dustindiazlopez / vqa-med-2019

Visual Question Answering in the Medical Domain VQA-Med 2019

VQA-Med-2019

• Website: https://www.imageclef.org/2019/medical/vqa/
• Mailing list: https://groups.google.com/d/forum/imageclef-vqa-med
• Paper: http://ceur-ws.org/Vol-2380/paper_272.pdf
• Results of the VQA-Med-2019 challenge on crowdAI: https://www.crowdai.org/challenges/imageclef-2019-vqa-med/leaderboards

Task:

VQA-Med 2019 focused on radiology images and four main categories of questions: Modality, Plane, Organ system and Abnormality. These categories are designed with different degrees of difficulty leveraging both classification and text generation approaches. 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 additional medical knowledge or domain-specific inference.

VQA-Med-2019 Data:

The VQA-Med-2019 dataset includes 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.

1. The training set: https://github.com/abachaa/VQA-Med-2019/blob/master/ImageClef-2019-VQA-Med-Training.zip

2. The validation set: https://github.com/abachaa/VQA-Med-2019/blob/master/ImageClef-2019-VQA-Med-Validation.zip

3. The VQA-Med-2019 test set and the reference answers are available here: https://github.com/abachaa/VQA-Med-2019/tree/master/VQAMed2019Test

Please see the readme files for more detailed information about the dataset and the categories of questions and answers: https://github.com/abachaa/VQA-Med-2019/blob/master/README-VQA-Med-2019-Data.txt https://github.com/abachaa/VQA-Med-2019/blob/master/VQAMed2019Test/README-VQA-Med-2019-TestSet.txt

Reference:

If you use the VQA-Med 2019 dataset, 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}
    }

Contact Information

Asma Ben Abacha: [email protected] https://sites.google.com/site/asmabenabacha/