Title :
Fast iterative segmentation of high resolution medical images
Author_Institution :
Dept. of Electr. Eng., Houston Univ., TX, USA
fDate :
6/1/1997 12:00:00 AM
Abstract :
Various applications in positron emission tomography (PET), single photon emission computed tomography (SPECT) and magnetic resonance imaging (MRI) require segmentation of 20 to 60 high resolution images of size 256×256 pixels in 3-9 seconds per image. This places particular constraints on the design of image segmentation algorithms. This paper proposes a quantized data representation and a quantised EM algorithm for estimating the parameters of a finite mixture density function to be used in a Bayes classifier for image segmentation. Both a Monte Carlo evaluation and an application to MRI images showed that the quantized EM algorithm can dramatically reduce the required computation time with negligible difference in mean estimation error and mean classification error
Keywords :
Bayes methods; Monte Carlo methods; biomedical NMR; error statistics; image representation; image resolution; image segmentation; iterative methods; maximum likelihood estimation; medical image processing; positron emission tomography; single photon emission computed tomography; 256 pixel; 65536 pixel; Bayes classifier; MRI; Monte Carlo evaluation; PET; SPECT; computation time; fast iterative segmentation; finite mixture density function; high resolution medical images; image segmentation; image segmentation algorithms; magnetic resonance imaging; mean classification error; mean estimation error; positron emission tomography; quantised EM algorithm; quantized data representation; single photon emission computed tomography; Algorithm design and analysis; Biomedical imaging; Density functional theory; Image resolution; Image segmentation; Magnetic resonance imaging; Parameter estimation; Pixel; Positron emission tomography; Single photon emission computed tomography;
Journal_Title :
Nuclear Science, IEEE Transactions on
