arXiv:2608. 10131v1 Announce Type: cross Abstract: Vision foundation models are increasingly used as reusable encoders in medical image computing, yet their high-dimensional spatial embeddings are difficult to inspect beyond downstream task performance or global dimensionality reduction.
arXiv:2608. 05960v1 Announce Type: cross Abstract: Routine CT interpretation is inherently comprehensive, capturing incidental findings across the entire scan volume.
arXiv:2608. 12689v1 Announce Type: cross Abstract: Multi-parametric magnetic resonance imaging (mpMRI) is a cornerstone for brain tumor diagnosis and treatment, yet current AI models face critical limitations: their lack of natural language interaction and interpretability impedes spatial information integration and cross-modal reasoning required clinically.
EXPOSE is a framework that applies Sparse Autoencoders to Vision Foundation Model embeddings in computational pathology, aiming to separate biological signals from domain‑specific noise. By training a sparse representation of VFM features and using a linear classifier to flag domain‑specific latent dimensions, the method masks these components before downstream relapse prediction, avoiding the need to retrain the backbone model. Experiments on a large prostate cancer dataset demonstrate that removing domain‑specific features improves cross‑domain performance and raises the Domain Robustness Index (DoRI).
arXiv:2608. 14924v1 Announce Type: cross Abstract: Spatial transcriptomics (ST) links tissue morphology with molecular programs, motivating multimodal pretraining methods that align histology images with gene expression.
The paper introduces BCARP, a Biology‑Guided Context‑Aware Residual Propagation framework that predicts spatial gene expression from H&E‑stained histology images. Rather than directly decoding expression from visual features, BCARP treats histology embeddings as morphological priors and iteratively refines spot representations in a transcriptomics‑oriented latent space, integrating nuclei‑guided local context and prototype‑guided global propagation. Evaluations on three spatial transcriptomics datasets and ten HEST‑1k cohorts show that BCARP outperforms existing baselines in mean correlation‑based metrics, demonstrating the effectiveness of biology‑guided local‑global residual propagation for histology‑to‑transcriptomics prediction.
arXiv:2601.09879v2 Announce Type: replace-cross Abstract: Recent progress in medical vision-language models (VLMs) has achieved strong performance on image-level text-centric tasks such as report gen...
Integrating 3D medical images with vision-language models (VLMs) holds substantial promise for computer-aided diagnosis. However, volumetric images generate prohibitively long visual-token sequences with considerable spatial and inter-slice redundancy.
arXiv:2609.31204v1 Announce Type: cross Abstract: Recent fMRI foundation models differ substantially in the spatial scale at which they represent brain activity. ROI- and connectivity-based models ar...
The paper introduces MSM‑Seg, a dual‑memory segmentation framework for 3D multi‑modal brain tumor segmentation. It combines a modality‑and‑slice memory attention module to capture cross‑modal and spatial‑slice dependencies, a multi‑scale category‑agnostic prompt encoder for whole‑tumor guidance, and a modality‑adaptive fusion decoder to integrate complementary decoding information. Experiments on various MRI datasets show that MSM‑Seg surpasses state‑of‑the‑art methods for metastases and glioma tumor segmentation.
arXiv:2606. 29600v1 Announce Type: cross Abstract: A faithful 3D world representation should account for layered geometry, where a single camera ray may contain multiple visible and geometrically valid surfaces.
arXiv:2607. 20993v1 Announce Type: cross Abstract: Large vision-language models are becoming increasingly dominant in 3D medical image interpretation, but we rarely know which internal units encode clinical findings or where that information lives in the representation.
arXiv:2607. 16325v1 Announce Type: cross Abstract: Foundation models provide powerful representations for brain MRI analysis, but their predictions remain difficult to interpret in anatomically meaningful terms.