What does this research mean for the field?

A multi-sequence MRI integration framework using the SwinUNETR-v2 backbone outperforms existing state-of-the-art segmentation methods in identifying multiple sclerosis lesions, achieving a peak Dice Similarity Coefficient of 90.7%. Novelty: ClaimNovelty.METHODOLOGICAL. Consensus alignment: ConsensusAlignment.NEUTRAL.

What question did this study set out to answer?

This study aims to develop a multi-sequence integration framework for accurate lesion segmentation in multiple sclerosis using AI.

May 30, 2026Open Access

A multi-sequence MRI integration framework using SwinUNETR-v2 for multiple sclerosis lesion segmentation

Key Points

This study aims to develop a multi-sequence integration framework for accurate lesion segmentation in multiple sclerosis using AI.
Developed a multi-sequence framework integrating four MRI modalities with SwinUNETR-v2.
Implemented a task-oriented input design with balanced foreground/background patch extraction.
Conducted evaluations using a five-fold controlled protocol.
Achieved peak Dice Similarity Coefficient (DSC) of 90.7% and mean DSC of 88.3%.
Outperformed state-of-the-art segmentation methods in all key metrics.
Demonstrated strong effectiveness in identifying MS lesions.

Abstract

Multiple Sclerosis (MS) is a chronic brain disease that affects the brain and spinal cord, where Magnetic Resonance Imaging (MRI) plays a key role in diagnosis. While manual analysis of brain MRIs is important, it is time-consuming and prone to human error. Artificial Intelligence (AI)-driven Computer Aided Diagnostic (CAD) systems have therefore gained traction due to their ability to provide more consistent and reliable assessments. This study presents a multi-sequence framework that integrates four MRI modalities with the SwinUNETR-v2 backbone for MS lesion segmentation. The main contribution is a task-oriented integration of multi-sequence input design (implemented as a four-channel volume representation), refined preprocessing (including balanced foreground/background patch extraction), and a weighted loss formulation under a controlled five-fold evaluation protocol. This approach achieved a peak DSC of 90.7% and a mean DSC of 88.3%. Moreover, when compared against other state-of-the-art segmentation methods—including AttentionUNet, DenseResidualUNet, SegResNet, FCNN, and nnUNet-v2—the multi-sequence SwinUNETR-v2 setup consistently outperformed these models across all key metrics, demonstrating strong effectiveness in identifying MS lesions.

KI fragen

Bookmark

View Full Paper