Quantitative analysis of pulmonary emphysema: three-dimensional fractal analysis of single-photon emission computed tomography images obtained with a carbon particle radioaerosol

Nonclinical research areas of future importance for clinical therapies: Exploring the concepts of nonlinearity in dentistry

Linear system analysis has been dominating medical and dental research, and most of the research achievements in these fields have come from applying a reductionist view of nature. However, biologic systems are fundamentally nonlinear with highly composite dynamics made up of numerous interacting elements and feedback loops, therefore studying them as linear models may not result in an accurate representation of their true features. The authors reviewed and utilized some of the principles of chaos and nonlinearity and extended them to clinical dentistry, from cracked tooth and flare-up after root canal procedures to the outcome of clinical treatments. Utilization of the concepts of chaos and sensitive dependence on initial conditions, and the concepts of self-organization, stigmergy, and fractals may help us to understand some of the puzzles that have not been solved by conventional linear models. The goal of this paper is to present some areas within nonclinical research that we believe will have important roles in the development of future clinical examination methods and therapies.

Impact of a combination of quantitative indices representing uptake intensity, shape, and asymmetry in DAT SPECT using machine learning: comparison of different volume of interest settings

BACKGROUND

We sought to assess the machine learning-based combined diagnostic accuracy of three types of quantitative indices obtained using dopamine transporter single-photon emission computed tomography (DAT SPECT)-specific binding ratio (SBR), putamen-to-caudate ratio (PCR)/fractal dimension (FD), and asymmetry index (AI)-for parkinsonian syndrome (PS). We also aimed to compare the effect of two different types of volume of interest (VOI) settings from commercially available software packages DaTQUANT (Q) and DaTView (V) on diagnostic accuracy.

METHODS

Seventy-one patients with PS and 40 without PS (NPS) were enrolled. Using SPECT images obtained from these patients, three quantitative indices were calculated at two different VOI settings each. SBR-Q, PCR-Q, and AI-Q were derived using the VOI settings from DaTQUANT, whereas SBR-V, FD-V, and AI-V were derived using those from DaTView. We compared the diagnostic value of these six indices for PS. We incorporated a support vector machine (SVM) classifier for assessing the combined accuracy of the three indices (SVM-Q: combination of SBR-Q, PCR-Q, and AI-Q; SVM-V: combination of SBR-V, FD-V, and AI-V). A Mann-Whitney U test and receiver-operating characteristics (ROC) analysis were used for statistical analyses.

RESULTS

ROC analyses demonstrated that the areas under the curve (AUC) for SBR-Q, PCR-Q, AI-Q, SBR-V, FD-V, and AI-V were 0.978, 0.837, 0.802, 0.906, 0.972, and 0.829, respectively. On comparing the corresponding quantitative indices between the two types of VOI settings, SBR-Q performed better than SBR-V (p = 0.006), whereas FD-V performed better than PCR-Q (p = 0.0003). No significant difference was observed between AI-Q and AI-V (p = 0.56). The AUCs for SVM-Q and SVM-V were 0.988 and 0.994, respectively; the two different VOI settings displayed no significant differences in terms of diagnostic accuracy (p = 0.48).

CONCLUSION

The combination of the three indices obtained using the SVM classifier improved the diagnostic performance for PS; this performance did not differ based on the VOI settings and software used.

Quantitative evaluation of the tracer distribution in dopamine transporter SPECT for objective interpretation

PURPOSE

Quantification of the tracer distribution would add objectivity to the visual assessments of dopamine transporter (DAT) single photon emission computed tomography (SPECT) data. Our study aimed to evaluate the diagnostic utility of fractal dimension (FD) as a quantitative indicator of tracer distribution and compared with the conventional quantitative value: specific binding ratio (SBR). We also evaluated the utility of the combined index SBR/FD (SBR divided by FD).

MATERIALS AND METHODS

We conducted both clinical and phantom studies. In the clinical study, 150 patients including 110 patients with Parkinsonian syndrome (PS) and 40 without PS were enrolled. In the phantom study, we used a striatal phantom with the striatum chamber divided into two spaces, representing the caudate nucleus and putamen. The SBR, FD, and SBR/FD were calculated and compared between datasets for evaluating the diagnostic utility. Mann-Whitney test and receiver-operating characteristics (ROC) analysis were used for analysis.

