Background: Assessing heterogeneity in lung images can be an important diagnosis tool. We present a novel and\nobjective method for assessing lung damage in a rat model of emphysema. We combined a three-dimensional\n(3D) computer graphics methodââ?¬â??octree decompositionââ?¬â??with a geostatistics-based approach for assessing spatial\nrelationshipsââ?¬â??the variogramââ?¬â??to evaluate disease in 3D computed tomography (CT) image volumes.\nMethods: Male, Sprague-Dawley rats were dosed intratracheally with saline (control), or with elastase dissolved\nin saline to either the whole lung (for mild, global disease) or a single lobe (for severe, local disease). Gated 3D\nmicro-CT images were acquired on the lungs of all rats at end expiration. Images were masked, and octree\ndecomposition was performed on the images to reduce the lungs to homogeneous blocks of 2 Ã?â?? 2 Ã?â?? 2, 4 Ã?â?? 4 Ã?â?? 4,\nand 8 Ã?â?? 8 Ã?â?? 8 voxels. To focus on lung parenchyma, small blocks were ignored because they primarily defined\nboundaries and vascular features, and the spatial variance between all pairs of the 8 Ã?â?? 8 Ã?â?? 8 blocks was calculated\nas the square of the difference of signal intensity. Variogramsââ?¬â??graphs of distance vs. varianceââ?¬â??were constructed,\nand results of a least-squares-fit were compared. The robustness of the approach was tested on images prepared\nwith various filtering protocols. Statistical assessment of the similarity of the three control rats was made with a\nKruskal-Wallis rank sum test. A Mann-Whitney-Wilcoxon rank sum test was used to measure statistical distinction\nbetween individuals. For comparison with the variogram results, the coefficient of variation and the emphysema\nindex were also calculated for all rats.\nResults: Variogram analysis showed that the control rats were statistically indistinct (p = 0.12), but there were\nsignificant differences between control, mild global disease, and severe local disease groups (p < 0.0001). A\nheterogeneity index was calculated to describe the difference of an individual variogram from the control average.\nThis metric also showed clear separation between dose groups. The coefficient of variation and the emphysema\nindex, on the other hand, did not separate groups.
Loading....