NIH publicizes dataset of 32K CT images, representing 4.4K patients
The National Institutes of Health (NIH)'s Clinical Center made a dataset of more than 32,000 annotated lesions identified on CT images representing 4,400 patients available to the public, according to a July 20 release.
The images in the dataset, named DeepLesion, have been anonymized.
"DeepLesion is unlike most lesion medical image datasets currently available, which can only detect one type of lesion," according to the release. "The database has great diversity—it contains all kinds of critical radiology findings from across the body, such as lung nodules, liver tumors, enlarged lymph nodes, and so on."
The dataset should be large enough to train a deep neural network and potentially allow for the creation of a universal lesion detector with one unified framework, according to the NIH. Additionally, studying relationships between different types of lesions and more accurately mearing sizes of all lesions in a patient are goals NIH researchers hope to achieve with the large dataset.
A recent graduate from Dominican University (IL) with a bachelor’s in journalism, Melissa joined TriMed’s Chicago team in 2017 covering all aspects of health imaging. She’s a fan of singing and playing guitar, elephants, a good cup of tea, and her golden retriever Cooper.