RESULTS

ROC analysis revealed that the FD value had high diagnostic performance [the areas under the curve (AUC) = 0.943] and the combined use of SBR and FD (SBR/FD) delivered better results than the SBR alone (AUC, 0.964 vs 0.899; p < 0.001). The sensitivity, specificity, and accuracy, respectively, were 79.1, 85.0, and 80.7% with SBR, 84.5, 97.5, and 88.0% with FD, and 92.7, 87.5, and 91.3% with SBR/FD.

CONCLUSION

Our results confirmed that the FD value is a useful diagnostic index, which reflects the tracer distribution in DAT SPECT images. The combined use of SBR and FD was more useful than either used alone.

Three-dimensional fractal analysis of 99mTc-MAA SPECT images in chronic thromboembolic pulmonary hypertension for evaluation of response to balloon pulmonary angioplasty: association with pulmonary arterial pressure

OBJECTIVE

Balloon pulmonary angioplasty (BPA) is used for inoperable chronic thromboembolic pulmonary hypertension (CTEPH), but its effect cannot be evaluated noninvasively. We devised a noninvasive quantitative index of response to BPA using three-dimensional fractal analysis (3D-FA) of technetium-99m-macroaggregated albumin (Tc-MAA) single-photon emission computed tomography (SPECT).

PATIENTS AND METHODS

Forty CTEPH patients who underwent pulmonary perfusion scintigraphy and mean pulmonary arterial pressure (mPAP) measurement by right heart catheterization before and after BPA were studied. The total uptake volume (TUV) in bilateral lungs was determined from maximum intensity projection Tc-MAA SPECT images. Fractal dimension was assessed by 3D-FA. Parameters were compared before and after BPA, and between patients with post-BPA mPAP more than 30 mmHg and less than or equal to 30 mmHg. Receiver operating characteristic analysis was carried out.

RESULTS

BPA significantly improved TUV (595±204-885±214 ml, P<0.001) and reduced the laterality of uptake (238±147-135±131 ml, P<0.001). Patients with poor therapeutic response (post-BPA mPAP≥30 mmHg, n=16) showed a significantly smaller TUV increase (P=0.044) and a significantly greater post-BPA fractal dimension (P<0.001) than the low-mPAP group. Fractal dimension correlated with mPAP values before and after BPA (P=0.013 and 0.001, respectively). A post-BPA fractal dimension threshold of 2.4 distinguished between BPA success and failure with 75% sensitivity, 79% specificity, 78% accuracy, and area under the curve of 0.85.

CONCLUSION

3D-FA using Tc-MAA SPECT pulmonary perfusion scintigraphy enables a noninvasive evaluation of the response of CTEPH patients to BPA.

Peripheral ventilation heterogeneity determines the extent of bronchoconstriction in asthma

In asthma, bronchoconstriction causes topographically heterogeneous airway narrowing, as measured by three-dimensional ventilation imaging. Computation modeling suggests that peripheral airway dysfunction is a potential determinant of acute airway narrowing measured by imaging. We hypothesized that the development of low-ventilation regions measured topographically by three-dimensional imaging after bronchoconstriction is predicted by peripheral airway function. Fourteen asthmatic subjects underwent ventilation single-photon-emission computed tomography/computed tomography scan imaging before and after methacholine challenge. One-liter breaths of Technegas were inhaled from functional residual capacity in upright posture before supine scanning. The lung regions with the lowest ventilation (Vent_low) were calculated using a thresholding method and expressed as a percentage of total ventilation (Vent_total). Multiple-breath nitrogen washout was used to measure diffusion-dependent and convection-dependent ventilation heterogeneity (Sacin and Scond, respectively) and lung clearance index (LCI), before and after challenge. Forced expiratory volume in 1 s (FEV₁) was 87.6 ± 15.8% predicted, and seven subjects had airway hyperresponsiveness. Vent_low at baseline was unrelated to spirometry or multiple-breath nitrogen washout indices. Methacholine challenge decreased FEV₁ by 23 ± 5% of baseline while Vent_low increased from 21.5 ± 2.3%Vent_total to 26.3 ± 6.7%Vent_total (P = 0.03). The change in Vent_low was predicted by baseline Sacin (r_s  = 0.60, P = 0.03) and by LCI (r_s  = 0.70, P = 0.006) but not by Scond (r_s  = 0.30, P = 0.30). The development of low-ventilation lung units in three-dimensional ventilation imaging is predicted by ventilation heterogeneity in diffusion-dependent airways. This relationship suggests that acinar ventilation heterogeneity in asthma may be of mechanistic importance in terms of bronchoconstriction and airway narrowing.NEW & NOTEWORTHY Using ventilation SPECT/CT imaging in asthmatics, we show induced bronchoconstriction leads to the development of areas of low ventilation. Furthermore, the relative volume of the low-ventilation regions was predicted by ventilation heterogeneity in diffusion-dependent acinar airways. This suggests that the pattern of regional airway narrowing in asthma is determined by acinar airway function.

