Website: https://www.imageclef.org/2019/medical/vqa/
Data and Results on crowdAI: https://www.crowdai.org/challenges/imageclef-2019-vqa-med
Mailing list: https://groups.google.com/d/forum/imageclef-vqa-med
VQA-Med 2019 focused on radiology images and four main categories of questions: Modality, Plane, Organ system and Abnormality. In this second edition of the VQA challenge, we targeted medical questions asking about one element only (e.g. what is the organ principally shown in this mri? in what plane is this mammograph taken? is this a t1 weighted, t2 weighted, or flair image? what is most alarming about this ultrasound?), and that can be answered from the image content without requiring medical knowledge or domain-specific inference. Thus, we did not target questions such as: "Is this modality safe for pregnant women?", "What is located immediately inferior to the right hemidiaphragm?", "What can be typically visualized in this plane?", "How would you measure the length of the kidneys?". These kinds of questions will be tackled in future editions of the challenge.
The datasets include a training set of 3,200 medical images with 12,792 Question-Answer (QA) pairs, a validation set of 500 medical images with 2,000 QA pairs, and a test set of 500 medical images with 500 questions. The training and validation sets are available here: https://www.crowdai.org/clef_tasks/13/task_dataset_files?challenge_id=53
Please see the readme files for more detailed information about the datasets and the categories of questions and answers.
The VQA-Med 2019 Test Set and the reference answers are available here: https://github.com/abachaa/VQA-Med-2019/tree/master/VQAMed2019Test
If you use the VQA-Med 2019 datasets, please cite our paper: "VQA-Med: Overview of the Medical Visual Question Answering Task at ImageCLEF 2019". Asma Ben Abacha, Sadid A. Hasan, Vivek V. Datla, Joey Liu, Dina Demner-Fushman, Henning Müller. CLEF 2019 Working Notes.
@Inproceedings{ImageCLEFVQA-Med2019,
author = {Asma {Ben Abacha} and Sadid A. Hasan and Vivek V. Datla and Joey Liu and Dina Demner-Fushman and Henning M\"uller},
title = {VQA-Med: Overview of the Medical Visual Question Answering Task at ImageCLEF 2019},
booktitle = {CLEF 2019 Working Notes},
series = {{CEUR} Workshop Proceedings},
year = {2019},
publisher = {CEUR-WS.org $<$http://ceur-ws.org$>$},
month = {September 9-12},
address = {Lugano, Switzerland}
}
Asma Ben Abacha: asma.benabacha@nih.gov https://sites.google.com/site/asmabenabacha/