Skip to Main Content
Thorax CT Image
+ copyright

NLST CT Images

The NLST CT image collection comes from over 25,000 people who were randomly assigned to the CT arm of NLST. The CT arm protocol was for three annual helical CT exams to screen for lung cancer: one at baseline (T0) and two more on the first and second anniversaries of randomization (T1 and T2). Only images from CT screening exams are available; images from follow-up scans (with higher radiation dose and image quality) are not in the collection. For each screening exam, the image collection contains (on average) one localizer image and 2 to 3 axial reconstructions of a single helical CT scan of the chest. Approximately 200,000 image series from 75,000 CT exams in 25,000 people are available.

CT images do not contain radiologist’s findings. Findings detected in trial screening can be found in the Spiral CT Abnormalities or Spiral CT Comparison Read Abnormalities datasets. Coordinates of findings on CT screens were not collected by the NLST trial.

CT image release is limited to 15,000 participants (of the investigator’s choice) per CDAS project. Investigators are encouraged not to take more images than necessary for their project. The CT images for 15,000 people consist of approximately 80,000 axial reconstructions of 40,000 CT scans and require 6 to 8 terabytes of storage space. Many projects have received only 3000 participants’ images. As the entire CT arm contains only ~600 participants with screen-detected lung cancer, there may be little added value in (for example) having 14,400 cancer-free participants (= 24 per lung cancer) rather than 2,400 cancer-free participants (= 4 per lung cancer). However, the value depends on the project goals, and investigators have discretion to use up to 15,000 people’s images per project.

NLST Pathology Images

The NLST pathology images come from ~450 lung cancer patients in the Lung Screening Study (LSS) subcomponent of NLST. They are images of H&E-stained slides that were obtained as part of a pathology specimen collection to construct tissue microarrays (TMAs). The ~1250 H&E slides (and corresponding images) each came from blocks of lung tissue that were resected during diagnosis and treatment of lung cancer and preserved by pathology labs.

A pathologist reviewed each image to identify histologies of interest for tissue sampling and inclusion in the TMAs. Each region of interest (ROI) was digitally annotated, and the histologies were recorded on case report forms. Data from the case report forms may be obtained by investigators.

How to Access

Access to images is a two-step process.

  • Step 1: Obtain permission. Create a project proposal for NLST data and images through CDAS. You will gain access after the request process is complete (request submitted, approved by NCI, Data Transfer Agreement completed, status set to “Delivered”). A single request can cover a research team of “Approved Users” if the requestor lists their names on the request.
  • Step 2: Obtain images. After obtaining permission (see step 1), you have two options for getting the images:
    1. Ship a hard drive drive to TCIA. (Contact to arrange this.)
    2. Use the TCIA Query Tool to get images. (Instructions)
      The Query Tool is a Web-based application developed and administered by The Cancer Imaging Archive (TCIA) group at University of Arkansas for Medical Sciences, where the images are stored.