Task 2: Lesion Attribute Detection: Training

Training data for Task 2: Lesion Attribute Detection may be downloaded from the challenge submission site.

The training data consists of 2594 images and 12970 corresponding ground truth response masks (5 for each image).

Please cite the use of this data as:

Our data was extracted from the “ISIC 2018: Skin Lesion Analysis Towards Melanoma Detection” grand challenge datasets [1][2].


[1] Noel C. F. Codella, David Gutman, M. Emre Celebi, Brian Helba, Michael A. Marchetti, Stephen W. Dusza, Aadi Kalloo, Konstantinos Liopyris, Nabin Mishra, Harald Kittler, Allan Halpern: “Skin Lesion Analysis Toward Melanoma Detection: A Challenge at the 2017 International Symposium on Biomedical Imaging (ISBI), Hosted by the International Skin Imaging Collaboration (ISIC)”, 2017; arXiv:1710.05006.


[2] Tschandl, P., Rosendahl, C. & Kittler, H. The HAM10000 dataset, a large collection of multi-source dermatoscopic images of common pigmented skin lesions. Sci. Data 5, 180161 doi:10.1038/sdata.2018.161 (2018).

Note: Earlier releases of the training data ground truth had issues with corrupted or missing data. Please ensure you are using the ISIC2018_Task2_Training_GroundTruth_v3.zip ground truth bundle for training.