Hemodynamic segmentation of brain perfusion images with delay and dispersion effects using an expectation-maximization algorithm.

Automatic identification of various perfusion compartments from dynamic susceptibility contrast magnetic resonance brain images can assist in clinical diagnosis and treatment of cerebrovascular diseases. The principle of segmentation methods was based on the clustering of bolus transit-time profiles to discern areas of different tissues. However, the cerebrovascular diseases may result in a delayed and dispersed local perfusion and therefore alter the hemodynamic signal profiles. Assessing the accuracy of the segmentation technique under delayed/dispersed circumstance is critical to accurately evaluate the severity of the vascular disease. In this study, we improved the segmentation method of expectation-maximization algorithm by using the results of hierarchical clustering on whitened perfusion data as initial parameters for a mixture of multivariate Gaussians model. In addition, Monte Carlo simulations were conducted to evaluate the performance of proposed method under different levels of delay, dispersion, and noise of signal profiles in tissue segmentation. The proposed method was used to classify brain tissue types using perfusion data from five normal participants, a patient with unilateral stenosis of the internal carotid artery, and a patient with moyamoya disease. Our results showed that the normal, delayed or dispersed hemodynamics can be well differentiated for patients, and therefore the local arterial input function for impaired tissues can be recognized to minimize the error when estimating the cerebral blood flow. Furthermore, the tissue in the risk of infarct and the tissue with or without the complementary blood supply from the communicating arteries can be identified.

Blind source separation of hemodynamics from magnetic resonance perfusion brain images using independent factor analysis.

Perfusion magnetic resonance brain imaging induces temporal signal changes on brain tissues, manifesting distinct blood-supply patterns for the profound analysis of cerebral hemodynamics. We employed independent factor analysis to blindly separate such dynamic images into different maps, that is, artery, gray matter, white matter, vein and sinus, and choroid plexus, in conjunction with corresponding signal-time curves. The averaged signal-time curve on the segmented arterial area was further used to calculate the relative cerebral blood volume (rCBV), relative cerebral blood flow (rCBF), and mean transit time (MTT). The averaged ratios for rCBV, rCBF, and MTT between gray and white matters for normal subjects were congruent with those in the literature.

Segmentation of diffusion-weighted brain images using expectation maximization algorithm initialized by hierarchical clustering.

Tissue segmentation based on diffusion-weighted images (DWI) provides complementary information of tissue contrast to the structural MRI for facilitating the tissue segmentation. In the previous literatures, DWI-based brain tissue segmentation was carried out using the parametric images, such as fractional anisotropy (FA) and apparent diffusion coefficient (ADC). However, the information of directions of neural fibers was very limited in the parametric images. To fully utilize the directional information, we propose a novel method to perform tissue segmentation directly on the DWI raw image data. Specifically, a hierarchical clustering (HC) technique was first applied on the down-sampled data to initialize the model parameters for each tissue cluster followed by automatic segmentation using the expectation maximization (EM) algorithm. The whole brain DWI raw data of five normal subjects were analyzed. The results demonstrated that HC-EM is effective in multi-tissue classification on DWI raw data.

Brain MR perfusion image segmentation using independent component analysis and hierarchical clustering.

Extraction of various perfusion components from dynamic-susceptibility-contrast (DSC) MR brain images is critical for the analysis of brain perfusion. According to the variation of temporal signal on different brain tissues, one can segment whole brain area into distinct blood supply patterns which are vital for the profound analysis of cerebral hemodynamics. In this study, independent component analysis (ICA) is used to project the perfusion image data into independent components from which each elucidated tissue cluster can be automatically segment out by using the hierarchical clustering (HC). Five normal subjects and a case of internal carotid artery stenosis subjects were analyzed. The results demonstrated that ICA-HC is effective in multi-tissue hemodynamic classification which improves differentiation of pathological and physiological hemodynamics.

Classification of hemodynamics from dynamic-susceptibility-contrast magnetic resonance (DSC-MR) brain images using noiseless independent factor analysis.

Dynamic-susceptibility-contrast (DSC) magnetic resonance imaging records signal changes on images when the injected contrast-agent particles pass through a human brain. The temporal signal changes on different brain tissues manifest distinct blood-supply patterns which are vital for the profound analysis of cerebral hemodynamics. Under the assumption of the spatial independence among these patterns, noiseless independent factor analysis (IFA) was first applied to decompose the DSC-MR data into different independent-factor images with corresponding signal-time curves. A major tissue type, such as artery, gray matter, white matter, vein, sinus, and choroid plexus, etc., on each independent-factor image was further segmented out by an optimal threshold. Based on the averaged signal-time curve on the arterial area, the cerebral hemodynamic parameters, such as relative cerebral blood volume (rCBV), relative cerebral blood flow (rCBF), and mean transit time (MTT), were computed and their averaged ratios between gray matter and white matter for normal subjects were in good agreement with those in the literature. Data of a stenosis patient before and after treatment were analyzed and the result illustrates that this method is effective in extracting spatiotemporal blood-supply patterns which improves differentiation of pathological and non-pathological hemodynamics.

Classification of spatiotemporal hemodynamics from brain perfusion MR images using expectation-maximization estimation with finite mixture of multivariate gaussian distributions.

The ability to cluster different perfusion compartments in the brain is critical for analyzing brain perfusion. This study presents a method based on a mixture of multivariate Gaussians (MoMG) and the expectation-maximization (EM) algorithm to dissect various perfusion compartments from dynamic susceptibility contrast (DSC) MR images so that each compartment comprises pixels of similar signal-time curves. This EM-based method provides an objective way to 1) delineate an area to serve as the in-plane arterial input function (AIF) of the feeding artery for adjacent tissues to better quantify the relative cerebral blood volume (rCBV), relative cerebral blood flow (rCBF), and mean transit time (MTT); 2) demarcate regions with abnormal perfusion derangement to facilitate diagnosis; and 3) obtain parametric maps with supplementary information, such as temporal scenarios and recirculation of contrast agent. Results from normal subjects show that perfusion cascade manifests (in order of appearance) the arteries, gray matter (GM), white matter (WM), veins and sinuses, and choroid plexus mixed with cerebrospinal fluid (CSF). The averaged rCBV, rCBF, and MTT ratios between GM and WM are in good agreement with those in the literature. Results from a patient with cerebral arteriovenous malformation (CAVM) showed distinct spatiotemporal characteristics between perfusion patterns, which allowed differentiation between pathological and nonpathological areas.