NEMESIS: Noise-suppressed Efficient MAE with Enhanced Superpatch Integration Strategy
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2026
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866910096924082176 |
|---|---|
| author | Kim, Kyeonghun Jung, Hyeonseok Han, Youngung Go, Hyunsu Choi, Eunseob Park, Seongbin Lim, Junsu Yang, Jiwon Lee, Sumin Hwang, Insung Liao, Ken Ying-Kai Kim, Nam-Joon |
| author_facet | Kim, Kyeonghun Jung, Hyeonseok Han, Youngung Go, Hyunsu Choi, Eunseob Park, Seongbin Lim, Junsu Yang, Jiwon Lee, Sumin Hwang, Insung Liao, Ken Ying-Kai Kim, Nam-Joon |
| contents | Volumetric CT imaging is essential for clinical diagnosis, yet annotating 3D volumes is expensive and time-consuming, motivating self-supervised learning (SSL) from unlabeled data. However, applying SSL to 3D CT remains challenging due to the high memory cost of full-volume transformers and the anisotropic spatial structure of CT data, which is not well captured by conventional masking strategies. We propose NEMESIS, a masked autoencoder (MAE) framework that operates on local 128x128x128 superpatches, enabling memory-efficient training while preserving anatomical detail. NEMESIS introduces three key components: (i) noise-enhanced reconstruction as a pretext task, (ii) Masked Anatomical Transformer Blocks (MATB) that perform dual-masking through parallel plane-wise and axis-wise token removal, and (iii) NEMESIS Tokens (NT) for cross-scale context aggregation. On the BTCV multi-organ classification benchmark, NEMESIS with a frozen backbone and a linear classifier achieves a mean AUROC of 0.9633, surpassing fully fine-tuned SuPreM (0.9493) and VoCo (0.9387). Under a low-label regime with only 10% of available annotations, it retains an AUROC of 0.9075, demonstrating strong label efficiency. Furthermore, the superpatch-based design reduces computational cost to 31.0 GFLOPs per forward pass, compared to 985.8 GFLOPs for the full-volume baseline, providing a scalable and robust foundation for 3D medical imaging. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_01612 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | NEMESIS: Noise-suppressed Efficient MAE with Enhanced Superpatch Integration Strategy Kim, Kyeonghun Jung, Hyeonseok Han, Youngung Go, Hyunsu Choi, Eunseob Park, Seongbin Lim, Junsu Yang, Jiwon Lee, Sumin Hwang, Insung Liao, Ken Ying-Kai Kim, Nam-Joon Computer Vision and Pattern Recognition Artificial Intelligence Volumetric CT imaging is essential for clinical diagnosis, yet annotating 3D volumes is expensive and time-consuming, motivating self-supervised learning (SSL) from unlabeled data. However, applying SSL to 3D CT remains challenging due to the high memory cost of full-volume transformers and the anisotropic spatial structure of CT data, which is not well captured by conventional masking strategies. We propose NEMESIS, a masked autoencoder (MAE) framework that operates on local 128x128x128 superpatches, enabling memory-efficient training while preserving anatomical detail. NEMESIS introduces three key components: (i) noise-enhanced reconstruction as a pretext task, (ii) Masked Anatomical Transformer Blocks (MATB) that perform dual-masking through parallel plane-wise and axis-wise token removal, and (iii) NEMESIS Tokens (NT) for cross-scale context aggregation. On the BTCV multi-organ classification benchmark, NEMESIS with a frozen backbone and a linear classifier achieves a mean AUROC of 0.9633, surpassing fully fine-tuned SuPreM (0.9493) and VoCo (0.9387). Under a low-label regime with only 10% of available annotations, it retains an AUROC of 0.9075, demonstrating strong label efficiency. Furthermore, the superpatch-based design reduces computational cost to 31.0 GFLOPs per forward pass, compared to 985.8 GFLOPs for the full-volume baseline, providing a scalable and robust foundation for 3D medical imaging. |
| title | NEMESIS: Noise-suppressed Efficient MAE with Enhanced Superpatch Integration Strategy |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence |
| url | https://arxiv.org/abs/2604.01612 |