Pattern analysis accounts for heterogeneity observed in MRI studies of tumor angiogenesis

MRI is a method of choice for assessing anatomical structures or angiogenesis-related parameters noninvasively during tumor progression. Typically, tumor tissue displays a high degree of heterogeneity that can be evaluated using pattern analysis (PA), which comprises shape and texture analysis. This work aims at implementing PA methods to study angiogenesis in a murine tumor model and testing its sensitivity with regard to detecting changes elicited by administration of a drug. Twelve balb/c-nude mice were injected subcutaneously with 10(6) C51 cells (colon carcinoma). A first group (N = 6) of animals was treated with dimethyloxalylglycine, a drug known to stabilize hypoxia-inducible-factor-α, which among other functions, is involved in angiogenesis. The second group (N = 6) was treated with saline. MRI experiments assessing tumor blood volume and permeability-maps (K(trans) ) were performed immediately before and 6 days after drug treatment. Data have been analyzed using standard histogram analysis and PA. Standard histogram analysis did not reveal any difference between the two groups, neither before nor after the treatment. In contrast, PA revealed significant differences between drug and placebo treated mice in the texture of the TBV and K(trans) maps after drug treatment, but not with regard to tumors shapes. The results indicated that in view of the heterogeneity of tumor tissue, standard histogram analysis appears insensitive in picking-up differences in response to treatment, while PA appears to be particularly sensitive to changes in texture.

Computer-assisted detection of infectious lung diseases: a review

Respiratory tract infections are a leading cause of death and disability worldwide. Although radiology serves as a primary diagnostic method for assessing respiratory tract infections, visual analysis of chest radiographs and computed tomography (CT) scans is restricted by low specificity for causal infectious organisms and a limited capacity to assess severity and predict patient outcomes. These limitations suggest that computer-assisted detection (CAD) could make a valuable contribution to the management of respiratory tract infections by assisting in the early recognition of pulmonary parenchymal lesions, providing quantitative measures of disease severity and assessing the response to therapy. In this paper, we review the most common radiographic and CT features of respiratory tract infections, discuss the challenges of defining and measuring these disorders with CAD, and propose some strategies to address these challenges.

Lung volume reduction surgery for emphysema: What the radiologist needs to know

Imaging plays a pivotal role in the selection of patients for the surgical treatment of emphysema. In this article, the imaging features of emphysema are reviewed along with the surgical options for treatment. Particular emphasis is given to lung volume reduction surgery as this technique has gained wide acceptance within the thoracic surgical community in recent years. Radiologists need to have an understanding of which patients may be potentially suitable for this technique.

Fractal analysis of internal and peripheral textures of small peripheral bronchogenic carcinomas in thin-section computed tomography: comparison of bronchioloalveolar cell carcinomas with nonbronchioloalveolar cell carcinomas

PURPOSE

To analyze the internal and peripheral textures of small peripheral bronchogenic carcinomas (<2 cm) in thin-section computed tomography (HRCT) images with fractal analysis.

METHOD

Thin-section computed tomography images from 70 patients with bronchogenic carcinomas (61 adenocarcinomas and 9 squamous cell carcinomas) were used. Regions of interest (ROIs) with a matrix size of 32 x 32 (0.326 mm per pixel) were selected manually on the lung-nodule interfaces and within the nodules on HRCT images. Three-dimensional density surfaces based on CT values of ROIs were characterized by fractal dimensions (FDs).

