Guardado en:
Detalles Bibliográficos
Autores principales: Wang, Jianwei, Wang, Qing, Ruan, Menglan, Ge, Rongjun, Yang, Chunfeng, Chen, Yang, Xie, Chunming
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:https://arxiv.org/abs/2512.08337
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866914189514113024
author Wang, Jianwei
Wang, Qing
Ruan, Menglan
Ge, Rongjun
Yang, Chunfeng
Chen, Yang
Xie, Chunming
author_facet Wang, Jianwei
Wang, Qing
Ruan, Menglan
Ge, Rongjun
Yang, Chunfeng
Chen, Yang
Xie, Chunming
contents Generating BOLD images from T1w images offers a promising solution for recovering missing BOLD information and enabling downstream tasks when BOLD images are corrupted or unavailable. Motivated by this, we propose DINO-BOLDNet, a DINOv3-guided multi-slice attention framework that integrates a frozen self-supervised DINOv3 encoder with a lightweight trainable decoder. The model uses DINOv3 to extract within-slice structural representations, and a separate slice-attention module to fuse contextual information across neighboring slices. A multi-scale generation decoder then restores fine-grained functional contrast, while a DINO-based perceptual loss encourages structural and textural consistency between predictions and ground-truth BOLD in the transformer feature space. Experiments on a clinical dataset of 248 subjects show that DINO-BOLDNet surpasses a conditional GAN baseline in both PSNR and MS-SSIM. To our knowledge, this is the first framework capable of generating mean BOLD images directly from T1w images, highlighting the potential of self-supervised transformer guidance for structural-to-functional mapping.
format Preprint
id arxiv_https___arxiv_org_abs_2512_08337
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle DINO-BOLDNet: A DINOv3-Guided Multi-Slice Attention Network for T1-to-BOLD Generation
Wang, Jianwei
Wang, Qing
Ruan, Menglan
Ge, Rongjun
Yang, Chunfeng
Chen, Yang
Xie, Chunming
Computer Vision and Pattern Recognition
Generating BOLD images from T1w images offers a promising solution for recovering missing BOLD information and enabling downstream tasks when BOLD images are corrupted or unavailable. Motivated by this, we propose DINO-BOLDNet, a DINOv3-guided multi-slice attention framework that integrates a frozen self-supervised DINOv3 encoder with a lightweight trainable decoder. The model uses DINOv3 to extract within-slice structural representations, and a separate slice-attention module to fuse contextual information across neighboring slices. A multi-scale generation decoder then restores fine-grained functional contrast, while a DINO-based perceptual loss encourages structural and textural consistency between predictions and ground-truth BOLD in the transformer feature space. Experiments on a clinical dataset of 248 subjects show that DINO-BOLDNet surpasses a conditional GAN baseline in both PSNR and MS-SSIM. To our knowledge, this is the first framework capable of generating mean BOLD images directly from T1w images, highlighting the potential of self-supervised transformer guidance for structural-to-functional mapping.
title DINO-BOLDNet: A DINOv3-Guided Multi-Slice Attention Network for T1-to-BOLD Generation
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2512.08337