Limits...
Rough-fuzzy clustering and unsupervised feature selection for wavelet based MR image segmentation.

Maji P, Roy S - PLoS ONE (2015)

Bottom Line: Image segmentation is an indispensable process in the visualization of human tissues, particularly during clinical analysis of brain magnetic resonance (MR) images.The dyadic wavelet analysis is used to extract the scale-space feature vector for each pixel, while the rough-fuzzy clustering is used to address the uncertainty problem of brain MR image segmentation.An unsupervised feature selection method is introduced, based on maximum relevance-maximum significance criterion, to select relevant and significant textural features for segmentation problem, while the mathematical morphology based skull stripping preprocessing step is proposed to remove the non-cerebral tissues like skull.

View Article: PubMed Central - PubMed

Affiliation: Biomedical Imaging and Bioinformatics Lab, Machine Intelligence Unit, Indian Statistical Institute, 203 B. T. Road, Kolkata, 700 108, India.

ABSTRACT
Image segmentation is an indispensable process in the visualization of human tissues, particularly during clinical analysis of brain magnetic resonance (MR) images. For many human experts, manual segmentation is a difficult and time consuming task, which makes an automated brain MR image segmentation method desirable. In this regard, this paper presents a new segmentation method for brain MR images, integrating judiciously the merits of rough-fuzzy computing and multiresolution image analysis technique. The proposed method assumes that the major brain tissues, namely, gray matter, white matter, and cerebrospinal fluid from the MR images are considered to have different textural properties. The dyadic wavelet analysis is used to extract the scale-space feature vector for each pixel, while the rough-fuzzy clustering is used to address the uncertainty problem of brain MR image segmentation. An unsupervised feature selection method is introduced, based on maximum relevance-maximum significance criterion, to select relevant and significant textural features for segmentation problem, while the mathematical morphology based skull stripping preprocessing step is proposed to remove the non-cerebral tissues like skull. The performance of the proposed method, along with a comparison with related approaches, is demonstrated on a set of synthetic and real brain MR images using standard validity indices.

Show MeSH
Original images of IBSR for volume no. 1, 2, 3, 4, 5, 11, 12, 13, 14, and 17.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4388859&req=5

pone.0123677.g003: Original images of IBSR for volume no. 1, 2, 3, 4, 5, 11, 12, 13, 14, and 17.

Mentions: All the methods are implemented in C language and run in LINUX environment having machine configuration Intel(R) Core(TM) i7-2600 CPU @3.40GHz×8 and 16 GB RAM. To analyze the performance of different algorithms and measures, the experimentation is done on some benchmark simulated MR images obtained from “BrainWeb: Simulated Brain Database” (www.bic.mni.mcgill.ca/brainweb/) and real MR images of “IBSR: Internet Brain Segmentation Repository” (www.cma.mgh.harvard.edu/ibsr/). All the image volumes of BrainWeb and IBSR are of size 256×256×181 and 256×128×256, respectively. The middle slice of each volume is considered for both qualitative and quantitative analysis. Figs 2 and 3 depict some of the original images of BrainWeb and IBSR data sets, respectively, while Figs 4, 5, 6, and 7 present the segmented images obtained using different methods, along with the ground truth images. The first and second columns of Figs 4, 5, 6, and 7 show the ground truth images and output images obtained using the proposed method, while remaining columns present the segmented images produced by different methods.


Rough-fuzzy clustering and unsupervised feature selection for wavelet based MR image segmentation.

Maji P, Roy S - PLoS ONE (2015)

Original images of IBSR for volume no. 1, 2, 3, 4, 5, 11, 12, 13, 14, and 17.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4388859&req=5

pone.0123677.g003: Original images of IBSR for volume no. 1, 2, 3, 4, 5, 11, 12, 13, 14, and 17.
Mentions: All the methods are implemented in C language and run in LINUX environment having machine configuration Intel(R) Core(TM) i7-2600 CPU @3.40GHz×8 and 16 GB RAM. To analyze the performance of different algorithms and measures, the experimentation is done on some benchmark simulated MR images obtained from “BrainWeb: Simulated Brain Database” (www.bic.mni.mcgill.ca/brainweb/) and real MR images of “IBSR: Internet Brain Segmentation Repository” (www.cma.mgh.harvard.edu/ibsr/). All the image volumes of BrainWeb and IBSR are of size 256×256×181 and 256×128×256, respectively. The middle slice of each volume is considered for both qualitative and quantitative analysis. Figs 2 and 3 depict some of the original images of BrainWeb and IBSR data sets, respectively, while Figs 4, 5, 6, and 7 present the segmented images obtained using different methods, along with the ground truth images. The first and second columns of Figs 4, 5, 6, and 7 show the ground truth images and output images obtained using the proposed method, while remaining columns present the segmented images produced by different methods.

Bottom Line: Image segmentation is an indispensable process in the visualization of human tissues, particularly during clinical analysis of brain magnetic resonance (MR) images.The dyadic wavelet analysis is used to extract the scale-space feature vector for each pixel, while the rough-fuzzy clustering is used to address the uncertainty problem of brain MR image segmentation.An unsupervised feature selection method is introduced, based on maximum relevance-maximum significance criterion, to select relevant and significant textural features for segmentation problem, while the mathematical morphology based skull stripping preprocessing step is proposed to remove the non-cerebral tissues like skull.

View Article: PubMed Central - PubMed

Affiliation: Biomedical Imaging and Bioinformatics Lab, Machine Intelligence Unit, Indian Statistical Institute, 203 B. T. Road, Kolkata, 700 108, India.

ABSTRACT
Image segmentation is an indispensable process in the visualization of human tissues, particularly during clinical analysis of brain magnetic resonance (MR) images. For many human experts, manual segmentation is a difficult and time consuming task, which makes an automated brain MR image segmentation method desirable. In this regard, this paper presents a new segmentation method for brain MR images, integrating judiciously the merits of rough-fuzzy computing and multiresolution image analysis technique. The proposed method assumes that the major brain tissues, namely, gray matter, white matter, and cerebrospinal fluid from the MR images are considered to have different textural properties. The dyadic wavelet analysis is used to extract the scale-space feature vector for each pixel, while the rough-fuzzy clustering is used to address the uncertainty problem of brain MR image segmentation. An unsupervised feature selection method is introduced, based on maximum relevance-maximum significance criterion, to select relevant and significant textural features for segmentation problem, while the mathematical morphology based skull stripping preprocessing step is proposed to remove the non-cerebral tissues like skull. The performance of the proposed method, along with a comparison with related approaches, is demonstrated on a set of synthetic and real brain MR images using standard validity indices.

Show MeSH