Core to rind distribution of severe emphysema predicts outcome of lung volume reduction surgery.

Computed tomography (CT) has shown that emphysema is more extensive in the inner (core) region than in the outer (rind) region of the lung. It has been suggested that the concentration of emphysematous lesions in the outer rind leads to a better outcome following lung volume reduction surgery (LVRS) because these regions tend to be more surgically accessible. The present study used a recently described, computer-based CT scan analysis to quantify severe emphysema (lung inflation > 10.2 ml gas/g tissue), mild/moderate emphysema (lung inflation = 10.2 to 6.0 ml gas/g tissue), and normal lung tissue (lung inflation < 6.0 ml gas/g tissue) present in the core and rind of the lung in 21 LVRS patients. The results show that the quantification of severe emphysema independently predicts change in maximal exercise response and FEV(1). We conclude that a greater extent of severe emphysema in the rind of the upper lung predicts greater benefit from LVRS because it identifies the lesions most accessible to removal by LVRS.

Visualization-based analysis of multiparameter models using environment for N-dimensional model analysis.

We present a methodology, based on N-dimensional computer visualization, for analyzing multiparameter models. This approach originally consisted of three steps: behavior analysis, sensitivity analysis, identifiability analysis. We have now developed a new way of calculating sensitivity based on the statistical measure of the coefficient of variation. Furthermore, we extended the methodology through the addition of an extra step, visual regression. Visual regression allows the user to visualize the process of actual parameter identification and presents a combined, empirical view of the first three steps in a single image. Next we applied this methodology to pulmonary capillary-transport models. Finally, we implemented the model analysis process as a stand-alone program. EN-DIMAN, the resulting software, allows researchers to carry out model analysis in a graphical user interface (GUI)-based environment.

Graphical lung analysis and simulation environment.

The study of transport across the pulmonary vasculature is an important aspect of the study of the lung. Models have often been used in conjunction with experimental work to further the information which can be obtained from experimental work alone. GLANSE was developed as an environment to carry out such analysis on microcomputers. The main model employed is a three region, homogeneous model which includes provisions for tracer diffusion in the extravascular region, hydrophilic and lipophyilic tracers as well as physiological parameters such as blood flow. Several heterogeneous models based on simplified versions of the three region model as well as two models which are not related to the three region model are also included. Computationally efficient routines for model simulations are used so as to enable their execution on microcomputers with large data sets. In addition, several methods for models analysis, such as parameter sensitivity and curve-fitting, as well as statistical analysis of results are also included. GLANSE has been tested and has been in use for several years for routine analysis of experimental data.

A visualization-based analysis method for multiparameter models of capillary tissue-exchange.

In order to successfully use a model for parameter identification, it must be carefully analyzed. Current analysis methods, however, are ad hoc and provide only partial information. We extended these methods through the application of stacked dimensions, a scientific visualization method. The end result of our extensions are multi-dimensional parametric model-images. These images depict a model as a function of all its parameters in a single graphic. We applied parametric model-images to model verification (behavioral analysis), sensitivity analysis, and identifiability analysis. We applied our methodology to the evaluation of pulmonary vascular capillary-transport models. Results have shown that the visualization-based method provides a more complete view of a model's behavior and its other characteristics. Furthermore, our method has also proven to be more computationally efficient than the traditional approaches.

Parameter identification in coronary pressure flow models: a graphical approach.

The confident identification of parameters is important in the practical application of physiological models. However, the task of parameter identification is often complicated by interactions among parameters and by the fact that the sensitivity of the model to changes in a given parameter is generally a function of all the other parameters. Here we illustrate a graphical approach to parameter identification that allows the modeler to visualize the behavior of the model, the sensitivity functions, and certain functions characteristic of parameter interdependence. The visual display can be generated over any desired portion of parameter space. The technique is applied to a simple, four-parameter, myocardial pump model of the coronary circulation. The results indicate that over specified ranges of parameters, it is possible to distinguish among the four parameters of the model: the ratio of proximal-to-distal resistance, alpha; the overall resistance of the vascular bed, R; the compliance of the vascular bed, C; and a parameter, kappa, relating tissue pressure to left ventricular pressure. It was found that in order to identify all parameters uniquely, it was necessary to regress upon both coronary inflow and outflow.

Deduction of pulmonary microvascular hematocrit from indicator dilution curves.

We have developed a new model describing the relationship between plasma and red cell tracers flowing through the lung. The model is the result of an analysis of the transport of radiolabeled plasma albumin between two flowing phases and shows that differences between red cell and plasma tracer curves are related to microvascular hematocrit. The model was tested in an isolated, blood-perfused dog lung preparation in which we injected 51Cr-labeled red cells and 125I-labeled plasma albumin into the pulmonary artery. From the tracer concentration-time curves at the venous outflow, we calculated hr, the ratio of microvascular hematocrit to large-vessel hematocrit. In 18 baseline experiments, hr = 0.92 +/- 0.01 (mn +/- sem) at a blood flow rate of 10.7 +/- 0.3 ml s-1. We determined the effects of (a) glass bead embolization, (b) alloxan, and (c) lobe ligation on hr. Embolization attenuated the separation between plasma and red cells (increased hr), probably as a consequence of passive vasodilation. Alloxan enhanced separation of plasma and red cells (decreased hr), possibly as a result of arteriolar vasoconstriction. Ligation of a fraction of the perfused tissue at constant flow did not cause significant change in hr in the remaining perfused tissue. The model assumes that large-vessel transit times are uniform and that all dispersion occurs in the microvasculature. A theoretical analysis apportioning dispersion between large and small vessels disclosed that the error associated with these assumptions is likely to be less than 15% of the measured hr. We conclude from this study that the microvascular hematocrit model describes experimental plasma and red cell curves. The results imply that hr can be readily deduced from tagged red cells and plasma and can be accounted for in calculating permeability-surface area in diffusing tracer experiments.