Automatic MRI adipose tissue mapping using overlapping mosaics

This paper presents an automatic method of correcting non-uniform RF coil response for the classification of body composition using MR imaging. By linear mosaic modelling, the smoothly but non-linearly varying bias field, which modulates tissue intensities within the image, was corrected. The overlapping between adjacent mosaics ensured consistent segmentation of body fat content and the effectiveness of the technique was validated by both phantom and in vivo experiments. Ten whole body composition data sets, each with 39 trans-axial slices, were acquired. Automatic segmentation results using the proposed technique were compared with those from manual delineations. The automatic segmentation method was found to be highly accurate and the mean percentage error between the two methods was less than 1.5%.

ERS transform for the automated detection of bronchial abnormalities on CT of the lungs

The identification of bronchi on Computed Tomography (CT) images of the lungs provides valuable clinical information in patients with suspected airways diseases including bronchiectasis, emphysema, or constrictive obliterative bronchiolitis. The automated recognition of the airways is, therefore, an important part of a diagnosis aid system for resolving potential ambiguities associated with intensity-based feature extractors. On CT images, near-perpendicular cross sections of bronchi normally appear as elliptical rings and this paper presents a novel technique for their recognition. The proposed method, the edge-radius-symmetry (ERS) transform, is based on the analysis of the distribution of edges in local polar coordinates. Pixels are ranked according to local edge (E) strength, radial (R), uniformity and local symmetry (S). A discrete implementation of the technique is provided which reduces the computational cost of the ERS transform by using a geometric approximation of the intensity patterns. The identification of the adjacent pulmonary vessels with template matching then allows for the automated measurement of bronchial dilatation and bronchial wall thickening. Computationally, the method compares favorably with other methods such as the Hough transform. Noise-sensitivity of the technique was evaluated on a set of synthetic images and nine patients under investigation for suspected airways disease. Agreement for the automated scoring of the presence and severity of bronchial abnormalities was demonstrated to be comparable to that of an experienced radiologist (kappa statistics kappa > 0.5 ).

Functional consequences of pleural disease evaluated with chest radiography and CT

PURPOSE

To identify a system for the quantification of pleural thickening with an acceptable level of interobserver variation and good functional correlation in individuals with pleural disease.

MATERIALS AND METHODS

The extent of pleural thickening and plaques was assessed in 50 patients by using the following: (a) a radiographic score based on the International Labour Office system, (b) a subjective simple computed tomographic (CT) score, (c) a subjective comprehensive CT score, (d) an objective nonautomated method, and (e) an objective computer-aided semiautomated method.

RESULTS

Similar correlations between the extent of diffuse pleural thickening and forced vital capacity were seen for each system (objective CT, r = -0.72, P <.001; simple CT, r = -0.69, P <.001; radiographic, r = -0.67, P <.001; comprehensive CT, r = -0.66, P <.001). Comparable correlations were observed for total lung capacity. After controlling for extent of diffuse pleural thickening, pleural plaque scores were functionally irrelevant.

CONCLUSION

Comparable functional-morphologic correlations were achieved by using different CT and radiographic scoring systems for pleural disease. A subjective simple CT system had the advantages of ease of application and potential to aid in the accurate assessment of the lung parenchyma, which may be important in individuals exposed to asbestos.

Gradient correction and classification of CT lung images for the automated quantification of mosaic attenuation pattern

PURPOSE

The detection of density differences, or "mosaic attenuation pattern," on CT images may be difficult when the regional inhomogeneity of the density of the lung parenchyma is subtle. The purpose of this work was to develop a fully automated method for the reproducible quantification of the underattenuated areas of the lung parenchyma. This technique may be useful in increasing the precision of investigation of structure/function relationships.

METHOD

Anatomical segmentation was achieved by a structure-filtering operator based on mathematical morphology. To compensate for the density gradient visible on lung CT scans, a model-based iterative deconvolution filter and an adaptive clustering algorithm were developed. Validation was performed with CT images from a lung phantom, 15 patients with constrictive obliterative bronchiolitis, and 8 normal subjects.

RESULTS

The accuracy of the estimate of the density gradient on phantom studies was 93.3%. The automated quantification of the areas of decreased attenuation on scans of constrictive obliterative bronchiolitis was within 8.2% from the average scoring of two experienced observers.

CONCLUSION

The proposed technique is fully automated and can accurately correct for density gradient and classify areas of decreased attenuation on lung CT images.

A comparison of methods for enhancing the detection of areas of decreased attenuation on CT caused by airways disease

PURPOSE

The purpose of this work was to investigate thin section CT image enhancement of subtle areas of decreased attenuation of the lung parenchyma in suspected airways disease.

METHOD

Forty-seven consecutive patients with chronic sputum production underwent pulmonary function tests and high resolution CT (HRCT). Single section inspiratory (INSP), expiratory (EXP), and minimum intensity projection (MINIP) images through the lower lobes were acquired. A histogram stretch was applied to the INSP and MINIP images, generating two further image formats. The five image types were compared for the extent of decreased attenuation, observer confidence, and correlations with pulmonary function tests.

RESULTS

Interobserver variation was lowest with MINIP images (mean weighted K: MINIP 0.70, INSP sections 0.65, other image formats < or =0.48). Observers were most confident with EXP and MINIP images. EXP sections identified more disease than MINIP images (p<0.001). Correlations with pulmonary function tests were similar for each image format.

CONCLUSION

The HRCT changes of small airways disease are enhanced with image postprocessing. MINIP images are associated with increased observer confidence and agreement as compared with HRCT alone.

Mechanism of mosaic attenuation of the lungs on computed tomography in induced bronchospasm

The purpose of this study was to investigate whether hypoxic pulmonary vasoconstriction is the major determinant of the computed tomography (CT) pattern of mosaic attenuation in asthmatic patients with induced bronchoconstriction. Thin-section CT was performed at suspended full inspiration immediately and 30 min after methacholine bronchoprovocation in 22 asthmatic subjects, who were randomly assigned to breathe room air (group A, n = 8), oxygen via nasal prongs at 5 l/min (group B, n = 8), and oxygen via face mask at 12 l/min (group C, n = 6). CT changes were quantified in terms of global lung density and density in hypodense and hyperdense areas. Lung parenchymal density increases were greatest in group C and greater in group B than in group A, globally (P = 0.03) and in hypodense regions (P = 0.01). On bivariate analysis, the only change in cross-sectional area was related to change in global density. In hypodense regions, density change was related both to reduction in cross-sectional area (P < 0.0005) and to oxygen administration (P = 0.01). After correction for changes in global lung density, only oxygen was independently related to density increase in hypodense areas (P = 0.02). In induced bronchoconstriction, the CT appearance of mosaic attenuation can be largely ascribed to hypoxic vasoconstriction rather than to changes in lung inflation.

Enhancement of subtle density differences of the lung parenchyma on CT

Several pulmonary conditions are characterized by the finding of areas of decreased attenuation relative to adjacent normal lung parenchyma on computed tomography. In particular, the identification of areas of the lung affected by small airways disease is difficult as the reduction in lung density is subtle and the regions are poorly marginated. This paper presents a novel wavelet analysis technique which increases the conspicuity of differences between normal and abnormal lung parenchyma.