{
  "abstract": "Background: Automated astrocyte quantification in human immunohistochemistry (IHC) is hindered by stain variability and dense clustering. Standard Transformers often miss faint somata due to agnostic initialization, while CNNs struggle with overlapping instances.\n\nMethods: We propose a Heatmap-Seeded Query Transformer coupling candidate sufficiency with global competitive assignment. Queries initialized from CNN-derived spatial peaks capture faint targets and are refined via a cross-scale Transformer decoder. We evaluated the model on ALDH1L1- and GFAP-stained human tissue \\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$(0.35\\mathrm{--}0.50\\,\\mu\\mathrm{m/px})$$\\end{document} against Faster R-CNN, YOLOv11, and Deformable DETR using COCO-style metric under the Olar et al. evaluation protocol, bootstrap-resampled FROC analysis, and targeted stratifications for local contrast and pair-separation recall.\n\nResults: Our method improved overall Average Precision (AP) across the vast majority of test cohorts, including a nearly 19% relative increase in high-resolution ALDH1L1 images. Crucially, stratified evaluations revealed significant improvements in localizing low-contrast targets and resolving densely crowded cell pairs without relying on heuristic filtering. FROC analysis confirmed superior sensitivity at clinically relevant low false-positive rates.\n\nConclusions: Integrating spatial priors with global reasoning effectively mitigates the sensitivity-precision trade-off. This yields a robust detector for astrocyte soma localization resilient to stain and resolution shifts, serving as a highly reliable upstream component for quantitative digital pathology workflows.\n\nSupplementary Information: The online version contains supplementary material available at https://doi.org/10.1186/s12880-026-02678-5.",
  "abstract_shown": true,
  "affiliations": [
    "Early Intervention Unit, Department of Psychiatry, Affiliated Nanjing Brain Hospital, Nanjing Medical University, Nanjing, China",
    "School of Biomedical Engineering and Informatics, Nanjing Medical University, Nanjing, Jiangsu China"
  ],
  "article_type": "research article",
  "arxiv": null,
  "author_total": 2,
  "authors": [
    {
      "affs": [
        0,
        1
      ],
      "name": "Jiayang Zhu"
    },
    {
      "affs": [
        0,
        1
      ],
      "name": "Xizhe Zhang",
      "orcid": "0000-0002-8684-4591"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/epssbc679gwn0.bib",
  "cited_by": 0,
  "doi": "10.1186/s12880-026-02678-5",
  "fulltext_url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC13757231/",
  "id": "epssbc679gwn0",
  "issns": [
    "1471-2342"
  ],
  "landing_url": "https://doi.org/10.1186/s12880-026-02678-5",
  "language": "en",
  "licence": "CC-BY-NC-ND-4.0",
  "licence_label": "CC BY-NC-ND 4.0",
  "licence_source": "article",
  "licence_url": "https://creativecommons.org/licenses/by-nc-nd/4.0/",
  "metadata_licence": null,
  "native_subjects": [
    [
      "heading",
      "Research"
    ],
    [
      "openalex",
      "1304|Cell Image Analysis Techniques"
    ]
  ],
  "openalex": "W7124427003",
  "pdf_url": null,
  "pmcid": "PMC13757231",
  "pmid": "43015688",
  "published": "2026-08-17",
  "publisher": "BMC",
  "ris": "https://research.draconitic.com/paper/epssbc679gwn0.ris",
  "sources": [
    {
      "first_seen": "2026-11-16T12:22:58Z",
      "native_id": "PMC13757231",
      "source": "pmc",
      "url": "https://doi.org/10.1186/s12880-026-02678-5"
    }
  ],
  "subjects": [
    {
      "field": "biochemistry",
      "origin": "openalex:1304|Cell Image Analysis Techniques",
      "subfield": "biophysics"
    }
  ],
  "title": "Heatmap-seeded spatially grounded query Transformers improve astrocyte soma detection across Immunostains and scan resolutions",
  "title_text": "Heatmap-seeded spatially grounded query Transformers improve astrocyte soma detection across Immunostains and scan resolutions",
  "type": "peer-reviewed",
  "updated_at": "2026-11-16T12:22:58Z",
  "url": "https://research.draconitic.com/paper/epssbc679gwn0",
  "venue": "BMC Medical Imaging",
  "versions": []
}