Validation of automated lobe segmentation on paired inspiratory-expiratory chest CT in 8-14 year-old children with cystic fibrosis

• DOI: 10.1371/journal.pone.0194557
• Publication Date: 2018-04-09T17:30:07Z
• Authors: Philip Konietzke; Oliver Weinheimer; Mark O. Wielpütz; Dasha Savage; Tiglath Ziyeh; Christin Tu; Beverly Newman; Craig J. Galbán; Marcus A. Mall; Hans-Ulrich Kauczor; Terry E. Robinson
• Keywords: Air trapping; Lobe; Densitometry

Objectives Densitometry on paired inspiratory and expiratory multidetector computed tomography (MDCT) for the quantification of air trapping is an important approach to assess functional changes in airways diseases such as cystic fibrosis (CF). For a regional analysis deficits, accurate lobe segmentation algorithm applicable scans beneficial. Materials methods We developed fully automated algorithm, subsequently validated automatically generated masks (ALM) against manually corrected (MLM). Paired CTs from 16 children with CF (mean age 11.1±2.4) acquired at 4 time-points (baseline, 3mon, 12mon, 24mon) 2 kernels (B30f, B60f) were segmented, resulting 256 ALM. After manual correction spatial overlap (Dice index) mean differences lung volume calculated ALM vs. MLM. Results The Dice index between MLM was 0.98±0.02 inspiratory, 0.86±0.07 CT. If 6 lobes segmented (lingula treated separate lobe), 0.97±0.02 0.83±0.08 lobar volumes accordance Bland Altman generally low, ranging CT 5.7±52.23cm3 right upper 17.41±14.92cm3 lower lobe. Higher noted even lower, 0±0.01 0.03±0.03 left Conclusions Automatic delivers excellent results good It may become component lobe-based deficits disease, reducing necessity user-interaction post-processing.