{
  "abstract": "Multimodal clinical data contain identifiers across diverse formats. We developed the Multimodal Anonymizer, a locally deployable framework combining multimodal language models, specialist networks, deterministic transformations and iterative verification. We compared 16 orchestrator configurations using automated evaluation on 250 constructed MIMIC-IV-derived patient bundles with synthetic identifiers. A selected local configuration underwent manual evaluation on these same bundles supplemented with head CT scans, face images, handwritten notes, audio recordings and synthetic German clinical notes from unrelated sources. Separate evaluations used the German GRASCCO corpus and 250 Charite partograms. On this expanded benchmark, patient-level deidentification sensitivity was 98.80% (95% CI, 97.20-100), per-identifier sensitivity 99.82% (95% CI, 99.76-99.88), patient-level critical clinical preservation 99.60% (95% CI, 98.80-100) and the clinical preservation score 99.61% (95% CI, 99.51-99.71). All partograms met the study-defined deidentification criterion. These findings support feasibility, while constructed bundles, synthetic identifiers, limited natural-data validation and absent component ablations constrain generalizability.",
  "abstract_shown": true,
  "affiliations": [
    "Charite Universitaetsmedizin Berlin, Institute for Artificial Intelligence in Medicine, Invalidenstrasse 120, 10115 Berlin, Germany"
  ],
  "article_type": "PUBLISHAHEADOFPRINT",
  "arxiv": null,
  "author_total": 12,
  "authors": [
    {
      "name": "A. Hirsch"
    },
    {
      "name": "F. W. Ten"
    },
    {
      "name": "K. S. Krueger"
    },
    {
      "name": "R. Geyer"
    },
    {
      "name": "T. Roeschl"
    },
    {
      "name": "M. Groeschel"
    },
    {
      "name": "P. Rostin"
    },
    {
      "name": "R. Eils"
    },
    {
      "name": "M. Spott"
    },
    {
      "name": "F. Prasser"
    },
    {
      "name": "A. Meyer"
    },
    {
      "affs": [
        0
      ],
      "name": "J. Madrid"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/84tfy09gbf3cy.bib",
  "cited_by": 1,
  "doi": "10.64898/2026.05.28.26353952",
  "fulltext_url": "https://www.medrxiv.org/content/10.64898/2026.05.28.26353952v2.full",
  "id": "84tfy09gbf3cy",
  "issns": [],
  "landing_url": "https://www.medrxiv.org/content/10.64898/2026.05.28.26353952v2",
  "language": null,
  "licence": "CC-BY-NC-ND",
  "licence_label": "CC BY-NC-ND",
  "licence_source": "article",
  "licence_url": "https://creativecommons.org/licenses/by-nc-nd/4.0/",
  "metadata_licence": null,
  "native_subjects": [
    [
      "medrxiv",
      "health informatics"
    ],
    [
      "openalex",
      "1702|Privacy-Preserving Technologies in Data"
    ]
  ],
  "openalex": "W7163660880",
  "pdf_url": "https://www.medrxiv.org/content/10.64898/2026.05.28.26353952v2.full.pdf",
  "pmcid": null,
  "pmid": null,
  "published": "2026-05-28",
  "publisher": null,
  "ris": "https://research.draconitic.com/paper/84tfy09gbf3cy.ris",
  "sources": [
    {
      "first_seen": "2026-10-30T12:18:31Z",
      "native_id": "10.64898/2026.05.28.26353952",
      "source": "medrxiv",
      "url": "https://www.medrxiv.org/content/10.64898/2026.05.28.26353952v3"
    }
  ],
  "subjects": [
    {
      "field": "medicine",
      "origin": "medrxiv:health informatics",
      "subfield": "epidemiology-public-health"
    }
  ],
  "title": "The Multimodal Anonymizer: a fully local multi-agent AI system for medical data deidentification",
  "title_text": "The Multimodal Anonymizer: a fully local multi-agent AI system for medical data deidentification",
  "type": "preprint",
  "updated_at": "2026-10-30T12:19:37Z",
  "url": "https://research.draconitic.com/paper/84tfy09gbf3cy",
  "venue": "medRxiv",
  "versions": []
}