Multimodal Visual Surrogate Compression for Alzheimer's Disease Classification
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866917232238395392 |
|---|---|
| author | Ding, Dexuan Peng, Ciyuan Kuantama, Endrowednes Guo, Jingcai Wu, Jia Yang, Jian Beheshti, Amin Yang, Ming-Hsuan Qi, Yuankai |
| author_facet | Ding, Dexuan Peng, Ciyuan Kuantama, Endrowednes Guo, Jingcai Wu, Jia Yang, Jian Beheshti, Amin Yang, Ming-Hsuan Qi, Yuankai |
| contents | High-dimensional structural MRI (sMRI) images are widely used for Alzheimer's Disease (AD) diagnosis. Most existing methods for sMRI representation learning rely on 3D architectures (e.g., 3D CNNs), slice-wise feature extraction with late aggregation, or apply training-free feature extractions using 2D foundation models (e.g., DINO). However, these three paradigms suffer from high computational cost, loss of cross-slice relations, and limited ability to extract discriminative features, respectively. To address these challenges, we propose Multimodal Visual Surrogate Compression (MVSC). It learns to compress and adapt large 3D sMRI volumes into compact 2D features, termed as visual surrogates, which are better aligned with frozen 2D foundation models to extract powerful representations for final AD classification. MVSC has two key components: a Volume Context Encoder that captures global cross-slice context under textual guidance, and an Adaptive Slice Fusion module that aggregates slice-level information in a text-enhanced, patch-wise manner. Extensive experiments on three large-scale Alzheimer's disease benchmarks demonstrate our MVSC performs favourably on both binary and multi-class classification tasks compared against state-of-the-art methods. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_21673 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Multimodal Visual Surrogate Compression for Alzheimer's Disease Classification Ding, Dexuan Peng, Ciyuan Kuantama, Endrowednes Guo, Jingcai Wu, Jia Yang, Jian Beheshti, Amin Yang, Ming-Hsuan Qi, Yuankai Computer Vision and Pattern Recognition High-dimensional structural MRI (sMRI) images are widely used for Alzheimer's Disease (AD) diagnosis. Most existing methods for sMRI representation learning rely on 3D architectures (e.g., 3D CNNs), slice-wise feature extraction with late aggregation, or apply training-free feature extractions using 2D foundation models (e.g., DINO). However, these three paradigms suffer from high computational cost, loss of cross-slice relations, and limited ability to extract discriminative features, respectively. To address these challenges, we propose Multimodal Visual Surrogate Compression (MVSC). It learns to compress and adapt large 3D sMRI volumes into compact 2D features, termed as visual surrogates, which are better aligned with frozen 2D foundation models to extract powerful representations for final AD classification. MVSC has two key components: a Volume Context Encoder that captures global cross-slice context under textual guidance, and an Adaptive Slice Fusion module that aggregates slice-level information in a text-enhanced, patch-wise manner. Extensive experiments on three large-scale Alzheimer's disease benchmarks demonstrate our MVSC performs favourably on both binary and multi-class classification tasks compared against state-of-the-art methods. |
| title | Multimodal Visual Surrogate Compression for Alzheimer's Disease Classification |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2601.21673 |