@article {272, title = {Intervertebral disc segmentation and volumetric reconstruction from peripheral quantitative computed tomography imaging}, journal = {IEEE Transactions on Biomedical Engineering}, volume = {56}, year = {2010}, pages = {2748 - 2751}, abstract = {

An automatic system for segmenting and constructing volumetric representations of excised intervertebral discs from peripheral quantitative computed tomography (PQCT) imagery is presented. The system is designed to allow for automatic quantitative analysis of progressive herniation damage to the intervertebral discs under flexion/extension motions combined with a compressive load. Automatic segmentation and volumetric reconstruction of intervertebral disc from PQCT imagery is a very challenging problem due to factors such as streak artifacts and unclear material density separation between contrasted intervertebral disc and surrounding bone in the PQCT imagery, as well as the formation of multiple contrasted regions under axial scans. To address these factors, a novel multiscale level set approach based on the Mumford-Shaft energy functional in iterative bilateral scale space is employed to segment the intervertebral disc regions from the PQCT imagery. A Delaunay triangulation is then performed based on the set of points associated with the intervertebral disc regions to construct the volumetric representation of the intervertebral disc. Experimental results show that the proposed system achieves segmentation and volumetric reconstructions of intervertebral discs with mean absolute distance error below 0.8 mm when compared to ground truth measurements. The proposed system is currently in operational use as a visualization tool for studying progressive intervertebral disc damage.

}, keywords = {Animals, Artificial Intelligence, Automated, automatic quantitative analysis, Computer-Assisted, computerised tomography, Delaunay triangulation, flexion-extension motions, image reconstruction, image segmentation, Imaging, intervertebral disc segmentation, Intervertebral Disk, Intervertebral Disk Displacement, iterative bilateral scale space, material density separation, medical image processing, Mumford-Shah energy functional, Pattern Recognition, peripheral quantitative computed tomography imaging, progressive herniation damage, Radiographic Image Enhancement, Radiographic Image Interpretation, Swine, Three-Dimensional, Tomography, volumetric reconstruction, X-Ray Computed}, issn = {0018-9294}, doi = {http://dx.doi.org/10.1109/TBME.2009.2027225}, author = {A Wong and A Mishra and J Yates and P Fieguth and D A. Clausi and J Callaghan} } @article {225, title = {Multivariate image segmentation using semantic region growing with adaptive edge penalty}, journal = {IEEE Transactions on Image Processing}, volume = {19}, year = {2010}, pages = {2157 - 2170}, abstract = {

Multivariate image segmentation is a challenging task, influenced by large intraclass variation that reduces class distinguishability as well as increased feature space sparseness and solution space complexity that impose computational cost and degrade algorithmic robustness. To deal with these problems, a Markov random field (MRF) based multivariate segmentation algorithm called $\#$x201C;multivariate iterative region growing using semantics $\#$x201D; (MIRGS) is presented. In MIRGS, the impact of intraclass variation and computational cost are reduced using the MRF spatial context model incorporated with adaptive edge penalty and applied to regions. Semantic region growing starting from watershed over-segmentation and performed alternatively with segmentation gradually reduces the solution space size, which improves segmentation effectiveness. As a multivariate iterative algorithm, MIRGS is highly sensitive to initial conditions. To suppress initialization sensitivity, it employs a region-level k -means (RKM) based initialization method, which consistently provides accurate initial conditions at low computational cost. Experiments show the superiority of RKM relative to two commonly used initialization methods. Segmentation tests on a variety of synthetic and natural multivariate images demonstrate that MIRGS consistently outperforms three other published algorithms.

}, keywords = {adaptive edge penalty, Algorithms, Automated, Computer-Assisted, image enhancement, image interpretation, image segmentation, iterative methods, Markov processes, Markov random field, MRF spatial context model, Multivariate Analysis, multivariate image segmentation, multivariate iterative region growing, Pattern Recognition, region-level k -means based initialization, Reproducibility of Results, semantic region growing, Semantics, Sensitivity and Specificity}, issn = {1057-7149}, doi = {http://dx.doi.org/10.1109/TIP.2010.2045708}, author = {K Qin and D A. Clausi} } @inproceedings {330, title = {Improved interactive medical image segmentation using Enhanced Intelligent Scissors (EIS)}, booktitle = {30th Annual Conference of the IEEE Engineering in Medicine and Biology Society}, year = {2008}, month = {08/2008}, address = {Vancouver, British Columbia, Canada}, abstract = {

