Loki-OT corrects lymphocyte mimicry by transferring region‑level tissue reasoning to individual cell predictions through Unbalanced Optimal Transport. It uses density priors from a pathology MLLM to guide ambiguous cell reassignment and distills the resulting transport plan into a lightweight MLP that learns context‑aware decision boundaries within pretrained cell‑foundation features. On the TCGA‑BRCA cohort, Loki‑OT outperformed a fully supervised PanopTILs classifier in patient‑level MAE and improved F1 scores in epithelium‑rich mimicry tissues.
arXiv:2609.00704v1 Announce Type: new Abstract: Functional tissue units (FTUs), including tertiary lymphoid structures (TLSs), blood vessels, and glands, encode localized immune, vascular, and epithe...
arXiv:2601.17228v2 Announce Type: replace Abstract: Deep learning models in computational pathology often fail to generalize across cohorts and institutions due to domain shift. Existing approaches e...
AdaptiveCDM is a modular framework for source‑free few‑shot domain adaptation in cell detection, enabling a pretrained model to adapt to new imaging domains using only a handful of labeled target images and no source data. It combines Resolution‑Aware Augmentation (RAug) to balance scarce, class‑imbalanced samples while preserving cellular morphology, and Category‑Aware Representation Learning (CARL) to strengthen class‑consistent proposals for better localization and classification. Experiments on M5 and Raabin‑WBC datasets show that AdaptiveCDM achieves competitive or superior mAP scores compared to state‑of‑the‑art methods under their respective supervision settings.
arXiv:2609.36429v1 Announce Type: new Abstract: Predicting gene expression from H&E-stained histology images offers a scalable alternative to costly spatial transcriptomics, yet most existing methods...
arXiv:2607. 25164v1 Announce Type: cross Abstract: A CT examination captures multiple organs, but many biomedical questions concern abnormalities, prognosis, or longitudinal change in a specific organ.
arXiv:2608. 10657v1 Announce Type: cross Abstract: Leukemia cell image classification is challenged by real-world domain shifts from acquisition, staining, illumination, and site protocols, causing single-dataset models to generalize poorly in real clinical scenarios.
The paper introduces Conserved Immune Topology (CIT), a lightweight spatial representation that enhances cross‑cancer MSI‑H prediction by augmenting pathology foundation‑model embeddings with immune‑related descriptors. CIT identifies immune‑associated tiles via unsupervised clustering and encodes features such as tertiary lymphoid structures, peritumoral immune reactions, tumor‑infiltrating lymphocyte density, and immune‑tumor mixing, all without requiring annotations or target‑domain data. In cross‑site and cross‑cancer experiments on CPTAC‑COAD and TCGA‑STAD cohorts, CIT improved zero‑shot TransMIL AUC from 0.6627 to 0.7161, demonstrating that spatial immune topology can provide an organ‑invariant representation for MSI‑H prediction.
arXiv:2505.03380v2 Announce Type: replace Abstract: Accurate delineation of tumors and surrounding organs-at-risk is essential for radiotherapy, surgery and treatment response assessment, yet remains...
arXiv:2605. 18419v2 Announce Type: replace-cross Abstract: Vision-language models (VLMs) can couple visual perception with open-ended clinical reasoning, making them attractive for computational histopathology.
arXiv:2606. 28392v1 Announce Type: cross Abstract: Accurate lesion segmentation in PET/CT is critical for oncology, yet remains challenging because physiologic tracer uptake and artifacts can mimic malignant signal.
arXiv:2601. 08127v2 Announce Type: replace-cross Abstract: Expert-annotated training data remains the critical bottleneck for AI in histopathology, particularly for rare pathologies where even dozens of cases may be unavailable.
arXiv:2606. 06983v1 Announce Type: cross Abstract: Computational pathology requires visual representations that transfer across diverse clinical endpoints and remain robust to variation in magnification, staining, scanner type, slide preparation, and input resolution.