Multivariate Statistical Model for 3D Image Segmentation with Application to Medical Images

Abstract

In this article we describe a statistical model that was developed to segment brain magnetic resonance images. The statistical segmentation algorithm was applied after a pre-processing stage involving the use of a 3D anisotropic filter along with histogram equalization techniques. The segmentation algorithm makes use of prior knowledge and a probabilitybased multivariate model designed to semi-automate the process of segmentation. The algorithm was applied to images obtained from the Center for Morphometric Analysis at Massachusetts General Hospital as part of the Internet Brain Segmentation Repository (IBSR). The developed algorithm showed improved accuracy over the k-means, adaptive Maximum Apriori Probability (MAP), biased MAP, and other algorithms. Experimental results showing the segmentation and the results of comparisons with other algorithms are provided. Results are based on an overlap criterion against expertly segmented images from the IBSR. The algorithm produced average results of approximately 80% overlap with the expertly segmented images (compared with 85% for manual segmentation and 55% for other algorithms).

Evaluation of automated brain MR image segmentation and volumetry methods

“…Several studies have been conducted where different skull-stripping and brain segmentation algorithms are compared, both with each other, and with manual tracing, or realistic digital brain phantoms [Barra and Boire, 2001;Byrum et al, 1996;Cuadra et al, 2005;Fennema-Notestine et al, 2006;Good et al, 2002;Grabowski et al, 2000;Greenspan et al, 2006;Heckemann et al, 2006;John et al, 2003;Kovacevic et al, 2002;Lemieux et al, 2003;Moretti et al, 2000;Rehm et al, 2004;Toga and Thompson, 2003;Wang and Doddrell, 2002;Warfield et al, 2004;Zaidi et al, 2006;Bezdek et al, 1993]. Also, the impact of MR image acquisitions protocols on tissue segmentation results and brain volumes has been investigated [e.g., Lundervold et al, 2000;Clark et al, 2006], and one study specifically addressed reproducibility of volumetry results over time of Chard et al [2002].…”

Hidden Markov models-based 3D MRI brain segmentation

“…3D anisotropic filtering as described in [13], using kZ5 and for 10 iterations. Previously in [14], we showed how Global Intensity level correction could be applied to MRI sequences. Also in [14], we showed how sudden intensity variations that appear in many MR sequences could be accounted for.…”

“…Previously in [14], we showed how Global Intensity level correction could be applied to MRI sequences. Also in [14], we showed how sudden intensity variations that appear in many MR sequences could be accounted for. The same techniques were used here for preprocessing of imaged prior to segmentation.…”

Automated detection of intercellular signaling in astrocyte networks using the converging squares algorithm

“…After the whole image sequence was completed, the combined peaks were clustered using the k-means clustering algorithm, which clusters objects by minimizing a squared error function (John et al, 2003) and a radius of 4-pixels. This step eliminated duplicate peaks and further refined peak location by averaging across multiple frames (Fig.…”