Hidenori Shikata

Characterization of the interstitial lung diseases via density-based and texture-based analysis of computed tomography images of lung structure and function1 ☆

RATIONALE AND OBJECTIVES Efforts to establish a quantitative approach to the computed tomography (CT)-based character ization of the lung parenchyma in interstitial lung disease (including emphysema) has been sought. The accuracy of these tools must be site independent. Multi-detector row CT has remained the gold standard for imaging the lung, and it provides the ability to image both lung structure as well as lung function. MATERIAL AND METHODS Imaging is via multi-detector row CT and protocols include careful control of lung volume during scanning. Characterization includes not only anatomic-based measures but also functional measures including regional parameters derived from measures of pulmonary blood flow and ventilation. Image processing includes the automated detection of the lungs, lobes, and airways. The airways provide the road map to the lung parenchyma. Software automatically detects the airways, the airway centerlines, and the branch points, and then automatically labels the airway tree segments with a standardized set of labels, allowing for intersubject as well intrasubject comparisons across time. By warping all lungs to a common atlas, the atlas provides the range of normality for the various parameters provided by CT imaging. RESULTS Imaged density and textural changes mark underlying structural changes at the most peripheral regions of the lung. Additionally, texture-based alterations in the parameters of blood flow may provide early evidence of pathologic processes. Imaging of stable xenon gas provides a regional measure of ventilation which, when coupled with measures of flow, provide for a textural analysis regional of ventilation-perfusion matching. CONCLUSION With the improved resolution and speed of CT imaging, the patchy nature of regional parenchymal pathology can be imaged as texture of structure and function. With careful control of imaging protocols and the use of objective image analysis methods it is possible to provide site-independent tools for the assessment of interstitial lung disease. There remains a need to validate these methods, which requires interdisciplinary and cross-institutional efforts to gather appropriate data bases of images along with a consensus on appropriate ground truths associated with the images. Furthermore, there is the growing need for scanner manufacturers to focus on not just visually pleasing images, but on quantitatifiably accurate images.

Automated segmentation of pulmonary vascular tree from 3D CT images

This paper describes an algorithm for automated segmentation of pulmonary vessels from thoracic 3D CT images. The lung region is roughly extracted based on thresholding and labeling in order to reduce computational cost in the following filtering step. Vessels are enhanced by application of a line-filter, which is based on a combination of eigen values of a Hessian matrix to provide higher response to vessels compared with the other structures. Initial segmentation is performed by thresholding of the filter output. Since extracted vessels may contain tiny holes and local discontinuities between segments, especially around branchpoints, tracking algorithm is used to fill these gaps. Though the results may still contain not only vessels but also parts of airway walls and noise, such structures can be eliminated by considering the number of branchpoints associated with each structure since vascular trees are characterized as objects with many branchpoints. Therefore, a thinning algorithm is applied to determine the number of branchpoints and the final segmentation is obtained by thresholding with regard to the number of branchpoints. We applied the algorithm to five healthy human scans and obtained visually promising results. In order to evaluate our segmentation results quantitatively, approximately 2,000 manually identified points inside the vascular tree were selected in each case to check how many were correctly included in the segmentation result. On average, 98% of the manually identified vessel points were properly marked as vessels. This result demonstrates the promising performance of our algorithm and its utility for further analyses.

Differences in regional wash-in and wash-out time constants for xenon-CT ventilation studies.

UNLABELLED Xenon-enhanced computed tomography (Xe-CT) has been used to measure regional ventilation by determining the wash-in (WI) and wash-out (WO) rates of stable Xe. We tested the common assumption that WI and WO rates are equal by measuring WO-WI in different anatomic lung regions of six anesthetized, supine sheep scanned using multi-detector-row computed tomography (MDCT). We further investigated the effect of tidal volume, image gating (end-expiratory EE versus end-inspiratory EI), local perfusion, and inspired Xe concentration on this phenomenon. RESULTS WO time constant was greater than WI in all lung regions, with the greatest differences observed in dependent base regions. WO-WI time constant difference was greater during EE imaging, smaller tidal volumes, and with higher Xe concentrations. Regional perfusion did not correlate with WI-WO. We conclude that Xe-WI rate can be significantly different from the WO rate, and the data suggest that this effect may be due to a combination of anatomic and fluid mechanical factors such as Rayleigh-Taylor instabilities set up at interfaces between two gases of different densities.

