Dino-NestedUNet: Unlocking Foundation Vision Encoders for Pathology Tumor Bulk Segmentation via Dense Decoding

• URL: http://arxiv.org/abs/2605.00894v1
• Date: Mon, 27 Apr 2026 23:22:14 GMT
• Title: Dino-NestedUNet: Unlocking Foundation Vision Encoders for Pathology Tumor Bulk Segmentation via Dense Decoding
• Abstract summary: This paper presents Dino-NestedUNet, a framework that couples a pre-trained DINOv3 encoder with a Nested Dense Decoder.<n>We observe consistent improvements over UNet++ and standard Dino-UNet variants, particularly under cross-domain shift.<n>These results suggest that dense decoding is a key ingredient for unlocking foundation encoders in boundary-sensitive pathology segmentation.
• Score: 7.637787421453472
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/
• Abstract: Vision foundation models (VFMs), such as DINOv3, provide rich semantic representations that are promising for computational pathology. However, many current adaptations pair frozen VFMs with lightweight decoders, creating a capacity mismatch that often limits boundary fidelity for infiltrative tumor bulk segmentation. This paper presents Dino-NestedUNet, a framework that couples a pre-trained DINOv3 encoder with a Nested Dense Decoder. Instead of sparse skip connections and linear upsampling, the proposed decoder forms a dense grid of intermediate pathways to enable continuous feature reuse and multi-scale recalibration, aligning high-level semantics with low-level morphological textures during reconstruction. We evaluate Dino-NestedUNet on three histopathology cohorts (multi-center CHTN, institutional OSU, and CAMELYON16) and observe consistent improvements over UNet++ and standard Dino-UNet variants, particularly under cross-domain shift. To further assess external generalization, we perform zero-shot evaluation by training on CHTN and directly testing on unseen TIGER WSIBULK and OSU CRC cohorts without fine-tuning. These results suggest that dense decoding is a key ingredient for unlocking foundation encoders in boundary-sensitive pathology segmentation.

