Eric A. Hoffman

Automatic lung segmentation for accurate quantitation of volumetric X-ray CT images

Segmentation of pulmonary X-ray computed tomography (CT) images is a precursor to most pulmonary image analysis applications. This paper presents a fully automatic method for identifying the lungs in three-dimensional (3-D) pulmonary X-ray CT images. The method has three main steps. First, the lung region is extracted from the CT images by gray-level thresholding. Then, the left and right lungs are separated by identifying the anterior and posterior junctions by dynamic programming. Finally, a sequence of morphological operations is used to smooth the irregular boundary along the mediastinum in order to obtain results consistent with these obtained by manual analysis, in which only the most central pulmonary arteries are excluded from the lung region. The method has been tested by processing 3-D CT data sets from eight normal subjects, each imaged three times at biweekly intervals with lungs at 90% vital capacity. The authors present results by comparing their automatic method to manually traced borders from two image analysts. Averaged over all volumes, the root mean square difference between the computer and human analysis is 0.8 pixels (0.54 mm). The mean intrasubject change in tissue content over the three scans was 2.75%/spl plusmn/2.29% (mean/spl plusmn/standard deviation).

Percent Emphysema, Airflow Obstruction, and Impaired Left Ventricular Filling

BACKGROUND Very severe chronic obstructive pulmonary disease causes cor pulmonale with elevated pulmonary vascular resistance and secondary reductions in left ventricular filling, stroke volume, and cardiac output. We hypothesized that emphysema, as detected on computed tomography (CT), and airflow obstruction are inversely related to left ventricular end-diastolic volume, stroke volume, and cardiac output among persons without very severe lung disease. METHODS We measured left ventricular structure and function with the use of magnetic resonance imaging in 2816 persons who were 45 to 84 years of age. The extent of emphysema (expressed as percent emphysema) was defined as the percentage of voxels below -910 Hounsfield units in the lung windows on cardiac computed tomographic scans. Spirometry was performed according to American Thoracic Society guidelines. Generalized additive models were used to test for threshold effects. RESULTS Of the study participants, 13% were current smokers, 38% were former smokers, and 49% had never smoked. A 10-point increase in percent emphysema was linearly related to reductions in left ventricular end-diastolic volume (-4.1 ml; 95% confidence interval [CI], -3.3 to -4.9; P<0.001), stroke volume (-2.7 ml; 95% CI, -2.2 to -3.3; P<0.001), and cardiac output (-0.19 liters per minute; 95% CI, -0.14 to -0.23; P<0.001). These associations were of greater magnitude among current smokers than among former smokers and those who had never smoked. The extent of airflow obstruction was similarly associated with left ventricular structure and function, and smoking status had similar modifying effects on these associations. Percent emphysema and airflow obstruction were not associated with the left ventricular ejection fraction. CONCLUSIONS In a population-based study, a greater extent of emphysema on CT scanning and more severe airflow obstruction were linearly related to impaired left ventricular filling, reduced stroke volume, and lower cardiac output without changes in the ejection fraction.

Practical reconstruction method for bioluminescence tomography

Bioluminescence tomography (BLT) is used to localize and quantify bioluminescent sources in a small living animal. By advancing bioluminescent imaging to a tomographic framework, it helps to diagnose diseases, monitor therapies and facilitate drug development. In this paper, we establish a direct linear relationship between measured surface photon density and an unknown bioluminescence source distribution by using a finite-element method based on the diffusion approximation to the photon propagation in biological tissue. We develop a novel reconstruction algorithm to recover the source distribution. This algorithm incorporates a priori knowledge to define the permissible source region in order to enhance numerical stability and efficiency. Simulations with a numerical mouse chest phantom demonstrate the feasibility of the proposed BLT algorithm and reveal its performance in terms of source location, density, and robustness against noise. Lastly, BLT experiments are performed to identify the location and power of two light sources in a physical mouse chest phantom.

