DALight-3D: A Lightweight 3D U-Net for Brain Tumor Segmentation from Multi-Modal MRI
Fuente:
arXiv
Salvato in:
| Autori principali: | , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2026
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866915983719923712 |
|---|---|
| author | Mishra, Nand Kumar Mishra, Dhruv Singh, Dr Manu Pratap |
| author_facet | Mishra, Nand Kumar Mishra, Dhruv Singh, Dr Manu Pratap |
| contents | Automatic brain tumor segmentation from multi-modal MRI remains challenging because volumetric models often incur substantial computational cost. This paper presents DALight-3D, a compact 3D U-Net variant that combines depthwise separable 3D convolutions, identifier-conditioned normalization, cross-slice attention, and adaptive skip fusion. The method is evaluated on the Medical Segmentation Decathlon Task01 BrainTumour benchmark under matched optimization settings against standard 3D U-Net, Attention U-Net, Residual 3D U-Net, and V-Net baselines. In the reported 50-epoch comparison, DALight-3D achieves a mean Dice of 0.727 with 2.22M parameters, compared with 0.710 Dice and 3.20M parameters for Residual 3D U-Net. Component-wise ablations show consistent performance degradation when SepConv, identifier-conditioned normalization, CSA, or SSFB is removed. These results indicate that DALight-3D offers a favorable accuracy-efficiency trade-off within the present benchmark setting. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_04518 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | DALight-3D: A Lightweight 3D U-Net for Brain Tumor Segmentation from Multi-Modal MRI Mishra, Nand Kumar Mishra, Dhruv Singh, Dr Manu Pratap Computer Vision and Pattern Recognition Machine Learning Neural and Evolutionary Computing Automatic brain tumor segmentation from multi-modal MRI remains challenging because volumetric models often incur substantial computational cost. This paper presents DALight-3D, a compact 3D U-Net variant that combines depthwise separable 3D convolutions, identifier-conditioned normalization, cross-slice attention, and adaptive skip fusion. The method is evaluated on the Medical Segmentation Decathlon Task01 BrainTumour benchmark under matched optimization settings against standard 3D U-Net, Attention U-Net, Residual 3D U-Net, and V-Net baselines. In the reported 50-epoch comparison, DALight-3D achieves a mean Dice of 0.727 with 2.22M parameters, compared with 0.710 Dice and 3.20M parameters for Residual 3D U-Net. Component-wise ablations show consistent performance degradation when SepConv, identifier-conditioned normalization, CSA, or SSFB is removed. These results indicate that DALight-3D offers a favorable accuracy-efficiency trade-off within the present benchmark setting. |
| title | DALight-3D: A Lightweight 3D U-Net for Brain Tumor Segmentation from Multi-Modal MRI |
| topic | Computer Vision and Pattern Recognition Machine Learning Neural and Evolutionary Computing |
| url | https://arxiv.org/abs/2605.04518 |