EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC-ND 4.0

Evaluating ChatGPT ‐5 for Detection of Barrett's Esophagus and Grading of Esophagitis: A Multiclass Endoscopic Image Analysis

Hamza R Khan iD, Ibraheem Mirza, Owais M Aftab iD, Yash Shah, Ahmed H Al-Khazraji iD

JGH Open · 2026-07 · 10 (7)

DOI: 10.1002/jgh3.70452 ↗

Kurz gefasst

Fachartikel in der Zeitschrift JGH Open, online erschienen am 22.07.2026. Verfasst von Hamza R Khan und 4 weiteren Autor:innen. Beteiligte Einrichtung: Division of Gastroenterology & Advanced Endoscopy Rutgers New Jersey Medical School Newark New Jersey USA. Die Arbeit nennt 23 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Aims

Gastroesophageal reflux disease can progress to reflux esophagitis and Barrett's esophagus (BE), making accurate endoscopic diagnosis important. Artificial intelligence tools like ChatGPT‐5 may assist image interpretation, though data on newer large language models remains limited. This study evaluated ChatGPT‐5 for BE detection and LA esophagitis severity classification. Methods and Results Endoscopic images from the HyperKvasir dataset were analyzed, including BE, esophagitis A, esophagitis B–D, and normal Z‐line images. Four standardized prompts were assessed: (1) BE versus normal, (2) esophagitis versus normal, (3) LA‐grade severity (A vs. B–D), and (4) BE versus severe esophagitis. ChatGPT‐5 was evaluated in auto mode. Two investigators analyzed 640 unique images, yielding 1280 evaluations. Sensitivity, specificity, positive/negative predictive values (PPV/NPV), F1 scores, and accuracy were calculated. In binary tasks, sensitivity was highest for severe esophagitis (B–D) (0.774). Binary accuracy was similar across tasks (~0.64), with the highest for severe esophagitis (0.655). In three‐class analyses, severe esophagitis performed best (sensitivity 0.506, specificity 0.761, accuracy 0.438), with performance improving alongside disease severity. PPV trended higher for severe esophagitis compared with normal mucosa ( p ~ 0.03), though significance was not retained after multiple‐comparison correction. In the BE‐severe esophagitis‐normal comparison, severe esophagitis achieved the highest sensitivity (0.590), specificity (0.790), and accuracy (0.521). PPV trended higher for severe esophagitis versus normal mucosa ( p ~ 0.05), though significance was not maintained after multiplicity adjustment. NPVs exceeded PPVs across all paradigms.

Conclusion

ChatGPT‐5 demonstrated moderate performance for esophageal image interpretation, performing best for severe esophagitis and worst for mild esophagitis/BE. Binary prompting outperformed multiclass formats, and the model functioned better as a rule‐out tool.

Schlagwörter

Stichwörter der Autor:innen

Barrett esophagusChatGPTartificial intelligenceendoscopyesophageal diseasesesophagitislarge language models

Autor:innen und Institutionen

  • Division of Gastroenterology & Advanced Endoscopy Rutgers New Jersey Medical School Newark New Jersey USA.
  • Department of Medicine SUNY Downstate Health Sciences Brooklyn New York USA.

ORCID-Profile: Hamza R Khan, Owais M Aftab, Ahmed H Al-Khazraji

Zitieren

APA

Khan, H. R., Mirza, I., Aftab, O. M., Shah, Y., & Al-Khazraji, A. H. (2026). Evaluating ChatGPT ‐5 for Detection of Barrett's Esophagus and Grading of Esophagitis: A Multiclass Endoscopic Image Analysis. JGH Open, 10(7). https://doi.org/10.1002/jgh3.70452

Vancouver

Khan HR, Mirza I, Aftab OM, Shah Y, Al-Khazraji AH. Evaluating ChatGPT ‐5 for Detection of Barrett's Esophagus and Grading of Esophagitis: A Multiclass Endoscopic Image Analysis. JGH Open. 2026;10(7). doi:10.1002/jgh3.70452

BibTeX
@article{Khan2026,
  author = {Khan, Hamza R and Mirza, Ibraheem and Aftab, Owais M and Shah, Yash and Al-Khazraji, Ahmed H},
  title = {Evaluating ChatGPT ‐5 for Detection of Barrett's Esophagus and Grading of Esophagitis: A Multiclass Endoscopic Image Analysis},
  journal = {JGH Open},
  year = {2026},
  volume = {10},
  number = {7},
  doi = {10.1002/jgh3.70452}
}
Kostenlos starten →