الملخص
Pretrained 2D foundation models offer a practical alternative to dedicated 3D pretraining for brain structural magnetic resonance imaging (sMRI), but their use on volumetric data requires bridging the mismatch between a 2D encoder and a 3D volume input. Existing methods typically encode slices independently and integrate their features afterwards. We introduce Multiscale Volumetric Reduction (MVR), a reduce-then-encode approach that compresses each anatomical view from (D) slices into (M << D) complementary 2D components before foundation-model encoding. MVR combines an uncentered-PCA base component derived from the original through-plane intensities with residual detail components constructed from multiscale spatial descriptors. The reduction is estimated from the training volumes without diagnostic labels or gradient-based optimization and remains fixed thereafter. The resulting components are independently processed by a shared frozen 2D foundation model and concatenated for linear probing. Under this frozen-encoder setting, MVR achieves strong overall performance across ADNI, OASIS, and ABIDE relative to the evaluated 2D-to-3D adaptation methods and simple input-reduction baselines, while also generalizing strongly from ADNI to AIBL.
الكلمات المفتاحية
الموضوع
بيانات النشر
- المجلة
- غير متاح
- وصول مفتوح
- وصول مفتوح أخضر
اقتبس هذه المقالة
APA 7
Ding, D., Qi, Y., Wang, B., Zhou, L., & Beheshti, A. (2026). Reduce, Then Encode: Multiscale Volumetric Reduction for 2D Foundation Models in Brain MRI. https://omanscience.com/ar/articles/reduce-then-encode-multiscale-volumetric-reduction-for-2d-foundation-models-in-brain-mri
MLA 9
Ding, Dexuan, et al. "Reduce, Then Encode: Multiscale Volumetric Reduction for 2D Foundation Models in Brain MRI." https://omanscience.com/ar/articles/reduce-then-encode-multiscale-volumetric-reduction-for-2d-foundation-models-in-brain-mri.
شيكاغو (المؤلف–التاريخ)
Ding, Dexuan, Yuankai Qi, Bogong Wang, Luping Zhou, and Amin Beheshti. 2026. "Reduce, Then Encode: Multiscale Volumetric Reduction for 2D Foundation Models in Brain MRI." https://omanscience.com/ar/articles/reduce-then-encode-multiscale-volumetric-reduction-for-2d-foundation-models-in-brain-mri.
هارفارد
Ding, D., Qi, Y., Wang, B., Zhou, L. and Beheshti, A. (2026) 'Reduce, Then Encode: Multiscale Volumetric Reduction for 2D Foundation Models in Brain MRI', Available at: https://omanscience.com/ar/articles/reduce-then-encode-multiscale-volumetric-reduction-for-2d-foundation-models-in-brain-mri.
فانكوفر
Ding D, Qi Y, Wang B, Zhou L, Beheshti A. Reduce, Then Encode: Multiscale Volumetric Reduction for 2D Foundation Models in Brain MRI. https://omanscience.com/ar/articles/reduce-then-encode-multiscale-volumetric-reduction-for-2d-foundation-models-in-brain-mri
IEEE
D. Ding, Y. Qi, B. Wang, L. Zhou, and A. Beheshti, "Reduce, Then Encode: Multiscale Volumetric Reduction for 2D Foundation Models in Brain MRI," https://omanscience.com/ar/articles/reduce-then-encode-multiscale-volumetric-reduction-for-2d-foundation-models-in-brain-mri.