XLSTM-HVED: Cross-Modal Brain Tumor Segmentation and MRI Reconstruction Method Using Vision XLSTM and Heteromodal Variational Encoder-Decoder

TL;DR

The paper introduces XLSTM-HVED, a novel model combining vision XLSTM and heteromodal variational encoder-decoder to improve brain tumor segmentation and MRI reconstruction, especially with missing modalities.

Contribution

It proposes a new cross-modal framework with a Self-Attention Variational Encoder and SFECA module for enhanced tumor segmentation and MRI modality reconstruction.

Findings

Outperforms existing methods on BraTS 2024 dataset

Effectively handles missing MRI modalities

Improves segmentation accuracy and reconstruction quality

Abstract

Neurogliomas are among the most aggressive forms of cancer, presenting considerable challenges in both treatment and monitoring due to their unpredictable biological behavior. Magnetic resonance imaging (MRI) is currently the preferred method for diagnosing and monitoring gliomas. However, the lack of specific imaging techniques often compromises the accuracy of tumor segmentation during the imaging process. To address this issue, we introduce the XLSTM-HVED model. This model integrates a hetero-modal encoder-decoder framework with the Vision XLSTM module to reconstruct missing MRI modalities. By deeply fusing spatial and temporal features, it enhances tumor segmentation performance. The key innovation of our approach is the Self-Attention Variational Encoder (SAVE) module, which improves the integration of modal features. Additionally, it optimizes the interaction of features between segmentation and reconstruction tasks through the Squeeze-Fusion-Excitation Cross Awareness (SFECA) module. Our experiments using the BraTS 2024 dataset demonstrate that our model significantly outperforms existing advanced methods in handling cases where modalities are missing. Our source code is available at https://github.com/Quanato607/XLSTM-HVED.