[783ff8]: / README.md

Download this file

12 lines (5 with data), 493 Bytes

VQA-RAD (Visual Question Answering in Radiology)

VQA-RAD consists of 3,515 question–answer pairs on 315 radiology images.

Source: A dataset of clinically generated visual questions and answers about radiology images

Homepage: https://osf.io/89kps/