Article Text

Download PDFPDF
GPT-4 and histopathological image detection and classification of colorectal adenomas
  1. Hinpetch Daungsupawong1,
  2. Viroj Wiwanitkit2
  1. 1 Private Academic Consultant, Vientiane, Lao People's Democratic Republic
  2. 2 DY Patil Vidhyapeeth, Pune, India
  1. Correspondence to Dr Hinpetch Daungsupawong, Private Academic Consultant, 10000 Vientiane, Lao People's Democratic Republic; hinpetchdaung{at}

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

We would like to respond to a comment on the published article entitled ‘Accuracy of GPT-4 in histopathological image detection and classification of colorectal adenomas’.1 In order to assess their competence in histopathological diagnosis, 100 colorectal polyp photomicrographs were analysed using the traditional and custom versions of the GPT-4. Pathologists evaluated the GPT-4 responses for descriptive accuracy and compared them with a reference standard.

According to the findings, GPT-4’s median adenoma detection sensitivity was 74%, while its median specificity was 36%. The percentage of polyps that correctly were classified as non-specific alterations was 16%, while the percentage …

View Full Text


  • Handling editor Runjan Chetty.

  • Contributors HD (50% ideas)—writing, analysing and approval. VW (50% ideas)—supervision and approval.

  • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

  • Competing interests None declared.

  • Provenance and peer review Not commissioned; internally peer reviewed.