RESUMO
Dynamic contrast-enhanced (DCE) imaging using MRI or CT is emerging as a promising tool for diagnostic imaging of cerebral disorders and the monitoring of tumor response to treatment. In this study, we present a robust and efficient deconvolution method based on a linearized model of the impulse residue function, which allows for the mapping of functional cerebral parameters such as cerebral blood flow, volume, mean transit time, and permeability. Monte Carlo simulation studies were performed to study the accuracy and stability of the proposed method, before applying it to clinical study cases of patients with cerebral tumors imaged using DCE CT. Functional parameter maps generated using the proposed method revealed the locations of the cerebral tumors and were found to be of sufficiently good clarity for marked regional differences in tissue vascularity and permeability to be assessed. In particular, tumor visualization and delineation were found to be better on the parameter maps that were indicative of the breakdown of the blood-brain barrier.
Assuntos
Neoplasias Encefálicas/irrigação sanguínea , Encéfalo/irrigação sanguínea , Lobo Frontal/irrigação sanguínea , Aumento da Imagem , Processamento de Imagem Assistida por Computador , Imageamento por Ressonância Magnética , Tomografia Computadorizada por Raios X , Adenoma/irrigação sanguínea , Adenoma/diagnóstico por imagem , Idoso , Artefatos , Velocidade do Fluxo Sanguíneo/fisiologia , Volume Sanguíneo/fisiologia , Neoplasias Encefálicas/diagnóstico por imagem , Meios de Contraste/administração & dosagem , Meios de Contraste/farmacocinética , Eficiência , Feminino , Lobo Frontal/diagnóstico por imagem , Humanos , Iohexol/farmacocinética , Modelos Lineares , Masculino , Neoplasias Meníngeas/irrigação sanguínea , Neoplasias Meníngeas/diagnóstico por imagem , Meningioma/irrigação sanguínea , Meningioma/diagnóstico por imagem , Microcirculação/fisiopatologia , Pessoa de Meia-Idade , Método de Monte Carlo , Neoplasias Hipofisárias/irrigação sanguínea , Neoplasias Hipofisárias/diagnóstico por imagem , Fluxo Sanguíneo Regional/fisiologia , Reprodutibilidade dos Testes , SoftwareRESUMO
Quantitative estimates of physiological parameters associated with cerebral blood flow can be derived from the analysis of dynamic contrast-enhanced (DCE) images, using an appropriate model of the underlying tissue impulse residue function. The theoretical formulation of a distributed parameter model of tissue microcirculation, which accounts for the effects of capillary permeability and transit time distribution, is presented here. This model considers a statistical distribution of capillary-tissue units, each described by a distributed parameter model that accounts for convective transport within the capillary and transcapillary axial diffusion. Monte Carlo simulations were performed to study the confidence of the parameter estimates, and the model was used to analyze DCE CT images of patient study cases with metastatic cerebral tumors. The tumors were found to yield significantly higher estimates than normal tissues for the parameters associated with the extravasation of tracer and for the standard deviation of capillary transit times. The proposed model can be used with DCE imaging to study the microcirculatory characteristics of cerebral tumors.