Segmentation of pulmonary vascular trees from thoracic 3D CT images

This paper describes an algorithm for extracting pulmonary vascular trees (arteries plus veins) from three-dimensional (3D) thoracic computed tomographic (CT) images. The algorithm integrates tube enhancement filter and traversal approaches which are based on eigenvalues and eigenvectors of a Hessian matrix to extract thin peripheral segments as well as thick vessels close to the lung hilum. The resultant algorithm was applied to a simulation data set and 44 scans from 22 human subjects imaged via multidetector-row CT (MDCT) during breath holds at 85% and 20% of their vital capacity. A quantitative validation was performed with more than 1000 manually identified points selected from inside the vessel segments to assess true positives (TPs) and 1000 points randomly placed outside of the vessels to evaluate false positives (FPs) in each case. On average, for both the high and low volume lung images, 99% of the points was properly marked as vessel and 1% of the points were assessed as FPs. Our hybrid segmentation algorithm provides a highly reliable method of segmenting the combined pulmonary venous and arterial trees which in turn will serve as a critical starting point for further quantitative analysis tasks and aid in our overall goal of establishing a normative atlas of the human lung.

Xenon gas flow patterns evaluated by high-speed multi-row detector CT

Regional lung ventilation can be measured via Xenon-enhanced computed tomography (Xe-CT) by determining washin (WI) and washout (WO) rates of stable Xe. It has been assumed that WI = WO, ignoring Xe solubility in blood and tissue and then other geometric isssues. We test this by measuring WO-WI in lung by Xe-CT. Also, we investigate the effect of tidal volume (TV) and end inspiratory (EI) vs end expiratory (EE) scan gating on WO and WI measurements. 3 anesthetized, supine sheep were scanned using multidetector-row computed tomography (MDCT). Imaging was gated to both EE and EI during a WI (33 breaths) and WO (20 breaths) maneuver using 55% Xe for WI and room air for WO. Time constants (TCs) of Xe WI and WO were obtained by exponential fitting. WO and WI TCs were compared: 1) apex and base 2) dependent, middle, and nondependent 3) EE and EI 4) three TVs. The vertical gradient of WO-WI showed WO > WI in dependent vs non-dependent regions. WO-WI in both dependent and nondependent region at the lung base and apex was larger when measured at EE compared to EI. As TV increases, the global WO-WI difference decreased. TV showed greater influence on WO than WI. Xe WO was longer than WI possibly reflecting Xe solubility in blood and tissue. Higher TVs and gating to EE provided greater effects on WO than WI TCs which may relate to the number of partial volumed conducting airways contributing to the regional voxel-based measures. We conclude that WO mode is more susceptible to errors caused by either xenon solubility or tidal volume than WI mode and EE scanning may more accurately reflect alveolar ventilation.

Three-dimensional and Four-dimensional Cardiopulmonary Image Analysis

Modern medical imaging equipment can provide data that describe the anatomy and function of structures in the body. Image segmentation techniques are needed to take this raw data and identify and delineate the relevant cardiovascular and pulmonary anatomy to put it into a form suitable for 3D and 4D modeling and simulation. These methods must be able to handle large multi-dimensional data sets, possibly limited in resolution, corrupted by noise and motion blur, and sometimes depicting unusual anatomy due to natural shape variation across the population or due to disease processes. This chapter describes modern techniques for robust, automatic image segmentation. Several applications in cardiovascular and pulmonary imaging are presented.