{
  "abstract": "Background/Objectives: Multimodal large language models are emerging tools for medical image interpretation, but their performance on isolated diffusion-weighted imaging (DWI) for acute ischemic stroke remains largely unexplored. We evaluated GPT-5 for stroke detection and vascular territory classification on preselected representative DWI images using a structured two-step prompting protocol. Methods: This single-center, retrospective, balanced case–control study included 220 cases (110 stroke patients and 110 stroke-negative controls). Four representative axial DWI slices per case were provided without apparent diffusion coefficient maps. Step 1 queried stroke presence and vascular territory; Step 2 queried the middle cerebral artery (MCA) segment-associated territory (M1–M4) following an MCA response. The reference standard was radiologist–neurologist consensus based on full multimodal magnetic resonance imaging, blinded to GPT-5 outputs. Diagnostic performance metrics and Cohen’s kappa (κ) were calculated. Results: Stroke detection showed a sensitivity of 100% (95% confidence interval [CI], 96.6–100%), specificity of 83.6% (95% CI, 75.6–89.4%), accuracy of 91.8%, and κ = 0.836. Accuracy was 93.6% for anterior/posterior circulation (κ = 0.838) and 90.9% for vascular territory (κ = 0.799). Conditional MCA segment-associated territory accuracy was 71.6% (53/74; κ = 0.602); end-to-end accuracy was 66.3% (53/80). Exploratory analysis of the 18 false-positive cases identified four overlapping error categories: artifactual and physiologic signal misinterpretation (72.2%), chronic lesion misclassification (27.8%), terminological inconsistency or unsupported report statements (77.8%), and etiological scenarios or unhedged definitive diagnoses (38.9%). Conclusions: GPT-5 achieved high sensitivity and a specificity of 83.6% for stroke classification on preselected representative DWI images using a vascular territory-based diagnostic framework. However, the 16.4% false-positive rate indicates that positive findings require radiologist confirmation. These findings support further evaluation of GPT-5 as a potential clinician-assisted image-interpretation tool but do not establish a clinical decision-support role.",
  "abstract_shown": true,
  "affiliations": [
    "Department of Emergency Medicine, Faculty of Medicine, Yozgat Bozok University, 66100 Yozgat, Turkey",
    "Department of Radiology, Faculty of Medicine, Yozgat Bozok University, 66100 Yozgat, Turkey",
    "Department of Neurology, Faculty of Medicine, Yozgat Bozok University, 66100 Yozgat, Turkey",
    "Department of Emergency Medicine, Dr. Nevruz Erez State Hospital, 76000 Iğdır, Turkey"
  ],
  "article_type": "research article",
  "arxiv": null,
  "author_total": 6,
  "authors": [
    {
      "affs": [
        0
      ],
      "name": "Emre Gökçen",
      "orcid": "0000-0002-6018-6105"
    },
    {
      "affs": [
        0
      ],
      "name": "Mikail Kuşdoğan",
      "orcid": "0000-0003-4075-8601"
    },
    {
      "affs": [
        1
      ],
      "name": "Mustafa Fatih Erkoc",
      "orcid": "0000-0002-6266-5177"
    },
    {
      "affs": [
        0
      ],
      "name": "Levent Albayrak",
      "orcid": "0000-0002-4288-8170"
    },
    {
      "affs": [
        2
      ],
      "name": "Güneş Seda Albayrak"
    },
    {
      "affs": [
        3
      ],
      "name": "Deniz Tanrıverdi"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/0hvvr3xjate0r.bib",
  "cited_by": 0,
  "doi": "10.3390/jcm15197716",
  "fulltext_url": "https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13666144/?tool=EBI",
  "id": "0hvvr3xjate0r",
  "issns": [
    "2077-0383"
  ],
  "landing_url": "https://europepmc.org/article/PMC/PMC13666144",
  "language": "eng",
  "licence": "CC-BY",
  "licence_label": "CC BY",
  "licence_source": "article",
  "licence_url": "https://creativecommons.org/licenses/by/4.0/",
  "metadata_licence": null,
  "native_subjects": [
    [
      "openalex",
      "2718|Artificial Intelligence in Healthcare and Education"
    ]
  ],
  "openalex": "W7220834083",
  "pdf_url": "https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13666144/pdf/?tool=EBI",
  "pmcid": "PMC13666144",
  "pmid": null,
  "published": "2026-10-01",
  "publisher": "MDPI AG",
  "ris": "https://research.draconitic.com/paper/0hvvr3xjate0r.ris",
  "sources": [
    {
      "first_seen": "2026-10-30T12:17:31Z",
      "native_id": "PMC:PMC13666144",
      "source": "europepmc",
      "url": "https://europepmc.org/article/PMC/PMC13666144"
    },
    {
      "first_seen": "2026-10-30T12:20:00Z",
      "native_id": "10.3390/jcm15197716",
      "source": "crossref",
      "url": "https://www.mdpi.com/2077-0383/15/19/7716"
    }
  ],
  "subjects": [
    {
      "field": "medicine",
      "origin": "openalex:2718|Artificial Intelligence in Healthcare and Education",
      "subfield": "epidemiology-public-health"
    }
  ],
  "title": "Diagnostic Accuracy of GPT-5 in Detecting and Localizing Acute Ischemic Stroke on Isolated Diffusion-Weighted Imaging: A Retrospective Case–Control Study",
  "title_text": "Diagnostic Accuracy of GPT-5 in Detecting and Localizing Acute Ischemic Stroke on Isolated Diffusion-Weighted Imaging: A Retrospective Case–Control Study",
  "type": "peer-reviewed",
  "updated_at": "2026-10-30T12:20:00Z",
  "url": "https://research.draconitic.com/paper/0hvvr3xjate0r",
  "venue": "Journal of clinical medicine",
  "versions": []
}