Registration-based estimates of local lung tissue expansion compared to xenon CT measures of specific ventilation

The main function of the respiratory system is gas exchange. Since many disease or injury conditions can cause biomechanical or material property changes that can alter lung function, there is a great interest in measuring regional lung ventilation and regional specific volume change. We describe a registration-based technique for estimating local lung expansion from multiple respiratory-gated CT images of the thorax. The degree of regional lung expansion is measured using the Jacobian (a function of local partial derivatives) of the registration displacement field, which we show is directly related to specific volume change. We compare the ventral-dorsal patterns of lung expansion estimated across five pressure changes to a xenon CT based measure of specific ventilation in five anesthetized sheep studied in the supine orientation. Using 3D image registration to match images acquired at 10 cm H(2)O and 15 cm H(2)O airway pressures gave the best match between the average Jacobian and the xenon CT specific ventilation (linear regression, average r(2)=0.73).

Atlas-driven lung lobe segmentation in volumetric X-ray CT images

High-resolution X-ray computed tomography (CT) imaging is routinely used for clinical pulmonary applications. Since lung function varies regionally and because pulmonary disease is usually not uniformly distributed in the lungs, it is useful to study the lungs on a lobe-by-lobe basis. Thus, it is important to segment not only the lungs, but the lobar fissures as well. In this paper, we demonstrate the use of an anatomic pulmonary atlas, encoded with a priori information on the pulmonary anatomy, to automatically segment the oblique lobar fissures. Sixteen volumetric CT scans from 16 subjects are used to construct the pulmonary atlas. A ridgeness measure is applied to the original CT images to enhance the fissure contrast. Fissure detection is accomplished in two stages: an initial fissure search and a final fissure search. A fuzzy reasoning system is used in the fissure search to analyze information from three sources: the image intensity, an anatomic smoothness constraint, and the atlas-based search initialization. Our method has been tested on 22 volumetric thin-slice CT scans from 12 subjects, and the results are compared to manual tracings. Averaged across all 22 data sets, the RMS error between the automatically segmented and manually segmented fissures is 1.96/spl plusmn/0.71 mm and the mean of the similarity indices between the manually defined and computer-defined lobe regions is 0.988. The results indicate a strong agreement between the automatic and manual lobe segmentations.

Characteristics of the turbulent laryngeal jet and its effect on airflow in the human intra-thoracic airways

A computational fluid dynamics technique is applied to understand the relative importance of the upper and intra-thoracic airways and their role in determining central airflow patterns with particular attention paid to the importance of turbulence. The geometry of the human upper respiratory tract is derived from volumetric scans of a volunteer imaged via multidetector-row computed tomography. Geometry 1 consists of a mouthpiece, the mouth, the oropharynx, the larynx, and the intra-thoracic airways of up to six generations. Geometry 2 comprises only the intra-thoracic airways. The results show that a curved sheet-like turbulent laryngeal jet is observed only in geometry 1 with turbulence intensity in the trachea varying from 10% to 20%, whereas the turbulence in geometry 2 is negligible. The presence of turbulence is found to increase the maximum localised wall shear stress by three-folds. The proper orthogonal decomposition analysis reveals that the regions of high turbulence intensity are associated with Taylor-Görtler-like vortices. We conclude that turbulence induced by the laryngeal jet could significantly affect airway flow patterns as well as tracheal wall shear stress. Thus, airflow modeling, particularly subject specific evaluations, should consider upper as well as intra-thoracic airway geometry.

Impaired Mucus Detachment Disrupts Mucociliary Transport in a Piglet Model of Cystic Fibrosis

A breathtaking tale of sticky mucus Patients with cystic fibrosis have difficulty breathing because their airways are clogged with thick mucus. Does this mucus accumulate because there is a defect in the way it is produced? Or does it accumulate because of other disease features, such as dehydration or airway wall remodeling? Distinguishing between these possibilities is important for future drug development. In a study of piglets with cystic fibrosis, Hoegger et al. identify mucus production as the primary defect (see the Perspective by Wine). The airway glands of the piglets synthesized strands of mucus normally, but the strands were never released and stayed tethered to the gland ducts. Science, this issue p. 818; see also p. 730 Lung disease in pigs with cystic fibrosis is caused by aberrant tethering of mucus to the airway glands that produce it. [Also see Perspective by Wine] Lung disease in people with cystic fibrosis (CF) is initiated by defective host defense that predisposes airways to bacterial infection. Advanced CF is characterized by a deficit in mucociliary transport (MCT), a process that traps and propels bacteria out of the lungs, but whether this deficit occurs first or is secondary to airway remodeling has been unclear. To assess MCT, we tracked movement of radiodense microdisks in airways of newborn piglets with CF. Cholinergic stimulation, which elicits mucus secretion, substantially reduced microdisk movement. Impaired MCT was not due to periciliary liquid depletion; rather, CF submucosal glands secreted mucus strands that remained tethered to gland ducts. Inhibiting anion secretion in non-CF airways replicated CF abnormalities. Thus, impaired MCT is a primary defect in CF, suggesting that submucosal glands and tethered mucus may be targets for early CF treatment.

