Gaussian mixture models (GMMs) constitute a well-known type of probabilistic neural networks. One of their many successful applications is in image segmentation, where spatially constrained mixture models have been trained using the expectation-maximization (EM) framework. In this letter, we elaborate on this method and propose a new methodology for the M-step of the EM algorithm that is based on a novel constrained optimization formulation. Numerical experiments using simulated images illustrate the superior performance of our method in terms of the attained maximum value of the objective function and segmentation accuracy compared to previous implementations of this approach.

In this paper, we propose a new methodology for analysis of microarray images. First, a new gridding algorithm is proposed for determining the individual spots and their borders. Then, a Gaussian mixture model (GMM) approach is presented for the analysis of the individual spot images. The main advantages of the proposed methodology are modeling flexibility and adaptability to the data, which are well-known strengths of GMM. The maximum likelihood and maximum a posteriori approaches are used to estimate the GMM parameters via the expectation maximization algorithm. The proposed approach has the ability to detect and compensate for artifacts that might occur in microarray images. This is accomplished by a model-based criterion that selects the number of the mixture components. We present numerical experiments with artificial and real data where we compare the proposed approach with previous ones and existing software tools for microarray image analysis and demonstrate its advantages.

No abstract

Abstract-In this study, we present an advanced Bayesian framework for the analysis of functional magnetic resonance imaging (fMRI) data that simultaneously employs both spatial and sparse properties. The basic building block of our method is the general linear regression model that constitutes a well-known probabilistic approach. By treating regression coefficients as random variables, we can apply an enhanced Gibbs distribution function that captures spatial constrains and at the same time allows sparse representation of fMRI time series. The proposed scheme is described as a maximum a posteriori approach, where the known expectation maximization algorithm is applied offering closed-form update equations for the model parameters. We have demonstrated that our method produces improved performance and functional activation detection capabilities in both simulated data and real applications. Index Terms-Expectation maximization (EM) algorithm, functional magnetic resonance imaging (fMRI) analysis, general linear regression model (GLM), Markov random field (MRF), relevance vector machine (RVM).

The purpose of the study was to evaluate brain myelination by measuring the magnetization transfer ratio (MTR) and to measure grey (GMV) and white matter volume (WMV) in macrocephalic children with neurofibromatosis type 1 (NF1). Seven NF1 patients (aged 0.65-16.67 years) and seven age- and gender-matched controls were studied. A three-dimensional (3D) gradient echo sequence with and without magnetization transfer (MT) prepulse was used for MTR assessment. Volume measurements of GM and WM were performed by applying segmentation techniques on T2-weighted turbo spin echo images (T2WI). MTR of unidentified bright objects (UBOs) on T2WI in cerebellar white matter (52.8+/-3.3), cerebral peduncles (48.5+/-1.5), hippocampus (52.6+/-1.1), internal capsule (55.7+/-0.3), globus pallidus (52.7+/-3.9), and periventricular white matter (52.6+/-1.2) was lower than in the corresponding areas of controls (64.6+/-2.5, 60.8+/-1.3, 56.4+/-0.9, 64.7+/-1.9, 59.2+/-2.3, 63.6+/-1.7, respectively; p<0.05). MTR of normal-appearing brain tissue in patients was not significantly different than in controls. Surface area (mm(2)) of the corpus callosum (809.1+/-62.8), GMV (cm(3)) (850.7+/-42.9), and white matter volume (WMV) (cm(3)) (785.1+/-85.2) were greater in patients than in controls (652.5+/-52.6 mm(2), 611.2+/-92.1 cm(3), 622.5+/-108.7 cm(3), respectively; p<0.05). To conclude, macrocephaly in NF1 patients is related to increased GMV and WMV and corpus callosum enlargement. MTR of UBOs is lower than that of normal brain tissue.