Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Quantifying white matter hyperintensity and brain volumes in heterogeneous clinical and low-field portable MRI

Pablo Laso, Stefano Cerri, Annabel Sorby-Adams, Jennifer Guo, Farrah Mateen, Philipp Goebl, Jiaming Wu, Peirong Liu, Hongwei Li, Sean I. Young, Benjamin Billot, Oula Puonti, Gordon Sze, Sam Payabavash, Adam DeHavenon, Kevin N. Sheth, Matthew S. Rosen, John Kirsch, Nicola Strisciuglio, Jelmer M. Wolterink, Arman Eshaghi, Frederik Barkhof, W. Taylor Kimberly, Juan Eugenio Iglesias

TL;DR

This work tackles the challenge of quantifying white matter hyperintensity (WMH) and brain anatomy in heterogeneous clinical and low-field portable MRI (pMRI) scans, where traditional high-resolution MRI-based methods underperform. The authors introduce WMH-SynthSeg, a contrast- and resolution-agnostic CNN trained with synthetic data via domain-randomization-like augmentation that does not require retraining for new acquisition settings. The method jointly segments WMH and 36 brain ROIs, using a 3D U-net with a four-term loss and multi-task outputs, achieving strong cross-field correlations (e.g., WMH $\rho=0.85$ and hippocampal $r=0.89$) and competitive Dice scores on high-field data ($$\text{Anatomy }\text{Dice} \approx 0.85, \text{WMH }\text{Dice} \approx 0.62$$) while outperforming SAMSEG and LST-LPA on low-field scans. The work demonstrates robust WMH and anatomy quantification across a range of resolutions and contrasts, including 64mT pMRI, with potential for scalable brain injury tracking in underserved settings; the method is publicly available through FreeSurfer.

Abstract

Brain atrophy and white matter hyperintensity (WMH) are critical neuroimaging features for ascertaining brain injury in cerebrovascular disease and multiple sclerosis. Automated segmentation and quantification is desirable but existing methods require high-resolution MRI with good signal-to-noise ratio (SNR). This precludes application to clinical and low-field portable MRI (pMRI) scans, thus hampering large-scale tracking of atrophy and WMH progression, especially in underserved areas where pMRI has huge potential. Here we present a method that segments white matter hyperintensity and 36 brain regions from scans of any resolution and contrast (including pMRI) without retraining. We show results on eight public datasets and on a private dataset with paired high- and low-field scans (3T and 64mT), where we attain strong correlation between the WMH ($ρ$=.85) and hippocampal volumes (r=.89) estimated at both fields. Our method is publicly available as part of FreeSurfer, at: http://surfer.nmr.mgh.harvard.edu/fswiki/WMH-SynthSeg.

Quantifying white matter hyperintensity and brain volumes in heterogeneous clinical and low-field portable MRI

TL;DR

This work tackles the challenge of quantifying white matter hyperintensity (WMH) and brain anatomy in heterogeneous clinical and low-field portable MRI (pMRI) scans, where traditional high-resolution MRI-based methods underperform. The authors introduce WMH-SynthSeg, a contrast- and resolution-agnostic CNN trained with synthetic data via domain-randomization-like augmentation that does not require retraining for new acquisition settings. The method jointly segments WMH and 36 brain ROIs, using a 3D U-net with a four-term loss and multi-task outputs, achieving strong cross-field correlations (e.g., WMH and hippocampal ) and competitive Dice scores on high-field data () while outperforming SAMSEG and LST-LPA on low-field scans. The work demonstrates robust WMH and anatomy quantification across a range of resolutions and contrasts, including 64mT pMRI, with potential for scalable brain injury tracking in underserved settings; the method is publicly available through FreeSurfer.

Abstract

Brain atrophy and white matter hyperintensity (WMH) are critical neuroimaging features for ascertaining brain injury in cerebrovascular disease and multiple sclerosis. Automated segmentation and quantification is desirable but existing methods require high-resolution MRI with good signal-to-noise ratio (SNR). This precludes application to clinical and low-field portable MRI (pMRI) scans, thus hampering large-scale tracking of atrophy and WMH progression, especially in underserved areas where pMRI has huge potential. Here we present a method that segments white matter hyperintensity and 36 brain regions from scans of any resolution and contrast (including pMRI) without retraining. We show results on eight public datasets and on a private dataset with paired high- and low-field scans (3T and 64mT), where we attain strong correlation between the WMH (=.85) and hippocampal volumes (r=.89) estimated at both fields. Our method is publicly available as part of FreeSurfer, at: http://surfer.nmr.mgh.harvard.edu/fswiki/WMH-SynthSeg.
Paper Structure (11 sections, 1 equation, 2 figures, 2 tables)

This paper contains 11 sections, 1 equation, 2 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Methods
  3. Synthetic training data
  4. Model architecture and training
  5. EXPERIMENTS AND RESULTS
  6. Datasets
  7. Competing methods
  8. Experimental setup
  9. Results
  10. Conclusion
  11. ACKNOWLEDGMENTS

Figures (2)

  • Figure 1: Input, ground truth, and automated segmentations of a sample high-field scan from the Singapore dataset. The top row shows the high-resolution axial view; the bottom row shows a lower resolution orthogonal view (in sagittal orientation).
  • Figure 2: (a) High-field 1mm isotropic FLAIR from MGH dataset. (b) LST segmentation, used as ground truth for WMH. (c) High-field 1mm T1w. (d) FreeSurfer segmentation of (c), used for ground truth for anatomy. (e) pMRI of the same subject at 2x2x5.8mm axial resolution. (f) WMH-SynthSeg segmentation. We note that, despite affine alignment of the high-field images to the pMRI, the anatomy on the slices is slightly different due to nonlinear distortion.