Quantitative analysis of inhomogeneity in ventilation SPET
2001 (English)In: European Journal of Nuclear Medicine, ISSN 0340-6997, E-ISSN 1432-105x, Vol. 28, no 12, 1795-1800 p.Article in journal (Refereed) Published
The aim of this study was to evaluate a method for quantification of inhomogeneity in ventilation single-photon emission tomography (SPET). Nine emphysematous patients, nine life-long non-smokers and nine smokers were included in the study. The SPET investigation was performed after 50 MBq (99m)Tc-Technegas had been inhaled by each subject in the supine position. A single-head gamma camera, equipped with a general-purpose parallel-hole collimator using 64 projections (20 s each) over 360 degrees, was used. Data were acquired in 128x128 matrices. Attenuation correction was applied based upon computed tomography (CT) density maps. Lung regions of interest were delineated manually on CT images and then positioned on SPET images. Several attenuation-corrected transaxial SPET slices (thickness 1 cm, spacing 3.5 cm) were reconstructed. Each SPET slice was divided into several 2x2x1 cm(3) elements. Inhomogeneity was assessed by the coefficient of variation (CV) of the pixel counts within these elements (micro-level) and the CV of the total counts of the elements (macro-level). Micro-level CVs in non-smokers varied between 1% and 41%, whereas they were dispersed over a wide range (1%-600%) in emphysematous patients. In seven smokers, the frequency distribution of micro-level CVs was within the normal range, whereas in the other two smokers the values were between the normal range and the range in emphysematous patients. The pooled mean values of micro-level CVs and macro-level CVs in each subject clearly separated the patients from the others. Parametric images of micro-level CV indicated the localisation and severity of ventilation inhomogeneity. We conclude that the present method enables quantification and localisation of regional inhomogeneity in ventilation SPET images.
Place, publisher, year, edition, pages
Springer, 2001. Vol. 28, no 12, 1795-1800 p.
Medical Image Processing
IdentifiersURN: urn:nbn:se:liu:diva-79209DOI: 10.1007/s002590100649PubMedID: 11734917OAI: oai:DiVA.org:liu-79209DiVA: diva2:539132