Search papers, labs, and topics across Lattice.
This paper addresses the challenge of automated multiple sclerosis (MS) lesion segmentation on 7-tesla (7T) MRI, where existing tools trained on lower-field data perform suboptimally due to contrast and artifact differences. The authors trained 3D UNETR and SegFormer transformer-based models on 7T FLAIR images at various resolutions, using expert-revised Lesion Segmentation Tool (LST) outputs as ground truth. Results show that the 7T-trained transformers, particularly SegFormer, achieved competitive overlap with state-of-the-art LST-AI while recovering additional small lesions, demonstrating the potential of transformer-based models for improved lesion detection in ultra-high field MS imaging.
Transformer-based models trained directly on high-resolution 7T MRI data can spot MS lesions that classical methods miss, but not without introducing new challenges in boundary variability and artifact-related false positives.
Ultra-high field 7-tesla (7T) MRI improves visualization of multiple sclerosis (MS) white matter lesions (WML) but differs sufficiently in contrast and artifacts from 1.5-3T imaging - suggesting that widely used automated segmentation tools may not translate directly. We analyzed 7T FLAIR scans and generated reference WML masks from Lesion Segmentation Tool (LST) outputs followed by expert manual revision. As external comparators, we applied LST-LPA and the more recent LST-AI ensemble, both originally developed on lower-field data. We then trained 3D UNETR and SegFormer transformer-based models on 7T FLAIR at multiple resolutions (0.5x0.5x0.5^3, 1.0x1.0x1.0^3, and 1.5x1.5x2.0^3) and evaluated all methods using voxel-wise and lesion-wise metrics from the BraTS 2023 framework. On the held-out test set at native 0.5x0.5x0.5^3 resolution, 7T-trained transformers achieved competitive overlap with LST-AI while recovering additional small lesions that were missed by classical methods, at the cost of some boundary variability and occasional artifact-related false positives. On a held-out 7 T test set, our best transformer model (SegFormer) achieved a voxel-wise Dice of 0.61 and lesion-wise Dice of 0.20, improving on the classical LST-LPA tool (Dice 0.39, lesion-wise Dice 0.02). Performance decreased for models trained on downsampled images, underscoring the value of native 7T resolution for small-lesion detection. By releasing our 7T-trained models, we aim to provide a reproducible, ready-to-use resource for automated lesion quantification in ultra-high field MS research (https://github.com/maynord/7T-MS-lesion-segmentation).