A novel interactive approach called Enhanced Intelligent Scissors (EIS) is presented for segmenting regions of interest in medical images. The proposed interactive medical image segmentation algorithm addresses the issues associated with segmenting medical images and allows for fast, robust, and flexible segmentation without requiring accurate manual tracing. A robust complex wavelet phase-based representation is used as an external local cost to address issues associated with contrast non-uniformities and noise typically found in medical images. The boundary extraction problem is formulated as a Hidden Markov Model (HMM) and the novel approach to the second-order Viterbi algorithm with state pruning is used to find the optimal boundary in a robust and efficient manner based on the extracted external and internal local costs, thus handling much inexact user boundary definitions than existing methods. Experimental results using MR and CT images show that the proposed algorithm achieves accurate segmentation in medical images without the need for accurate boundary definition as per existing Intelligent Scissors methods. Furthermore, usability testing indicate that the proposed algorithm requires significantly less user interaction than Intelligent Scissors.

}, keywords = {Algorithms, Artificial Intelligence, Automated, Computer Simulation, Computer-Assisted, Diagnostic Imaging, Humans, image enhancement, image interpretation, Information Storage and Retrieval, Markov Chains, Models, Pattern Recognition, Reproducibility of Results, Statistical, Visible Human Projects}, doi = {http://dx.doi.org/10.1109/IEMBS.2008.4649855}, author = {A Mishra and A Wong and W Zhang and D A. Clausi and P Fieguth} } @article {316, title = {IRGS: Image segmentation using edge penalties and region growing}, journal = {IEEE Transactions on Pattern Analysis and Machine Intelligence}, volume = {30}, year = {2008}, pages = {2126 - 2139}, abstract = {

This paper proposes an image segmentation method named iterative region growing using semantics (IRGS), which is characterized by two aspects. First, it uses graduated increased edge penalty (GIEP) functions within the traditional Markov random field (MRF) context model in formulating the objective functions. Second, IRGS uses a region growing technique in searching for the solutions to these objective functions. The proposed IRGS is an improvement over traditional MRF based approaches in that the edge strength information is utilized and a more stable estimation of model parameters is achieved. Moreover, the IRGS method provides the possibility of building a hierarchical representation of the image content, and allows various region features and even domain knowledge to be incorporated in the segmentation process. The algorithm has been successfully tested on several artificial images and synthetic aperture radar (SAR) images.

}, keywords = {Algorithms, artificial images, Artificial Intelligence, Automated, Computer-Assisted, edge strength information, GIEP, graduated increased edge penalty, image enhancement, image interpretation, image segmentation, IRGS, iterative methods, iterative region growing, Markov processes, Markov random field context model, model parameter estimation, MRF, Pattern Recognition, random processes, Reproducibility of Results, Sensitivity and Specificity, Subtraction Technique, synthetic aperture radar images}, doi = {10.1109/TPAMI.2008.15}, author = {Q Yu and D A. Clausi} } @inproceedings {331, title = {Shape-guided active contour based segmentation and tracking of lumbar vertebrae in video fluoroscopy using complex wavelets}, booktitle = {30th Annual International Conference of the IEEE Engineering in Medicine and Biology Society}, year = {2008}, month = {08/2008}, address = {Vancouver, British Columbia, Canada}, abstract = {

This paper presents a novel shape-guided active contour based approach for segmenting and tracking lumbar vertebrae in video fluoroscopy using complex-valued wavelets. representations. Due to low radiation exposure levels, fluoroscopic images are characterized by low signal-to-noise ratios, low contrast resolution, and illumination non-homogeneities both spatially and temporally, making current methods ill-suited for segmenting and tracking lumbar vertebrae based on existing energy functionals. Furthermore, current methods perform poorly in situations characterized by high curvature as found in the structure of lumbar spine vertebrae. In this paper, a novel iterative estimation approach is used to determine an external energy functional based on complex wavelets. A shaped-guided algorithm is used to evolve the contour around a lumbar spine vertebra based on the complex wavelet energy. The high curvature exhibited by the lumbar spine vertebra is addressed through a novel importance sampling scheme. Experimental results show that the proposed algorithm achieves significantly better segmentation and tracking performance for lumbar spine vertebrae in fluoroscopic images when compared to existing techniques.

