boneage-training-dataset.csv -> contains all of the filenames, gender, and age of 12,611 samples.
train_boneage.csv -> I took 90% of the above as training set
validation_boneage.csv -> remaining 10% of the above as validation set
random_train_boneage.csv -> subsample of the train_boneage.csv
random_validation_boneage.csv -> subsample of the validation_boneage.csv
boneage-training-dataset (zipped folder, attached in this email) -> contains all 12,611 images
Further reading: https://www.rsna.org/education/ai-resources-and-training/ai-image-challenge/RSNA-Pediatric-Bone-Age-Challenge-2017
https://www.biorxiv.org/content/biorxiv/early/2018/06/20/234120.full.pdf