NIH Deep Lesion Dataset

tensorflow / datasets

TFDS is a collection of datasets ready to use with TensorFlow, Jax, ...

https://www.tensorflow.org/datasets

Apache License 2.0

4.3k stars 1.54k forks source link

NIH Deep Lesion Dataset #853

Open Ouwen opened 5 years ago

Ouwen commented 5 years ago

Name of dataset: NIH Deep Lesion Dataset
URL of dataset: https://www.nih.gov/news-events/news-releases/nih-clinical-center-releases-dataset-32000-ct-images
License of dataset: Unknown
Short description of dataset and use case(s): The National Institutes of Health’s Clinical Center has made a large-scale dataset of CT images publicly available to help the scientific community improve detection accuracy of lesions. While most publicly available medical image datasets have less than a thousand lesions, this dataset, named DeepLesion, has over 32,000 annotated lesions identified on CT images. The images, which have been thoroughly anonymized, represent 4,400 unique patients, who are partners in research at the NIH.

Folks who would also like to see this dataset in tensorflow/datasets, please thumbs-up so the developers can know which requests to prioritize.

And if you'd like to contribute the dataset (thank you!), see our guide to adding a dataset.

Ouwen commented 5 years ago

@jason-zl190

zl190 commented 5 years ago

@jason-zl190

Working on it!

suyashkumar commented 5 years ago

For reference, here's the download URLs of the 56 4GB zip files: https://nihcc.app.box.com/v/DeepLesion/file/316932195591 they all contain PNGs. Would be happy to assist with this!

See also: the code @Ouwen and I used at Gradient Health to populate our cloud bucket https://github.com/suyashkumar/deeplesion-gcp-loader

zl190 commented 5 years ago

For reference, here's the download URLs of the 56 4GB zip files: https://nihcc.app.box.com/v/DeepLesion/file/316932195591 they all contain PNGs. Would be happy to assist with this!

See also: the code @Ouwen and I used at Gradient Health to populate our cloud bucket https://github.com/suyashkumar/deeplesion-gcp-loader

Thanks for the help.

zl190 commented 5 years ago

@Ouwen I create a pull request last week. chexpert, coco and stanford_dogs are the references. The current version depends on manually download and user needs to provide csv files containing split info. I shall include the download function and split logic in the next version. They were tested locally already.

Ouwen commented 5 years ago

https://github.com/tensorflow/datasets/pull/1006

zl190 commented 4 years ago

@Ouwen I create a new pull request and closed the old one because it's outdated. I also create a new issue for it #1225 . Would you mind be my reviewer? This version no longer needs to manually download and place the data.