Where to find normal CT volumes

[ieee8023]

Although the ideal non-COVID examples would be those suspected of COVID but do not test positive for it like with other types of pneumonia. But in any case here are some other datasets:

Lung CT Segmentation Challenge 2017 (LCTSC)
60 patients
https://academictorrents.com/details/0a3611528c9172383656cb1b6a07cfb7f095eb82

LNDb CT scan dataset (training)
They have lesions though. 294 scans.
https://academictorrents.com/details/e3c196b07c8ea94ac5fca872bccf2cc035f4e88d

Non-Small Cell Lung Cancer CT Scan Dataset (NSCLC-Radiomics-Genomics)
183 scans. They have cancer though.
https://academictorrents.com/details/95b58ebfc1952780cfe2102dd7290889feefad66

LUng Nodule Analysis (LUNA16) All Images
888 scans, they have lesions.
https://academictorrents.com/details/58b053204337ca75f7c2e699082baeb57aa08578

Data Science Bowl 2017 Lung Cancer Detection (DSB3)
Unclear how many scans.
https://academictorrents.com/details/015f31a94c600256868be155358dc114157507fc

Comment to add more sources.

[ncovgt2020]

Two public chest X-ray datasets for computer-aided screening of pulmonary diseases.
Normal/Abnormal tuberculosis X-ray:
https://openi.nlm.nih.gov/faq#faq-tb-coll
The CT volumes (in DICOM format) can be obtained directly from the authors (only available for 1 of the sets).

[jscheithe]

Check out the Lung CT datasets from https://www.cancerimagingarchive.net/

TCIA is a great resource for any kind of (cancer related) medical imaging.

[hasibzunair]

ImageCLEFmed Tuberculosis Challenge (2019)
335 chest CT scans. Labels are CT-wise scale.
https://www.crowdai.org/challenges/imageclef-2019-tuberculosis-severity-scoring

ImageCLEFmed Tuberculosis Challenge (2020)
403 chest CT scans. Labels are on a lung-wise scale rather than CT-wise scale.
https://www.imageclef.org/2020/medical/tuberculosis/