Pulmonary nodule size evaluation with chest tomosynthesis and CT: A phantom study

S. S. Shim, Yu Whan Oh, K. A. Kong, Y. J. Ryu, Y. Kim, D. H. Jang

Research output: Contribution to journalArticlepeer-review

6 Citations (Scopus)

Abstract

Objective:We compared digital tomosynthesis (TOMO) and chest CT in terms of assessing the sizes of nodules located in zones where evaluation by simple radiography is limited. Methods: A total of 48 images comprising phantom nodules of four sizes in six different locations were used. Nodule size measurement errors for measurements using TOMO and CT images compared with the actual size from each observer were calculated. The inter-and intraobserver repeatability of the measured values and the agreement between the two techniques were assessed using the method described by Bland and Altman. Results: The mean measurement errors for all of the nodules and four observers were 20.84mm [standard deviation (SD), 0.60mm] on TOMO and 20.18mm (SD, 0.71mm) on CT images. The mean measurement errors for the different observers ranged from 21.11 to 20.55mm for TOMO and from 20.39 to 0.08mm for CT. Assessing the agreement between nodule size measurements using TOMO and CT resulted in mean measurement errors of 20.65mm, with a 95% limit of agreement of 22.53 to 1.22mm for comparison of TOMO with CT. Conclusion: Our results suggest that nodule sizes obtained using TOMO and chest CT are comparable, even for nodules located in areas where the size measurement is limited on simple radiography. Advances in knowledge: TOMO and CT can be used interchangeably, even for nodules located in a blind area on simple radiography.

Original languageEnglish
Article number20140040
JournalBritish Journal of Radiology
Volume88
Issue number1047
DOIs
Publication statusPublished - 2015 Mar 1

ASJC Scopus subject areas

  • Radiology Nuclear Medicine and imaging

Fingerprint Dive into the research topics of 'Pulmonary nodule size evaluation with chest tomosynthesis and CT: A phantom study'. Together they form a unique fingerprint.

Cite this