David A. Lynch

IMAGING OF ASTHMA AND ALLERGIC BRONCHOPULMONARY MYCOSIS

The chest radiograph of the patient with asthma is characterized by bronchial wall thickening and hyperinflation. On CT scanning of patients with asthma one may see airway wall thickening, thickened centrilobular structures, and focal or diffuse hyperlucency. Apparent bronchial dilation may be seen, but the diagnosis of bronchiectasis should be made with caution. Quantification of changes in the airway wall and lung parenchyma may be valuable in understanding the mechanisms of asthma and in evaluating the effects of treatment. Central bronchiectasis occurs in most, but not all, cases of ABPA. Patchy airspace opacity may be the sole radiologic manifestation of ABPA in some cases. Other fungi can rarely cause a similar syndrome. The challenge for the radiologist evaluating the images of a patient with asthma is to find complications, such as ABPA, or alternative diagnoses.

HIGH-RESOLUTION CT OF IDIOPATHIC INTERESTITIAL PNEUMONIAS

Important recent changes have occurred in our understanding of the IIPs. IPF (characterized histologically as UIP) is recognized as a progressive disease with a relatively poor prognosis, and with a characteristic CT appearance. The radiologist must be able to distinguish between UIP and the other IIPs. Complications of IPF include accelerated progression, lung cancer, and secondary infection. NSIP has a better prognosis than IPF, and has ground-glass attenuation as its salient CT feature. COP (formerly known as BOOP) is included as an IIP because its clinical, physiologic, and imaging features overlap with those of the other IIPs. It is characterized on CT by consolidation and ground-glass attenuation. AIP is the idiopathic form of ARDS. LIP and DIP are less common IIPs, both characterized by ground-glass attenuation.

Automated discrimination and quantification of idiopathic pulmonary fibrosis from normal lung parenchyma using generalized fractal dimensions in high-resolution computed tomography images

RATIONALE AND OBJECTIVESnWe computed generalized fractal dimensions for high-resolution computed tomography (HRCT) images to investigate their value in the discrimination and quantification of idiopathic pulmonary fibrosis (IPF) from normal lung parenchyma.nnnMETHODSnA probability distribution that was based on the pixel value in each image was used to compute capacity, information, and higher fractal dimensions for a series of 52 HRCT slices obtained from four patients. Qualitative classification of normal, mild, moderate, and severe IPF cases was achieved by computing the following parameter: DD = D0 - 2D1 + D2, where D0, D1, and D2 represents the capacity, information, and pair correlation dimensions, respectively. A multiple linear regression analysis using morphometric quantification for the set of 52 slices was tested for all possible combinations of the parameters D0, D1, D2, and D3. The generalizability of the model was tested by predicting the extent of IPF for each patient from a regression model computed with the remaining slices in the database.nnnRESULTSnThe best regression results were obtained using the independent parameters D1 and D2 to quantify the extent of diseased lung parenchyma. The technique was tested with 48 slices from 12 new patients. The results indicated that the extent of IPF could be predicted within the confidence limits given by the regression analysis.nnnCONCLUSIONnThe extent of IPF can be predicted well within the 90% confidence interval given by the model. The width of the confidence interval decreases as the number of slices used in the linear regression model increases. This operator-independent quantitative technique may be useful in the follow-up of patients with IPF.

Asthma: An Imaging Update

Significant advances continue in the subjective and quantifiable imaging features of asthma. Radiologists need to be aware of not only the general features, but also potential asthma mimics as well as complications.

Right Ventricular Phenotypic Characteristics in Subjects With Primary Pulmonary Hypertension or Idiopathic Dilated Cardiomyopathy

BACKGROUNDnStudies of animal models and human subjects with cardiomyopathies suggest that cardiac myocyte and ventricular chamber remodeling show distinct phenotypic characteristics that may be dependent on specific signaling pathways.nnnMETHODS AND RESULTSnIn this study, we characterize right ventricular (RV) chamber size, end-diastolic thickness, myocardial mass, and ejection fraction (EF) in human subjects with chronic heart failure from primary pulmonary hypertension (PPH; n = 10) and idiopathic dilated cardiomyopathy (IDC; n = 10). Subjects underwent gated cardiac magnetic resonance imaging (MRI), and the RVs were phenotypically classified based on the presence or absence of hypertrophy (increased mass), systolic dysfunction (reduced EF), and degree of wall thickness (concentric v eccentric pattern of hypertrophy). Within this schema, five abnormal phenotypes could be identified. In PPH subjects, in whom the RV is subjected to the uniform insult of chronic pressure overload, four different abnormal phenotypes were identified.nnnCONCLUSIONSnThese data indicate that distinct structural/functional ventricular chamber phenotypes may be classified by MRI, and that a uniform insult can result in multiple RV phenotypes.