RESULTS

When all the bronchogenic carcinomas were divided into bronchioloalveolar cell carcinomas (BACs) and other bronchogenic carcinomas (nonBACs), there were significant differences between BACs and nonBACs in the FDs obtained from the internal textures (mean: 2.38 +/- 0.05 versus 2.19 +/- 0.05; P< 0.0001) and in the FDs obtained from the peripheral textures (mean: 2.16 +/- 0.01 versus 2.06 +/- 0.01; P< 0.0001).

CONCLUSION

The textures of BACs that reveal ground-glass opacities are more complicated than those of nonBACs. The FDs can differentiate between small localized BACs, which have a good prognosis, and nonBACs, which have a poor prognosis. Fractal analysis is promising for characterization of small peripheral pulmonary bronchogenic carcinomas based on radiographic features of HRCT images.

Radiologic assessment of emphysema for lung volume reduction surgery

Radiologic imaging is vital to determining whether lung volume reduction surgery (LVRS) is a potential therapeutic option for patients with severe chronic obstructive pulmonary disease. The importance of imaging in patient selection for LVRS has been emphasized by numerous studies relating preoperative imaging features to postoperative outcomes. This article reviews the radiologic features assessed in the LVRS evaluation, summarizes the relationships between preoperative radiologic features and postoperative outcomes, and addresses the limitations of imaging in guiding patient selection.

Measurement of heterogeneous distribution on Technegas SPECT images by three-dimensional fractal analysis

This review article describes a method for quantifying heterogeneous distribution on Technegas (99mTc-carbon particle radioaerosol) SPECT images by three-dimensional fractal analysis (3D-FA). Technegas SPECT was performed to quantify the severity of pulmonary emphysema. We delineated the SPECT images by using five cut-offs (15, 20, 25, 30 and 35% of the maximal voxel radioactivity), and measured the total number of voxels in the areas surrounded by the contours obtained with each cut-off level. We calculated fractal dimensions from the relationship between the total number of voxels and the cut-off levels transformed into natural logarithms. The fractal dimension derived from 3D-FA is the relative and objective measurement, which can assess the heterogeneous distribution on Technegas SPECT images. The fractal dimension strongly correlate pulmonary function in patients with emphysema and well documented the overall and regional severity of emphysema.

Pulmonary perfusion patterns and pulmonary arterial pressure

PURPOSE

To use artificial intelligence methods to determine whether quantitative parameters describing the perfusion image can be synthesized to make a reasonable estimate of the pulmonary arterial (PA) pressure measured at angiography.

MATERIALS AND METHODS

Radionuclide perfusion images were obtained in 120 patients with normal chest radiographs who also underwent angiographic PA pressure measurement within 3 days of the radionuclide study. An artificial neural network (ANN) was constructed from several image parameters describing statistical and boundary characteristics of the perfusion images. With use of a leave-one-out cross-validation technique, this method was used to predict the PA systolic pressure in cases on which the ANN had not been trained. A Pearson correlation coefficient was determined between the predicted and measured PA systolic pressures.

RESULTS

ANN predictions correlated with measured pulmonary systolic pressures (r = 0.846, P <.001). The accuracy of the predictions was not influenced by the presence of pulmonary embolism. None of the 51 patients with predicted PA pressures of less than 29 mm Hg had pulmonary hypertension at angiography. All 13 patients with predicted PA pressures greater than 48 mm Hg had pulmonary hypertension at angiography.

CONCLUSION

Meaningful information regarding PA pressure can be derived from noninvasive radionuclide perfusion scanning. The use of image analysis in concert with artificial intelligence methods helps to reveal physiologic information not readily apparent at visual image inspection.

Fractal analysis of small peripheral pulmonary nodules in thin-section CT: evaluation of the lung-nodule interfaces

OBJECTIVES

To analyze the lung-nodule interfaces on small peripheral pulmonary nodules (<2 cm) in thin-section CT (HRCT) images with fractal analysis.

METHODS

Thin-section CT images from 70 patients with bronchogenic carcinomas (61 adenocarcinomas and 9 squamous cell carcinomas) and 47 patients with benign pulmonary nodules (23 hamartomas, 13 organizing pneumonias, and 11 tuberculomas) were used. For calculation of fractal dimensions (FDs), the authors used a box-counting method for binary- and gray-scale images of nodules. FD(two-dimensional [2D]) was an FD obtained from the binary image, and FD(three-dimensional [3D]) was an FD obtained from the gray-scale image.

