of their replication, the openness of the source work, and additional deliverables. Click here for more information. Propose your OHBM abstract(s) as source work for future OHBM meetings by selecting one of the following options:
Please indicate below if your study was a "resting state" or "task-activation” study.
Healthy subjects only or patients (note that patient studies may also involve healthy subjects):
Was this research conducted in the United States?
Were any human subjects research approved by the relevant Institutional Review Board or ethics panel? NOTE: Any human subjects studies without IRB approval will be automatically rejected.
Were any animal research approved by the relevant IACUC or other animal research panel? NOTE: Any animal studies without IACUC approval will be automatically rejected.
Please indicate which methods were used in your research:
For human MRI, what field strength scanner do you use?
Which processing packages did you use for your study?
Provide references using APA citation style.
Humphreys, C. A., Smith, C., & Wardlaw, J. M. (2021). Correlations in post-mortem imaging-histopathology studies of sporadic human cerebral small vessel disease: A systematic review. Neuropathology and Applied Neurobiology, 47(7), 910–930.
Kuchling, J., Backner, Y., Oertel, F. C., Raz, N., Bellmann-Strobl, J., Ruprecht, K., Paul, F., Levin, N., Brandt, A. U., & Scheel, M. (2018). Comparison of probabilistic tractography and tract-based spatial statistics for assessing optic radiation damage in patients with autoimmune inflammatory disorders of the central nervous system. NeuroImage: Clinical, 19, 538–550.
Matijevic, S., & Ryan, L. (2021). Tract Specificity of Age Effects on Diffusion Tensor Imaging Measures of White Matter Health. Frontiers in Aging Neuroscience, 13, 628865.
Rheault, F., Poulin, P., Caron, A. V., St-Onge, E., & Descoteaux, M. (2020). Common misconceptions, hidden biases and modern challenges of dMRI tractography. Journal of Neural Engineering, 17(1), 011001.
Schilling, K. G., Nath, V., … Landman, B. A. (2018). Limits to anatomical accuracy of diffusion tractography using modern approaches. NeuroImage, 185, 1.
St-Onge, E., Schilling, K. G., & Rheault, F. (2023). BundleSeg: A Versatile, Reliable and Reproducible Approach to White Matter Bundle Segmentation. In M. Karaman, R. Mito, E. Powell, F. Rheault, & S. Winzeck (Eds.), Computational Diffusion MRI (pp. 47–57). Springer Nature Switzerland.
Theaud, G., Dilharreguy, B., Catheline, G., & Descoteaux, M. (2017, January 1). Impact of white-matter hyperintensities on tractography. 25th Annual Meeting of the International Society for Magnetic Resonance in Medicine (ISMRM).
Theaud, G., Houde, J.-C., Boré, A., Rheault, F., Morency, F., & Descoteaux, M. (2020). TractoFlow: A robust, efficient and reproducible diffusion MRI pipeline leveraging Nextflow & Singularity. NeuroImage, 218, 116889.
Tsuchida, A., Boutinaud, P., Verrecchia, V., Tzourio, C., Debette, S., & Joliot, M. (2023). Early detection of white matter hyperintensities using SHIVA‐WMH detector. Human Brain Mapping, 45(1), e26548.