}, keywords = {Algorithms, Artificial Intelligence, Automated, Computer-Assisted, Female, Fluoroscopy, Humans, Lumbar Vertebrae, Male, Pattern Recognition, Radiographic Image Enhancement, Radiographic Image Interpretation, Reproducibility of Results, Sensitivity and Specificity, Video Recording, Young Adult}, doi = {http://dx.doi.org/10.1109/IEMBS.2008.4649290}, author = {A Wong and A Mishra and P Fieguth and D A. Clausi and N M. Dunk and J Callaghan} } @inproceedings {349, title = {Watershed deconvolution for cell segmentation}, booktitle = {30th Annual Conference of the IEEE Engineering in Medicine and Biology Society}, year = {2008}, abstract = {

Cell segmentation and/or localization is the first stage of a (semi)automatic tracking system. We addressed the cell localization problem in our previous work where we characterized a typical blood stem cell in a microscopic image as an approximately circular object with dark interior and bright boundary. We also addressed the modelling of adjacent and dividing cells in our previous work as a deconvolution method to model individual blood stem cell as well as adjacent and dividing blood stem cells where an optimization algorithm was combined with a template matching method to segment cell regions and locate the cell centers. Our previous cell deconvolution method is capable of modelling different cell types with changes in the model parameters. However in cases where either a complex parameterized shape is needed to model a specific cell type, or in place of cell center localization, an exact cell segmentation is needed, this method will not be effective. In this paper we propose a method to achieve cell boundary segmentation. Considering cell segmentation as an inverse problem, we assume that cell centers are located in advance. Then, the cell segmentation will be solved by finding cell regions for optimal representation of cell centers while a template matching method is effectively employed to localize cell

}, keywords = {Algorithms, Artificial Intelligence, Automated, Cells, Computer-Assisted, Cultured, hematopoietic stem cells, Humans, image enhancement, image interpretation, Microscopy, Pattern Recognition, Reproducibility of Results, Sensitivity and Specificity}, doi = {http://dx.doi.org/10.1109/IEMBS.2008.4649168}, author = {N Nezamoddin Kachouie and P Fieguth and E Jervis} } @article {415, title = {Design-based texture feature fusion using Gabor filters and co-occurrence probabilities}, volume = {14}, year = {2005}, pages = {925 - 936}, abstract = {

A design-based method to fuse Gabor filter and grey level co-occurrence probability (GLCP) features for improved texture recognition is presented. The fused feature set utilizes both the Gabor filter\&$\#$39;s capability of accurately capturing lower and mid-frequency texture information and the GLCP\&$\#$39;s capability in texture information relevant to higher frequency components. Evaluation methods include comparing feature space separability and comparing image segmentation classification rates. The fused feature sets are demonstrated to produce higher feature space separations, as well as higher segmentation accuracies relative to the individual feature sets. Fused feature sets also outperform individual feature sets for noisy images, across different noise magnitudes. The curse of dimensionality is demonstrated not to affect segmentation using the proposed the 48-dimensional fused feature set. Gabor magnitude responses produce higher segmentation accuracies than linearly normalized Gabor magnitude responses. Feature reduction using principal component analysis is acceptable for maintaining the segmentation performance, but feature reduction using the feature contrast method dramatically reduced the segmentation accuracy. Overall, the designed fused feature set is advocated as a means for improving texture segmentation performance.

}, keywords = {Algorithms, Artificial Intelligence, Automated, Computer Simulation, Computer-Assisted, design-based texture feature fusion, feature contrast, feature reduction, Gabor filter, grey level cooccurrence probability, image classification, image enhancement, image interpretation, image segmentation, image segmentation classification, image texture, Information Storage and Retrieval, Models, Pattern Recognition, principal component analysis, probability, Statistical, Subtraction Technique, texture recognition}, issn = {1057-7149}, doi = {http://dx.doi.org/10.1109/TIP.2005.849319}, author = {D A. Clausi and H Deng} } @article {416, title = {Modeling emotional content of music using system identification}, journal = {IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics}, volume = {36}, year = {2005}, pages = {588 - 599}, abstract = {

Research was conducted to develop a methodology to model the emotional content of music as a function of time and musical features. Emotion is quantified using the dimensions valence and arousal, and system-identification techniques are used to create the models. Results demonstrate that system identification provides a means to generalize the emotional content for a genre of music. The average R2 statistic of a valid linear model structure is 21.9\% for valence and 78.4\% for arousal. The proposed method of constructing models of emotional content generalizes previous time-series models and removes ambiguity from classifiers of emotion.

