Enhanced parameter estimation from noisy PET data: Part I--methodology.

RATIONALE AND OBJECTIVES: The reliability of positron emission tomographic (PET) images depends on the number of annihilation events that are detected. Short image durations are required to capture rapid tracer dynamics, and the resultant images are noisy. Consequently, direct parameter estimation from time-activity curves at high resolution often is unreliable. If adjacent voxels are combined into larger regions of interest the reliability of parameter estimation may be improved, but at the expense of decreased spatial resolution. In this report, a method is presented that provides an alternative to degrading image resolution. MATERIALS AND METHODS: Following the approach of Kimura et al, voxels are grouped not by spatial proximity, but by the similarity of their kinetics. Parameter estimation is performed on these groups, and derived parameters are assigned to all members of the group. Spatial information thus is preserved, but at the expense of parametric discretization. An improvement to the method of Kimura et al is described, in which data are grouped using principal components derived from artificial data. RESULTS: The application of the method is demonstrated by analysis of PET images of human lungs obtained by the nitrogen-13 infusion-washout technique. In a comparison of the accuracy of parameter estimates, the enhanced method is shown to outperform the original method at all noise levels, with the difference increasing as the amount of noise increases. The robustness of this parameter estimation method in the presence of noise is described in part II of this report in this issue of Academic Radiology. CONCLUSION: A method is described that provides demonstrably robust parameter estimates from noisy PET data, while not compromising image resolution.

Enhanced parameter estimation from noisy PET data: Part II--evaluation.

RATIONALE AND OBJECTIVES: Positron emission tomography (PET) is a minimally invasive imaging modality that provides three-dimensional distribution data for a radioactive tracer concentration within the body. Local functional parameters are estimated from these images by fitting tracer kinetic data with mathematical models. However, in some applications, the reliability of parameter estimates may be hindered by the presence of noise. In the accompanying report in this issue of Academic Radiology, a novel method using principal component analysis (PCA) was presented and used for deriving parametric images of lung function from imaged tracer kinetics of intravenously injected nitrogen 13 (13NN) in saline solution. The PCA method averages 13NN concentrations from groups of voxels (volume elements) selected for their similarity in kinetics, rather than their spatial proximity. The goal of this study is to conduct a Monte Carlo simulation to evaluate the robustness to noise of parameters derived by means of the PCA method. MATERIALS AND METHODS: This evaluation involved: (1) generating "noise-free" synthetic PET images from experimental PET data, (2) adding noise to these images, (3) applying the PCA method to yield parametric images, and (4) comparing these parametric images with original noise-free images. RESULTS: Local parameters recovered by using the PCA method deviated from noise-free parameters on average by less than 1% for up to 32-fold of expected noise levels. These deviations were much less than those (>10%) recovered by using a direct curve-fitting method. CONCLUSION: The novel PCA approach provides robust parametric lung functional images while preserving the spatial resolution of the original images.

Self-organized patchiness in asthma as a prelude to catastrophic shifts.

Asthma is a common disease affecting an increasing number of children throughout the world. In asthma, pulmonary airways narrow in response to contraction of surrounding smooth muscle. The precise nature of functional changes during an acute asthma attack is unclear. The tree structure of the pulmonary airways has been linked to complex behaviour in sudden airway narrowing and avalanche-like reopening. Here we present experimental evidence that bronchoconstriction leads to patchiness in lung ventilation, as well as a computational model that provides interpretation of the experimental data. Using positron emission tomography, we observe that bronchoconstricted asthmatics develop regions of poorly ventilated lung. Using the computational model we show that, even for uniform smooth muscle activation of a symmetric bronchial tree, the presence of minimal heterogeneity breaks the symmetry and leads to large clusters of poorly ventilated lung units. These clusters are generated by interaction of short- and long-range feedback mechanisms, which lead to catastrophic shifts similar to those linked to self-organized patchiness in nature. This work might have implications for the treatment of asthma, and might provide a model for studying diseases of other distributed organs.

Quantification of regional ventilation-perfusion ratios with PET.

