Validation of a radiolological report generation model for chest radiographs
Principal Investigator
Name
Hyungjin Kim
Degrees
M.D., Ph.D.
Institution
Seoul National University Hospital
Position Title
Clinical associate professor
Email
About this CDAS Project
Study
PLCO
(Learn more about this study)
Project ID
PLCOI-1404
Initial CDAS Request Approval
Dec 4, 2023
Title
Validation of a radiolological report generation model for chest radiographs
Summary
One of the most innovative developments in artificial intelligence is the large language model (LLM). Vision-language models, a subset of LLMs, have the capability to perform tasks such as image captioning and interpretation. In the realm of medical imaging, particularly for chest radiographs — the most commonly utilized imaging technique — this technology holds significant potential. This project is focused on the external validation of a vision-language model specifically designed for generating reports of chest radiographs. The PLCO dataset will be employed as one of the external validation sets. Reports generated by the model will be evaluated for both detection accuracy and acceptability.
Aims
This project aims to externally validate a specialized vision-language model for creating reports on chest radiographs, using the PLCO dataset, and will assess the model's accuracy and acceptability in report generation.
Collaborators
Eui Jin Hwang, Seoul National University Hospital
Amy Hong, Kakao Brain
Jong Hyuk Lee, Seoul National University Hospital
Ju Gang Nam, Seoul National University Hospital