Related papers

• Bridging Vision Foundation Model Priors with CLIP for Spatial-aware Few-shot Anomaly Detection in Medical Images [48.646448077284276]
  We propose Spatial-FAD, a framework for effective few-shot medical anomaly detection.<n>It improves lesion localization by combining VFM spatial priors with CLIP semantics.<n>Our method achieves an average improvement of over 11.4% in Dice score and 1.8% in AUC.
  arXiv  Detail & Related papers  (2026-09-11T05:21:01Z)
• A deep dictionary network-based foundation model for ultra-low-dose CT denoising [49.70293041079032]
  Ultra-low-dose computed tomography (ULDCT) reduces radiation exposure but suffers from severe noise that degrades diagnostic image quality.<n>Existing deep learning-based denoising methods are typically trained in an organ-specific fashion, resulting in limited generalization across heterogeneous multi?organ imaging scenarios.<n>We propose an architecture?able foundation model based on the deep dictionary network (DDN) for unified multi-organ ULDCT denoising.
  arXiv  Detail & Related papers  (2026-09-11T05:14:08Z)
• Rad-JEPA 3D: Radiology Joint-Embedding Predictive Model for 3D Computed Tomography [10.665332135327205]
  We introduce Rad-JEPA 3D, a joint-embedding predictive framework that learns volumetric CT representations.<n>To improve the quality of intermediate representations, we propose Hidden States Orthogonal Regularization (HSOR)<n>HSOR aligns student-teacher hidden states and reduces feature redundancy throughout the encoder.
  arXiv  Detail & Related papers  (2026-07-28T19:00:17Z)
• AMN: An Adaptive Multi-Scale Fusion Network with Boundary and Uncertainty Modeling for Nuclei Segmentation [0.0]
  We present AMN, a dual-encoder segmentation framework that jointly leverages a Swin Transformer and a ResNet-50 feature pyramid.<n>AMN is trained with a multi-objective loss combining class-weighted focal loss, boundary-aware loss with positive-pixel emphasis, and a novel uncertainty-modulated classification term that suppresses overconfident erroneous predictions.
  arXiv  Detail & Related papers  (2026-05-31T06:46:23Z)
• GRAFNet: Multiscale Retinal Processing via Guided Cortical Attention Feedback for Enhancing Medical Image Polyp Segmentation [6.834321209531585]
  We propose GRAFNet, a biologically inspired architecture that emulates the hierarchical organisation of the human visual system.<n>GRAFNet integrates three key modules: (1) a Guided Asymmetric Attention Module (GAAM) that mimics orientation-tuned cortical neurones to polyp boundaries, (2) a MultiScale Retinal Module (MSRM) that replicates retinal ganglion cell pathways for parallel multi-feature analysis, and (3) a Guided Cortical Attention Feedback Module (GCAFM) that applies predictive coding for iterative refinement.
  arXiv  Detail & Related papers  (2026-02-15T17:29:37Z)
• DINO-SAE: DINO Spherical Autoencoder for High-Fidelity Image Reconstruction and Generation [47.409626500688866]
  We present the DINO Spherical Autoencoder (DINO-SAE), a framework that bridges semantic representation and pixel-level reconstruction.<n>Our approach achieves state-of-the-art reconstruction quality, reaching 0.37 rFID and 26.2 dB PSNR, while maintaining strong semantic alignment to the pretrained VFM.
  arXiv  Detail & Related papers  (2026-01-30T12:25:34Z)
• Decoder-Free Supervoxel GNN for Accurate Brain-Tumor Localization in Multi-Modal MRI [1.209913077217557]
  SVGFormer is a decoder-free pipeline that partitions the volume into a semantic graph of supervoxels.<n>Its hierarchical encoder learns rich node representations by combining a patch-level Transformer with a supervoxel-level Graph Attention Network.<n>Our results establish that a graph-based, encoder-only paradigm offers an accurate and inherently interpretable alternative for 3D medical image representation.
  arXiv  Detail & Related papers  (2026-01-20T15:13:04Z)
• Meta-cognitive Multi-scale Hierarchical Reasoning for Motor Imagery Decoding [43.32839547082765]
  This work investigates a hierarchical and meta-cognitive decoding framework for four-class electroencephalogram (EEG) signals.<n>We introduce a multi-scale hierarchical signal processing module that reorganizes backbone features into temporal multi-scale representations.<n>We instantiate this framework on three standard EEG backbones and evaluate four-class MI decoding using the BCI Competition IV-2a dataset.
  arXiv  Detail & Related papers  (2025-11-11T06:32:23Z)
• Unleashing Vision Foundation Models for Coronary Artery Segmentation: Parallel ViT-CNN Encoding and Variational Fusion [12.839049648094893]
  coronary artery segmentation is critical for computeraided diagnosis of coronary artery disease (CAD)<n>We propose a novel framework that leverages the power of vision foundation models (VFMs) through a parallel encoding architecture.<n>The proposed framework significantly outperforms state-of-the-art methods, achieving superior performance in accurate coronary artery segmentation.
  arXiv  Detail & Related papers  (2025-07-17T09:25:00Z)
• Topology-Constrained Learning for Efficient Laparoscopic Liver Landmark Detection [46.2391319253146]
  Liver landmarks provide crucial anatomical guidance to the surgeon during laparoscopic liver surgery.<n>TopoNet is a novel topology-constrained learning framework for laparoscopic liver landmark detection.<n>Our framework adopts a snake-CNN dual-path encoder to simultaneously capture detailed RGB texture information and depth-informed topological structures.
  arXiv  Detail & Related papers  (2025-07-01T07:35:36Z)
• TransNuSeg: A Lightweight Multi-Task Transformer for Nuclei Segmentation [6.369485141013728]
  We make the first attempt at a pure Transformer framework for nuclei segmentation, called TransNuSeg. To eliminate the divergent predictions from different branches in previous work, a novel self distillation loss is introduced to explicitly impose consistency regulation between branches. Experiments on two datasets, including MoNuSeg, have shown that our methods can outperform state-of-the-art counterparts by 2-3% Dice with 30% fewer parameters.
  arXiv  Detail & Related papers  (2023-07-16T14:12:54Z)
• From NeurODEs to AutoencODEs: a mean-field control framework for width-varying Neural Networks [68.8204255655161]
  We propose a new type of continuous-time control system, called AutoencODE, based on a controlled field that drives dynamics. We show that many architectures can be recovered in regions where the loss function is locally convex.
  arXiv  Detail & Related papers  (2023-07-05T13:26:17Z)
• Omni-supervised Point Cloud Segmentation via Gradual Receptive Field Component Reasoning [41.83979510282989]
  We bring the first omni-scale supervision method to point cloud segmentation via the proposed gradual Receptive Field Component Reasoning (RFCR) Our method brings new state-of-the-art performances for S3DIS as well as Semantic3D and ranks the 1st in the ScanNet benchmark among all the point-based methods.
  arXiv  Detail & Related papers  (2021-05-21T08:32:02Z)
• G-MIND: An End-to-End Multimodal Imaging-Genetics Framework for Biomarker Identification and Disease Classification [49.53651166356737]
  We propose a novel deep neural network architecture to integrate imaging and genetics data, as guided by diagnosis, that provides interpretable biomarkers. We have evaluated our model on a population study of schizophrenia that includes two functional MRI (fMRI) paradigms and Single Nucleotide Polymorphism (SNP) data.
  arXiv  Detail & Related papers  (2021-01-27T19:28:04Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.