Login / Signup

Preliminary assessment of automated radiology report generation with generative pre-trained transformers: comparing results to radiologist-generated reports.

Takeshi NakauraNaofumi YoshidaNaoki KobayashiKaori ShiraishiYasunori NagayamaHiroyuki UetaniMasafumi KidohMasamichi HokamuraYoshinori FunamaToshinori Hirai
Published in: Japanese journal of radiology (2023)
Our preliminary study suggests that GPT-3.5 and GPT-4 have the possibility to generate radiology reports with high readability and reasonable image findings from very short keywords; however, concerns persist regarding the accuracy of impressions and differential diagnoses, thereby requiring verification by radiologists.
Keyphrases
  • artificial intelligence
  • deep learning
  • machine learning
  • adverse drug
  • high throughput
  • healthcare
  • body composition