Overview

Schema Version

https://atlas.rsna.org/schemas/2025-11/dataset.json

Name

National Lung Screening Trial (NLST) CT images

Link

https://biometry.nci.nih.gov/cdas/learn/nlst/images/

Indexing

Keywords: lung cancer screening, low-dose chest CT, NLST, Lung-RADS, pulmonary nodules

Content: CH, CT

RadLex: RID50134, RID50149, RID50138, RID10553, RID50137

SNOMED: 363358000, 786838002

Organization(s)

National Cancer Institute (NCI)

Ethical review

Institutional review board approval was granted for each dataset in their respective locations, and all datasets were de-identified prior to transfer (as stated for datasets used in the study).

Comments

Publicly available low-dose chest CT dataset used in this study for case selection (negative cases) within the U.S. reader study.

References

[1] National Lung Screening Trial (NLST). "National Lung Screening Trial (NLST)". . 2014-01-01. Available from: https://www.cancer.gov/types/lung/research/nlst

[2] National Lung Screening Trial Research Team; Aberle DR, Adams AM, et al.. "Reduced lung-cancer mortality with low-dose computed tomographic screening". N Engl J Med. 2011-01-01. PMID: 21714641. PMCID: PMC4356534.

[3] Pinsky PF, Gierada DS, Black W, et al.. "Performance of Lung-RADS in the National Lung Screening Trial: a retrospective assessment". Ann Intern Med. 2015-01-01. PMID: 25664444. PMCID: PMC4705835.

Dataset

Motivation

Used to evaluate an AI assistant for lung cancer screening within randomized retrospective multireader multicase studies.

Sampling

From NLST, all cases with negative biopsies (n=54) were first chosen; then 150 additional negative cases were selected based on nodule size categories (no nodules, 2–6 mm, 6–8 mm, 8–15 mm, and 15 mm+; 30 per category).

Partitioning scheme

This study selected a subset of NLST negative cases for the U.S. reader study; cases were enriched for challenging findings.

Relationships between instances

For reader studies, each case could include the current LDCT and available prior studies.

External data

Used alongside additional U.S., Canadian, and Japan datasets under license or research agreements in this study.

Confidentiality

De-identified DICOM images used for research; de-identified prior to transfer as stated for datasets used in the study.

Sensitive data

No PHI; datasets used in the study were de-identified prior to transfer.