UNLABELLED: The topographic matching of alveolar ventilation (V(A)) and perfusion (Q) is the main determinant of gas exchange efficiency of the lung. However, no pulmonary functional imaging technique has been shown to predict whole-lung gas exchange in health and disease. This study aims to present a PET-based method to estimate regional alveolar ventilation-to-perfusion ratios (V(A)/Q) predictive of arterial blood gases. METHODS: The method is based on the regional tracer kinetics of (13)N-nitrogen ((13)NN) after an intravenous bolus injection during a breath-hold period and subsequent washout from the lungs with resumption of breathing. The method takes into account the presence of inter- and intraregional nonuniformities at length scales smaller than the imaging spatial resolution. An algorithm used regional tracer washout to classify regional V(A)Q/ uniformity. Intraregional V(A)/Q mismatch in nonuniform regions was described with a 2-compartment model. Regional V(A)/Q estimates were combined into a whole-lung distribution of V(A)/Q ratios and were used to compute global arterial blood gases. The method was applied to 3-dimensional PET data from anesthetized and mechanically ventilated sheep before and after methacholine bronchoconstriction (n = 3) and pulmonary embolism (n = 3) and after saline lung lavage (n = 3). RESULTS: PET images revealed regional changes in ventilation and perfusion consistent with the different disease models. Quantification of the images using PET-derived V(A)Q/ distributions showed unimodal and narrow distributions in control conditions that became wider and unimodal after pulmonary embolism and saline lung lavage and bimodal after bronchoconstriction. Images of regional gas exchange allowed for visualization of regional gas exchange. Arterial blood gases estimated from the PET-based V(A)/Q distributions closely agreed with measured values (partial pressure of oxygen, arterial [PaO(2)]: r(2) = 0.97, P < 0.001; partial pressure of carbon dioxide, arterial [PaCO(2)]: r(2) = 0.96, P < 0.001). CONCLUSION: Tracer kinetics analysis of PET images after an intravenous injection of (13)NN provides a quantitative assessment of regional V(A)/Q heterogeneity including that corresponding to length scales smaller than the spatial resolution of the imaging method. Quantification of V(A)/Q mismatch obtained with the presented technique is directly related to severity of gas exchange impairment as determined by arterial blood gases.

Changes in regional ventilation after autologous blood clot pulmonary embolism.

BACKGROUND: Previous studies have suggested that pulmonary embolism (PE) and pulmonary artery occlusion result in a shift in alveolar ventilation away from unperfused regions. This study aimed to directly assess changes in regional specific ventilation (sV(A)) due to autologous blood clot PE using positron emission tomography. METHODS: Pulmonary embolism was created in six anesthetized, paralyzed, and mechanically ventilated sheep by injecting cylindrical clots of autologous blood (7 mm in diameter and height). Clots were progressively infused into a central vein until a stable mean pulmonary artery pressure between 30 and 40 mmHg was achieved. A multislice positron emission tomography camera was used to image 15 contiguous, 6.5-mm-thick transverse cross-sections of the chest beginning just above the diaphragm. sV(A) from perfused regions (sV(A),(p)) was assessed as the ventilatory turnover rate of the tracer NN after central venous injection of NN-labeled saline. RESULTS: Pulmonary embolism obstructed flow to 64% of imaged areas. Before PE, (sV(A),(p))was equivalent in areas that would remain perfused and those that would become embolized after PE (0.021 +/- 0.007 0.021 +/- 0.006 s(-1); P = nonsignificant). After PE, sV(A),(p) of areas remaining perfused increased to 0.033 +/- 0.011 s (-1) (P < 0.005). This effect on regional sV(A),(p) could have been caused by active redistribution of sV(A),(p) or by a reduction in tracer concentration of perfused areas due to the dead space common to perfused and embolized regions. Model simulations indicated that the common dead-space effect could only explain a small part of the sV(A),(p) increase. CONCLUSIONS: An increase in sV(A),(p) of perfused regions occurs following PE with 7-mm autologous blood clots. This increase is most likely caused by a shift in ventilation away from embolized areas mediated by hypocapnic pneumoconstriction.