Generating dermatopathology reports from gigapixel whole slide images with HistoGPT

Manuel Tran, Paul Schmidle, Ruifeng Ray Guo, Sophia J. Wagner, Valentin Koch, Valerio Lupperger, Brenna Novotny, Dennis H. Murphree, Heather D. Hardway, Marina D’Amato, Judith Lefkes, Daan J. Geijs, Annette Feuchtinger, Alexander Böhner, Robert Kaczmarczyk, Tilo Biedermann, Avital L. Amir, Antien L. Mooyaart, Francesco Ciompi, Geert Litjens, Chen Wang, Nneka I. Comfere, Kilian Eyerich (), Stephan A. Braun (), Carsten Marr () and Tingying Peng ()
Additional contact information
Manuel Tran: Helmholtz Munich
Paul Schmidle: University of Freiburg
Ruifeng Ray Guo: Mayo Clinic
Sophia J. Wagner: Helmholtz Munich
Valentin Koch: Technical University of Munich
Valerio Lupperger: MLL Munich Leukemia Laboratory
Brenna Novotny: Mayo Clinic
Dennis H. Murphree: Mayo Clinic
Heather D. Hardway: Mayo Clinic
Marina D’Amato: Radboud University Medical Center
Judith Lefkes: Radboud University Medical Center
Daan J. Geijs: Radboud University Medical Center
Annette Feuchtinger: Helmholtz Munich
Alexander Böhner: Technical University of Munich
Robert Kaczmarczyk: Technical University of Munich
Tilo Biedermann: Technical University of Munich
Avital L. Amir: Radboud University Medical Center
Antien L. Mooyaart: Erasmus University Medical Center
Francesco Ciompi: Radboud University Medical Center
Geert Litjens: Radboud University Medical Center
Chen Wang: Mayo Clinic
Nneka I. Comfere: Mayo Clinic
Kilian Eyerich: University of Freiburg
Stephan A. Braun: University Hospital Münster
Carsten Marr: Helmholtz Munich
Tingying Peng: Helmholtz Munich

Nature Communications, 2025, vol. 16, issue 1, 1-17

Abstract: Abstract Histopathology is the reference standard for diagnosing the presence and nature of many diseases, including cancer. However, analyzing tissue samples under a microscope and summarizing the findings in a comprehensive pathology report is time-consuming, labor-intensive, and non-standardized. To address this problem, we present HistoGPT, a vision language model that generates pathology reports from a patient’s multiple full-resolution histology images. It is trained on 15,129 whole slide images from 6705 dermatology patients with corresponding pathology reports. The generated reports match the quality of human-written reports for common and homogeneous malignancies, as confirmed by natural language processing metrics and domain expert analysis. We evaluate HistoGPT in an international, multi-center clinical study and show that it can accurately predict tumor subtypes, tumor thickness, and tumor margins in a zero-shot fashion. Our model demonstrates the potential of artificial intelligence to assist pathologists in evaluating, reporting, and understanding routine dermatopathology cases.

Date: 2025
References: Add references at CitEc
Citations:

Downloads: (external link)
https://www.nature.com/articles/s41467-025-60014-x Abstract (text/html)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:nat:natcom:v:16:y:2025:i:1:d:10.1038_s41467-025-60014-x

Ordering information: This journal article can be ordered from
https://www.nature.com/ncomms/

DOI: 10.1038/s41467-025-60014-x

Access Statistics for this article

Nature Communications is currently edited by Nathalie Le Bot, Enda Bergin and Fiona Gillespie

More articles in Nature Communications from Nature
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().