Aims and objectives
Lung nodule volumetry is a perspective alternative for conventional diameter measurements in follow-up studies.
The method is highly reproducible and more accurate,
especially for non spherical nodules.
But if nodule it is quite difficult for software to segment nodule that has contact with vessels or pleura.
There are two way for radiologist to solve this problem: not use such nodules in analysis or manually correct automatic segmentation results.
The objective of this study was to evaluate variability of lung nodules volumetry after manual correction of...
Methods and materials
At Russian Scientific Center for Radiology and Surgical Technologies (Saint-Petersburg) database CT images of 7 patients with disseminated metastatic lung lesions were evaluated.
We have chosen 27 perivascular and parapleural nodules.
Images were segmented and analized in Mango v4.0.1 (http://ric.uthscsa.edu/mango/).
Segmentation was made with following algorithm.
Creation of sphere ROI with diameter 2x larger then nodule diameter .
Exclusion from ROI voxels with density that is different from nodule density using histogram tool Fig. 2 Fig. 3.
Exclusion of structures that are not connected with...
Results
Mean nodules volume was in range 19.5-16493.2 mm3.
There were no significant differences found between VC of different observers (p=0.365).
The best linear model that describes dependency of VC from nodule size and contact area was:
CV = exp(α*Effective diameter + β*Contact area+i),
where α,β,i are model coefficients .
Model coefficient
α
-0.08
β
4.57
i
-1.40
R2 for this model was 0.72.
Shapiro-Wilk test of model residuals has shown their normal distribution (p=0.62).
Also there wasn't found any heteroscedasticity of residuals or their correlation...
Conclusion
It was shown variation of volumetry isn't affected by radiologist experience.
Exponential model gives us good prediction of volumetry variation from nodule size and the area of contact between nodule and other structures like vessels and pleura.
The main limitation was that we evaluate variability of volumetry on single studies,
while volumetry mainly used in follow-up studies.
References
Quantitative Imaging Biomarkers: A Review of Statistical Methods for Computer Algorithm Comparisons/ Nancy A.
Obuchowski,
Anthony P.
Reeves,
Erich P.
Huang [et al.].— Vol.
24,
no.
1.— P.
68–106.—
pmid : 24919829.