{
  "abstract": "Medical AI has produced many radiology models, particularly for chest X-rays (CXR), each excelling at isolated tasks like lesion detection or report generation. However, these models have disparate capabilities and limited generalizability due to training on restricted datasets, making clinical integration challenging. Large language models (LLMs) now enable interfacing heterogeneous models within agentic frameworks that automatically interpret and unify outputs in natural language. In this work, we present RadFabric, an agentic AI system that orchestrates fourteen specialized open-source CXR analytics models and two Vision-Language Models (VLM) through a modular protocol. RadFabric includes an Anatomical Interpretation Agent that grounds visual findings in anatomical context, and a trainable reasoning agent that synthesizes these anatomically-enriched outputs with VLM-generated radiology reports into transparent, step-by-step diagnoses, even when model outputs are heterogeneous or conflicting. This architecture enables explainable, robust diagnoses across common and rare pathologies while facilitating extensibility through additional agents. Evaluation results on the MIMIC-CXR dataset shows that RadFabric can achieve an AUC of 85.18% on task of detecting different legion types from the given CXR, outperforming all state-of-art CXR models. Notably, the reasoning agent particularly improves detection of uncommon findings, demonstrating enhanced interpretability, generalizability, and clinical applicability.",
  "abstract_shown": true,
  "affiliations": [
    "Department of Electrical Engineering, City University of Hong Kong, Hong Kong, SAR China",
    "Department of Radiology, Massachusetts General Hospital, Boston, MA USA",
    "Department of Computer Science and Engineering, The Chinese University of Hong Kong, Hong Kong, SAR China",
    "School of Computing, University of Georgia, Athens, GA USA",
    "Harvard University, Boston, MA USA",
    "Department of Computer Science and Engineering, Lehigh University, Bethlehem, PA USA",
    "Department of Electronic Engineering, The Chinese University of Hong Kong, Hong Kong, SAR China"
  ],
  "article_type": "research article",
  "arxiv": null,
  "author_total": 21,
  "authors": [
    {
      "affs": [
        0,
        1
      ],
      "name": "Wenting Chen"
    },
    {
      "affs": [
        2
      ],
      "name": "Yi Dong"
    },
    {
      "affs": [
        3
      ],
      "name": "Zhaojun Ding"
    },
    {
      "affs": [
        3
      ],
      "name": "Yucheng Shi"
    },
    {
      "affs": [
        1
      ],
      "name": "Fang Zeng"
    },
    {
      "affs": [
        3
      ],
      "name": "Yifan Zhou"
    },
    {
      "affs": [
        3
      ],
      "name": "Huaqin Zhao"
    },
    {
      "affs": [
        1
      ],
      "name": "Yijun Luo"
    },
    {
      "affs": [
        1
      ],
      "name": "Tianyu Lin"
    },
    {
      "affs": [
        2
      ],
      "name": "Yihang Su"
    },
    {
      "affs": [
        1
      ],
      "name": "Yichen Wu"
    },
    {
      "affs": [
        4
      ],
      "name": "Jasmine Liu"
    },
    {
      "affs": [
        5
      ],
      "name": "Kai Zhang"
    },
    {
      "affs": [
        2
      ],
      "name": "Wenxuan Wang"
    },
    {
      "affs": [
        3
      ],
      "name": "Zhen Xiang"
    },
    {
      "affs": [
        3
      ],
      "name": "Tianming Liu"
    },
    {
      "affs": [
        3
      ],
      "name": "Ninghao Liu"
    },
    {
      "affs": [
        1
      ],
      "name": "Quanzheng Li"
    },
    {
      "affs": [
        5
      ],
      "name": "Lichao Sun"
    },
    {
      "affs": [
        0,
        6
      ],
      "name": "Yixuan Yuan"
    },
    {
      "affs": [
        1
      ],
      "name": "Xiang Li"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/bc0m72dw8gynf.bib",
  "cited_by": 4,
  "doi": "10.1038/s41746-026-02994-8",
  "fulltext_url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC13740248/",
  "id": "bc0m72dw8gynf",
  "issns": [
    "2398-6352"
  ],
  "landing_url": "https://doi.org/10.1038/s41746-026-02994-8",
  "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",
      "Article"
    ],
    [
      "openalex",
      "2718|Artificial Intelligence in Healthcare and Education"
    ]
  ],
  "openalex": "W7168428995",
  "pdf_url": "https://www.nature.com/articles/s41746-026-02994-8_reference.pdf",
  "pmcid": "PMC13740248",
  "pmid": "42457975",
  "published": "2026-07-15",
  "publisher": "Nature Publishing Group",
  "ris": "https://research.draconitic.com/paper/bc0m72dw8gynf.ris",
  "sources": [
    {
      "first_seen": "2026-11-10T12:21:32Z",
      "native_id": "PMC13740248",
      "source": "pmc",
      "url": "https://doi.org/10.1038/s41746-026-02994-8"
    }
  ],
  "subjects": [
    {
      "field": "medicine",
      "origin": "openalex:2718|Artificial Intelligence in Healthcare and Education",
      "subfield": "epidemiology-public-health"
    }
  ],
  "title": "Interpretable agentic AI system with localized reasoning for radiology",
  "title_text": "Interpretable agentic AI system with localized reasoning for radiology",
  "type": "peer-reviewed",
  "updated_at": "2026-11-10T12:21:32Z",
  "url": "https://research.draconitic.com/paper/bc0m72dw8gynf",
  "venue": "NPJ Digital Medicine",
  "versions": []
}