RESULTS

The FD(2D)s of hamartomas were smaller than those of other nodules ( < 0.05). The FD(3D)s obtained from the gray-scale images of organizing pneumonias and tuberculomas were greater than those of bronchogenic carcinomas ( < 0.0001) and hamartomas ( < 0.0001). In bronchogenic carcinomas, FD(3D)s of adenocarcinomas were greater than those of squamous cell carcinomas ( < 0.05).

CONCLUSIONS

Fractal dimensions reflect the characteristics of the lung-nodule interfaces of small peripheral pulmonary nodules. The FD(2D)s revealed the irregularities of the contours. On the other hand, FD(3D)s revealed the complexities of the heterogeneous textures. With use of FD(2D) and FD(3D), it may be possible to distinguish bronchogenic carcinomas from benign pulmonary nodules. Moreover, FD(3D) may make it possible to distinguish between adenocarcinomas and squamous cell carcinomas.

Relationship between regional severity of emphysema and coronary heart disease

We analyzed the relationship between regional severity of emphysema, which was evaluated by three-dimensional fractal analysis (3D-FA) of Technegas SPECT images, and coronary heart disease (CHD). For 22 patients with emphysema who underwent Technegas SPECT, we followed up CHD events. The follow-up period was 5.4+/-0.5 (mean +/- SD) years. We defined the upper-lung fractal dimension (U-FD) and lower-lung fractal dimension (L-FD) obtained with 3D-FA of Technegas SPECT images as the regional severity of emphysema. FD became greater with the progression of emphysematous change. During the follow-up period, CHD events occurred in 6 (27%) of the 22 patients. The ratio of U-FD to L-FD for patients with CHD events (0.87+/-0.22) was significantly smaller than for patients without CHD events (1.52+/-0.38) (p = 0.0015). These findings suggest that severer emphysema in the lower lung indicates a higher risk of CHD than that in the upper lung.

Fractal analysis of nuclear medicine images for the diagnosis of pulmonary emphysema: interpretations, implications, and limitations

OBJECTIVE

The purpose of this study was to investigate, on images obtained in nuclear medicine examinations, the physical meanings and consequent implications of fractal analysis developed in a recent study that was reported to be effective in quantifying the heterogeneous distribution of carbon particle radioaerosol in the lungs.

MATERIALS AND METHODS

Fractal dimensions were computed for 108 sets of radionuclide imaging data from 28 patients according to the methods in a previous report, and were then correlated with the ratio of tissue areas segmented at two thresholds (15% and 35% of maximal radioactivity).

RESULTS

Fractal dimension was found to linearly correlate with the ratio natural logarithm of tissue areas segmented at two different threshold levels (n = 108, r = 0.999), with regression slope accurately predicted (error = 0.06%). Bland-Altman analysis showed that fractal dimensions ranging from 0.2 to 1.9 can be explained by this area ratio with disagreement of only 5.13% at two standard deviations; thus, fractal dimension seems to be an over-simplified parameter unrelated to spatial heterogeneity of radioaerosol distribution.

CONCLUSION

The analysis of this study suggested that the fractal dimension defined in a previous report was limited to the indication of the percentage area of low-radioactivity regions with respect to total tissue area in the image. Because the fractal dimension partially reflects, but is not specific to, a certain degree of focal spots of low radioactivity, we suggest using fractal analysis in clinical practice only with careful control and thorough understanding of the physical meanings.

Computed tomographic diagnosis of chronic obstructive pulmonary disease

Recently, two-dimensional high-resolution computed tomography (2D-HRCT) and volumetric high-resolution computed tomography (VHRCT) have allowed great progress to be made in estimating small regions of hypoattenuation in the lung. VHRCT allows minimum-intensity projection (MINIP) imaging, which seems to be much more suitable than conventional 2D-HRCT for detecting subtle hypoattenuating regions, although the reliability of MINIP imaging data has not been conclusively shown. Three-dimensional computed tomography (3D-CT) images reconstructed from helical computed tomography (CT) acquisition over the whole lung field have been increasingly used to determine the absolute volume occupied by emphysematous changes. However, depending on the threshold CT values used for image reconstruction, whole-lung 3D-CT images may be distorted by considerable differences between the x-ray beam thickness and the effective section thickness, resulting in erroneous quantitation of emphysematous lesions by helical 3D-CT.