}, keywords = {arousal dimension, Artificial Intelligence, Automated, average R/sup 2/ statistic, Computer Simulation, emotion recognition, Emotions, Humans, linear model structure, Models, music, music emotional content modeling, Pattern Recognition, Psychological, Psychometrics, system identification, time series, valence dimension}, issn = {1083-4419}, doi = {http://dx.doi.org/10.1109/TSMCB.2005.862491}, author = {M Korhonen and D A. Clausi and E Jernigan} } @article {418, title = {Statistical processing of large image sequences}, journal = {IEEE Transactions on Image Processing}, volume = {14}, year = {2005}, pages = {80 - 93}, abstract = {

The dynamic estimation of large-scale stochastic image sequences, as frequently encountered in remote sensing, is important in a variety of scientific applications. However, the size of such images makes conventional dynamic estimation methods, for example, the Kalman and related filters, impractical. We present an approach that emulates the Kalman filter, but with considerably reduced computational and storage requirements. Our approach is illustrated in the context of a 512 times; 512 image sequence of ocean surface temperature. The static estimation step, the primary contribution here, uses a mixture of stationary models to accurately mimic the effect of a nonstationary prior, simplifying both computational complexity and modeling. Our approach provides an efficient, stable, positive-definite model which is consistent with the given correlation structure. Thus, the methods of this paper may find application in modeling and single-frame estimation.

}, keywords = {Algorithms, Artificial Intelligence, Automated, Biological, Cluster Analysis, computational complexity, Computer Graphics, Computer Simulation, Computer-Assisted, correlation methods, correlation structure, dynamic estimation method, image enhancement, image interpretation, image sequences, Imaging, Information Storage and Retrieval, Kalman filter, Kalman filters, large-scale stochastic image sequence estimation, Models, Movement, Numerical Analysis, ocean surface temperature, Pattern Recognition, Remote Sensing, Reproducibility of Results, Sensitivity and Specificity, Signal Processing, single-frame estimation, static estimation step, Statistical, statistical analysis, statistical image processing, stochastic processes, Subtraction Technique, Three-Dimensional, Video Recording}, issn = {1057-7149}, doi = {http://dx.doi.org/10.1109/TIP.2004.838703}, author = {F M. Khellah and P Fieguth and M J. Murray and M R. Allen} } @article {690, title = {Gaussian MRF rotation-invariant features for image classification}, journal = {IEEE Transactions on Pattern Analysis and Machine Intelligence}, volume = {26}, year = {2004}, pages = {951 - 955}, abstract = {

Features based on Markov random field (MRF) models are sensitive to texture rotation. This paper develops an anisotropic circular Gaussian MRF (ACGMRF) model for retrieving rotation-invariant texture features. To overcome the singularity problem of the least squares estimate method, an approximate least squares estimate method is designed and implemented. Rotation-invariant features are obtained from the ACGMRF model parameters using the discrete Fourier transform. The ACGMRF model is demonstrated to be a statistical improvement over three published methods. The three methods include a Laplacian pyramid, an isotropic circular GMRF (ICGMRF), and gray level cooccurrence probability features.

}, keywords = {Algorithms, anisotropic circular Gaussian MRF model, Artificial Intelligence, Automated, Computer Simulation, Computer-Assisted, discrete Fourier transform, discrete Fourier transforms, Gaussian MRF rotation-invariant features, gray level cooccurrence probability features, image classification, image enhancement, image interpretation, image texture, Laplacian pyramid, Markov Chains, Markov processes, Markov Random Field models, Models, Normal Distribution, Pattern Recognition, Rotation, rotation-invariant texture features, Statistical, statistical improvement, texture rotation}, issn = {0162-8828}, doi = {http://dx.doi.org/10.1109/TPAMI.2004.30}, author = {H Deng and D A. Clausi} }