Challenges in the Use of NLST Image Data for Quantitative Algorithm Development

One of the products of the National Lung Screening Trial (NLST) is a public database of all the CT images performed during the trial. With 73,118 studies and 203,099 image series, this is the largest publicly available database of chest CT images and is being used for the development and testing of quantitative algorithms, such as lung nodule measurement, detection, and diagnosis. However, the NLST screened participants with CT between 2002 and 2006; these scans are 15-20 years old. Minimum quality guidelines for scan acquisition and reconstruction have been developed by the Quantitative Imaging Biomarker Alliance (QIBA) of the Radiological Society of North America for tasks related to quantitative measurements in the context of LDCT.