Image segmentation applied to CT examination of lymphangioleiomyomatosis

The purpose of this study is to use modern image segmentation techniques to quantitate cyst area and number within a complete CT examination of the lungs. Lymphangioleiomyomatosis (LAM) was chosen because this disease produces many well defined thin- walled cysts of varying sizes throughout the lungs that provide a good test for 2D image segmentation techniques, which are used to separate LAM cysts from the normal lung tissue. Quantitative measures of the lung, such as cyst area versus frequency, are then automatically extracted. Three women with LAM were examined using CT slices obtained at 20 mm intervals, with 1 to 1.5 mm collimation, and a pixel size of 0.4 - 0.5 mm. Our segmentation algorithm operates in several stages. First, masks for each lung are automatically generated, thus allowing only lung pixels to be considered for the cyst segmentation. Next, we threshold the data under the masks at a level of -900 Hounsfield units. The threshold segments LAM cysts from normal lung tissue and other structures, such as pulmonary veins and arteries. In order to determine the size of individual cysts, we grow all regions having brightness values lower than the threshold within the masked regions. These regions, which correspond to cysts, are then sorted by size, and a cyst histogram for each patient is computed.

The Impact of Sources of Variability on Parametric Response Mapping of Lung CT Scans.

Parametric response mapping (PRM) of inspiration and expiration computed tomography (CT) images improves the radiological phenotyping of chronic obstructive pulmonary disease (COPD). PRM classifies individual voxels of lung parenchyma as normal, emphysematous, or nonemphysematous air trapping. In this study, bias and noise characteristics of the PRM methodology to CT and clinical procedures were evaluated to determine best practices for this quantitative technique. Twenty patients of varying COPD status with paired volumetric inspiration and expiration CT scans of the lungs were identified from the baseline COPDGene cohort. The impact of CT scanner manufacturer and reconstruction kernels were evaluated as potential sources of variability in PRM measurements along with simulations to quantify the impact of inspiration/expiration lung volume levels, misregistration, and image spacing on PRM measurements. Negligible variation in PRM metrics was observed when CT scanner type and reconstruction were consistent and inspiration/expiration lung volume levels were near target volumes. CT scanner Hounsfield unit drift occurred but remained difficult to ameliorate. Increasing levels of image misregistration and CT slice spacing were found to have a minor effect on PRM measurements. PRM-derived values were found to be most sensitive to lung volume levels and mismatched reconstruction kernels. As with other quantitative imaging techniques, reliable PRM measurements are attainable when consistent clinical and CT protocols are implemented.

Bleomycin-induced lung disease in an animal model: correlation between computed tomography-determined abnormalities and lung function.

RATIONALE AND OBJECTIVESnThe authors evaluated whether specific types of computed tomographic (CT) abnormalities could be correlated with physiologic impairment in animals with bleomycin-induced lung injury.nnnMETHODSnLung injury was induced in 20 rabbits by means of intratracheal administration of bleomycin (3 U per kilogram of body weight), followed by 100% oxygen for 2 minutes. The animals underwent high-resolution CT scanning at 14 (n = 4), 28 (n = 6), or 56 (n = 10) days after injury. CT morphometry was used to determine the extent of abnormal lung. Physiologic evaluation was performed before injury and before scanning.nnnRESULTSnThe overall extent of abnormal lung and of parenchymal opacification on CT scans did not correlate with any physiologic variable. The extent of interstitial thickening correlated significantly with total lung capacity (r = -.783, P = .0005), airway pressure at maximal lung volume (r = .836, P = .0001), and alveolar-arterial oxygen gradient (r = .613, P = .004).nnnCONCLUSIONnCT findings of interstitial thickening are associated with impaired gas exchange and lung stiffness in rabbits.

The idiopathic interstitial pneumonias: an update and review.

Idiopathic interstitial pneumonias (IIPs) are a group of disorders with distinct histologic and radiologic appearances and no identifiable cause. The IIPs comprise 8 currently recognized entities. Each of these entities demonstrates a prototypical imaging and histologic pattern, although in practice the imaging patterns may overlap, and some interstitial pneumonias are not classifiable. To be considered an IIP, the disease must be idiopathic; however, each pattern may be secondary to a recognizable cause, most notably collagen vascular disease, hypersensitivity pneumonitis, or drug reactions. The diagnosis of IIP requires the correlation of clinical, imaging, and pathologic features.

Convolutional Neural Network Based COPD and Emphysema Classifications Are Predictive of Lung Cancer Diagnosis.

Lung cancer is a leading cause of mortality and morbidity for patients suffering from Chronic Obstructive Pulmonary Disease (COPD). Both the presence of visually assessed emphysema on CT scans and abnormal pulmonary function tests are associated with the development of lung cancer. Based on recent results showing that convolutional neural networks (CNNs) applied to CT scans can predict spirometrically-defined COPD ((frac{FEV_{1}}{FVC}<0.7)), we hypothesized that CNN-based classification of COPD and emphysema is predictive of lung cancer development in the National Lung Cancer Screening (NLST) cohort. We trained spirometric COPD and visual emphysema CNN classifiers using data from the COPDGene study. The classifiers were then used to generate COPD and emphysema scores ((CS_{CNN}) and (ES_{CNN}), respectively) on 7347 CT scans from the NLST study. Cox proportional hazards regression was used to model the effects of (CS_{CNN}), (ES_{CNN}), age, body mass index, education, gender, smoking pack-years, and years since smoking cessation on lung cancer diagnosis. It was found that, individually, both (CS_{CNN}) and (ES_{CNN}) were statistically significant predictors (p < 0.000 and p < 0.000, respectively) of lung cancer diagnosis hazard.