An accurate method for tissue type detection and classification on T2- and diffusion-weighted brain MRI based on the image intensity exponential decay is presented. Relaxation weighted images and diffusion coded images were both obtained at equal intervals of image setting parameters, allowing for an appropriate data set for Prony method application. Furthermore, diffusion coded images were obtained for three orthogonal gradient directions, allowing for image data parametrized by the Apparent Diffusion Coefficient (ADC) or independently by each orthogonal diffusion tensor component. Image noise was reduced by the application of morphological operators to select representative decaying behaviors and geometrical regularization of the decay data points by adjusting its geometrical properties to those that are expected from noiseless data, i.e., monotonous and convex behavior. Data points were fitted by an over determined Prony interpolation procedure. Relaxation rate and diffusion coefficient distributions were obtained and tissue classification was performed by means of the determination of principal relaxation rate and diffusion coefficients or diffusion modes using a suitable mathematical morphology operator, i.e., watershed or similar. Image segmentation was performed by linear regression analysis on a pixel by pixel basis assuming that the pixel intensity decay is composed by a linear superposition of the relaxation rate or diffusion modes previously obtained from the correspondent distribution functions. The main advantage of the proposed method rests in
its accuracy and speed of calculation with respect to other methods such as Inverse Laplace Transform algorithms, making it suitable for on line application on imaging data.