Hierarchical Text-Guided Brain Tumor Segmentation via Sub-Region-Aware Prompts
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_ | 1866918401523318784 |
|---|---|
| author | Mohammadi, Bahram Huy, Ta Duc Sheikholeslami, Afrouz Chen, Qi Phan, Vu Minh Hieu White, Sam To, Minh-Son Zhang, Xuyun Beheshti, Amin Zhou, Luping Qi, Yuankai |
| author_facet | Mohammadi, Bahram Huy, Ta Duc Sheikholeslami, Afrouz Chen, Qi Phan, Vu Minh Hieu White, Sam To, Minh-Son Zhang, Xuyun Beheshti, Amin Zhou, Luping Qi, Yuankai |
| contents | Brain tumor segmentation remains challenging because the three standard sub-regions, i.e., whole tumor (WT), tumor core (TC), and enhancing tumor (ET), often exhibit ambiguous visual boundaries. Integrating radiological description texts with imaging has shown promise. However, most multimodal approaches typically compress a report into a single global text embedding shared across all sub-regions, overlooking their distinct clinical characteristics. We propose TextCSP (text-modulated soft cascade architecture), a hierarchical text-guided framework that builds on the TextBraTS baseline with three novel components: (1) a text-modulated soft cascade decoder that predicts WT->TC->ET in a coarse-to-fine manner consistent with their anatomical containment hierarchy. (2) sub-region-aware prompt tuning, which uses learnable soft prompts with a LoRA-adapted BioBERT encoder to generate specialized text representations tailored for each sub-region; (3) text-semantic channel modulators that convert the aforementioned representations into channel-wise refinement signals, enabling the decoder to emphasize features aligned with clinically described patterns. Experiments on the TextBraTS dataset demonstrate consistent improvements across all sub-regions against state-of-the-art methods by 1.7% and 6% on the main metrics Dice and HD95. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_21083 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Hierarchical Text-Guided Brain Tumor Segmentation via Sub-Region-Aware Prompts Mohammadi, Bahram Huy, Ta Duc Sheikholeslami, Afrouz Chen, Qi Phan, Vu Minh Hieu White, Sam To, Minh-Son Zhang, Xuyun Beheshti, Amin Zhou, Luping Qi, Yuankai Computer Vision and Pattern Recognition Brain tumor segmentation remains challenging because the three standard sub-regions, i.e., whole tumor (WT), tumor core (TC), and enhancing tumor (ET), often exhibit ambiguous visual boundaries. Integrating radiological description texts with imaging has shown promise. However, most multimodal approaches typically compress a report into a single global text embedding shared across all sub-regions, overlooking their distinct clinical characteristics. We propose TextCSP (text-modulated soft cascade architecture), a hierarchical text-guided framework that builds on the TextBraTS baseline with three novel components: (1) a text-modulated soft cascade decoder that predicts WT->TC->ET in a coarse-to-fine manner consistent with their anatomical containment hierarchy. (2) sub-region-aware prompt tuning, which uses learnable soft prompts with a LoRA-adapted BioBERT encoder to generate specialized text representations tailored for each sub-region; (3) text-semantic channel modulators that convert the aforementioned representations into channel-wise refinement signals, enabling the decoder to emphasize features aligned with clinically described patterns. Experiments on the TextBraTS dataset demonstrate consistent improvements across all sub-regions against state-of-the-art methods by 1.7% and 6% on the main metrics Dice and HD95. |
| title | Hierarchical Text-Guided Brain Tumor Segmentation via Sub-Region-Aware Prompts |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2603.21083 |