A Primer in Compartment Modeling



A Primer in Compartment Modeling


Cyrill Burger

Alfred Buck





Introduction

Imaging in nuclear medicine is based on the application of radioactively labeled compounds (tracers) that are distributed throughout the body by the vascular system and take part in physiologic processes according to the properties of the tracer molecule. Each tracer is designed to assess a specific property of tissue (e.g., perfusion, metabolic turnover, or the concentration of a specific receptor). To obtain optimal images, a protocol defines how the tracer is administered and at what timepoints single images or a sequence of images is acquired. The information contained in the images represents the average tracer concentration during the image acquisition, and in most cases it cannot be directly transformed into a quantitative parameter of the target process. For instance, “perfusion-weighted” images cannot be translated directly into images representing tissue perfusion in milliliters per minute per 100 mL of tissue.
Although functionally weighted images bear valuable diagnostic information, some situations require absolute quantification, such as therapy monitoring and group comparisons. Hence there is an interest in using nuclear medicine techniques for the quantitative assessment of tissue parameters not only for the scientific evaluation of tracer characteristics but also for clinical purposes.






Figure 13.1 PET acquisition setup for quantitative imaging. A series of dynamic images is acquired to observe the tracer distribution processes. Additionally, the tracer concentration in plasma is assessed by withdrawing blood samples throughout the scanning period.

The transformation of the measured data into absolute metabolic parameters is based on physiologic models, often compartment models. Hereby the relevant transport and uptake parameters are mathematically modeled and a prescription derived that predicts the tracer concentration in tissue given a measured plasma concentration. Quantification of the measurement then consists of solving the following inverse problem: Which model parameters predict the measured tissue concentration most closely? It must be possible to find a model that adequately describes the processes with a small number of parameters. Otherwise, with too many parameters in the model, there is a high chance that different parameter combinations will result in approximately the same predicted tissue response, and an optimization algorithm will then have difficulty finding a definite optimal solution. Whether such a model—one that is considerably simplified with respect to the underlying physiology—still provides valuable information is a question that must be answered for each tracer individually.

Compartment modeling has the following requirements:



  • A tracer with favorable properties regarding distribution and uptake dynamics;


  • Sampling of the tracer concentration in blood plasma (input curve) from the time of tracer administration until the end of image acquisition;


  • Observation of the dynamic tracer distribution in tissue by a series of image acquisitions (Fig. 13.1); and


  • Appropriate correction of the image information (e.g., attenuation correction) and conversion into absolute tracer concentration in becquerels per milliliter (as a result of which the time course of the tracer activity concentration [TAC, tissue time–activity curve] is available in each image pixel).


Compartment Models

To describe a physiologic system, it is often helpful to decompose it into a number of interacting subsystems, called compartments. Each compartment represents a homogeneous distribution of material and may exchange material with other compartments. Note that compartments
should be understood not as a physical volume but rather as a mass of well-mixed material that behaves uniformly. Examples of frequently used compartments include these:



  • Tracer in the arterial plasma that can be extracted into the tissue;


  • Free tracer in tissue that can be bound or that may diffuse back into the blood;


  • Tracer in tissue that has been nonspecifically bound to other than the targeted cell components; and


  • Tracer in tissue that has been specifically bound, for instance, at the target receptor site.

Compartment models are visualized by diagrams in which the rectangles symbolize compartments and the arrows represent material exchange, as illustrated in Figure 13.2. Although the mechanisms of material transport may be diverse, models that can be reasonably analyzed with standard mathematical methods assume first-order processes. As a consequence, the change of tracer concentration in one of the compartments is a linear function of the concentrations in all other compartments. Put into mathematical terms,


It is assumed that the observed system is not disturbed by the administration of the tracer. Because of the tiny amounts of tracer material applied in nuclear medicine investigations, this assumption is highly justified in most cases.






Figure 13.2 Three different configurations of compartment models. A: One-tissue compartment model. B: Two-tissue compartment model. C: Three-tissue compartment model. The Ci indicates the tracer concentrations in the compartments, and the ki the transfer coefficients of the material exchange between them.

To illustrate the properties of compartment models, let us consider a tracer that is not freely diffusible and is injected intravenously as a bolus. When the tracer arrives at the heart, it is well mixed with blood and distributed by the arterial circulation. It finally arrives at the capillary bed, where exchange with the tissue can take place. Nonextracted tracer is transported back to the heart, from which a new circulation starts. This highly simplified “physiologic” model can be translated into a one-tissue compartment model, as illustrated in Figure 13.2A. It is assumed that, because of mixing, the arterial tracer concentration is the same throughout the body, so it can be measured in a peripheral artery. The tracer concentration in tissue increases by the extraction of tracer from the blood plasma, and as this is described by a first-order process, the transfer of material is proportional to the concentration in plasma CA(t). Conversely, it is reduced by a backward transfer, which is proportional to the concentration in tissue C1(t). Both processes compete, so the change over time of the net tracer concentration in tissue (dC1(t)/dt) can be expressed by the following differential equation:


The two transfer coefficients, K1 and k2, have a somewhat different meaning, indicated by using capital and small letters. K1 includes a perfusion-dependent component and has units of milliliter per minute per milliliter of tissue, whereas k2 [min-1] indicates the fraction of mass transferred per unit of time. For example, if k2 equals 0.1[min-1], the material leaves the compartment at a rate of 10% per minute. The plasma concentration CA(t) is not modeled but is considered a measured quantity and appears as the input curve driving the system.

By solving differential equation (2), the following equation for the concentration of tracer in tissue can be derived:

Only gold members can continue reading. Log In or Register to continue

Stay updated, free articles. Join our Telegram channel

Jul 27, 2016 | Posted by in GENERAL RADIOLOGY | Comments Off on A Primer in Compartment Modeling

Full access? Get Clinical Tree

Get Clinical Tree app for offline access