ECR 2012 / B-0417
A scoring system for tomosynthesis in pulmonary cystic fibrosis
Keywords:
Lung, Digital radiography
Authors:
K. Vult von Steyern, I. Björkman-Burtscher, P. Höglund, G. C. Bozovic, M. Wiklund, M. Geijer; Lund/SE
DOI:
10.1594/ecr2012/B-0417
Results
- Observer agreements for the tomosynthesis score were almost perfect for the total score with square-weighted kappa >0.90,
and generally substantial to almost perfect for subscores [3].
- Observer agreements for the total score for the Brasfield score were also almost perfect (square-weighted kappa 0.80,
0.81 and 0.85).
- Correlation between the tomosynthesis score and the Brasfield score was good for the three observers (Kendall’s rank correlation tau 0.68,
0.77 and 0.78).
- The total disease severity scores were generally higher for tomosynthesis than for radiography in percentage of the maximum score (Figs.
8-10,
patient 3).
Many patients that were scored normal for linear markings with radiography were scored for presence of bronchial wall thickening with tomosynthesis.
Patients without apparent nodular-cystic lesions on radiographs had bronchiectases or mucus plugs detectable with tomosynthesis (Figs.
11-12,
patient 4,
Figs.
13-15,
patient 5).
- Diversity in the tomosynthesis scores was greater than in the Brasfield scores.