Clinical Significance of Symptoms in Smokers with Preserved Pulmonary Function

BACKGROUND Currently, the diagnosis of chronic obstructive pulmonary disease (COPD) requires a ratio of forced expiratory volume in 1 second (FEV1) to forced vital capacity (FVC) of less than 0.70 as assessed by spirometry after bronchodilator use. However, many smokers who do not meet this definition have respiratory symptoms. METHODS We conducted an observational study involving 2736 current or former smokers and controls who had never smoked and measured their respiratory symptoms using the COPD Assessment Test (CAT; scores range from 0 to 40, with higher scores indicating greater severity of symptoms). We examined whether current or former smokers who had preserved pulmonary function as assessed by spirometry (FEV1:FVC ≥0.70 and an FVC above the lower limit of the normal range after bronchodilator use) and had symptoms (CAT score, ≥10) had a higher risk of respiratory exacerbations than current or former smokers with preserved pulmonary function who were asymptomatic (CAT score, <10) and whether those with symptoms had different findings from the asymptomatic group with respect to the 6-minute walk distance, lung function, or high-resolution computed tomographic (HRCT) scan of the chest. RESULTS Respiratory symptoms were present in 50% of current or former smokers with preserved pulmonary function. The mean (±SD) rate of respiratory exacerbations among symptomatic current or former smokers was significantly higher than the rates among asymptomatic current or former smokers and among controls who never smoked (0.27±0.67 vs. 0.08±0.31 and 0.03±0.21 events, respectively, per year; P<0.001 for both comparisons). Symptomatic current or former smokers, regardless of history of asthma, also had greater limitation of activity, slightly lower FEV1, FVC, and inspiratory capacity, and greater airway-wall thickening without emphysema according to HRCT than did asymptomatic current or former smokers. Among symptomatic current or former smokers, 42% used bronchodilators and 23% used inhaled glucocorticoids. CONCLUSIONS Although they do not meet the current criteria for COPD, symptomatic current or former smokers with preserved pulmonary function have exacerbations, activity limitation, and evidence of airway disease. They currently use a range of respiratory medications without any evidence base. (Funded by the National Heart, Lung, and Blood Institute and the Foundation for the National Institutes of Health; SPIROMICS ClinicalTrials.gov number, NCT01969344.).

In vivo mouse studies with bioluminescence tomography.

Bioluminescence tomography (BLT) is a new molecular imaging mode, which is being actively developed to reveal molecular and cellular signatures as labeled by bioluminescent probes in a living small animal. This technology can help diagnose diseases, evaluate therapies, and facilitate drug development with mouse models. In this paper, we describe in vivo mouse experiments with BLT, and propose the reconstruction procedure of bioluminescent sources from optical data measured on the body surface of the mouse using a modality fusion approach. The results show the feasibility of our methodology for localization and quantification of the bioluminescent activities in vivo.

Severe asthma: lessons learned from the National Heart, Lung, and Blood Institute Severe Asthma Research Program.

The National Heart, Lung, and Blood Institute Severe Asthma Research Program (SARP) has characterized over the past 10 years 1,644 patients with asthma, including 583 individuals with severe asthma. SARP collaboration has led to a rapid recruitment of subjects and efficient sharing of samples among participating sites to conduct independent mechanistic investigations of severe asthma. Enrolled SARP subjects underwent detailed clinical, physiologic, genomic, and radiological evaluations. In addition, SARP investigators developed safe procedures for bronchoscopy in participants with asthma, including those with severe disease. SARP studies revealed that severe asthma is a heterogeneous disease with varying molecular, biochemical, and cellular inflammatory features and unique structure-function abnormalities. Priorities for future studies include recruitment of a larger number of subjects with severe asthma, including children, to allow further characterization of anatomic, physiologic, biochemical, and genetic factors related to severe disease in a longitudinal assessment to identify factors that modulate the natural history of severe asthma and provide mechanistic rationale for management strategies.