19 research outputs found
Brownian motion curve-based textural classification and its application in cancer diagnosis
No full textObjective: To develop an automated diagnostic methodology based on textural features of the oral mucosal epithelium to discriminate normal and oral submucous fibrosis (OSF).Study Design: A total of 83 normal and 29 OSF images from histopathologic sections of the oral mucosa are considered. The proposed diagnostic mechanism consists of two parts: feature extraction using Brownian motion curve (BMC) and design of a suitable classifier. The discrimination ability of the features has been substantiated by statistical tests. An error back-propagation neural network (BPNN) is used to classify OSF vs. normal.Results: In development of an automated oral cancer diagnostic module, BMC has played an important role in characterizing textural features of the oral images. Fisher's linear discriminant analysis yields 100% sensitivity and 85% specificity, whereas BPNN leads to 92.31% sensitivity and 100% specificity, respectively.Conclusion: In addition to intensity and morphology-based features, textural features are also very important, especially in histopathologic diagnosis of oral cancer. In view of this, a set of textural features are extracted using the BMC for the diagnosis of OSF. Finally, a textural
Diagnosis of Hashimoto's thyroiditis in ultrasound using tissue characterization and pixel classification
No full textHashimoto's thyroiditis is the most common type of inflammation of the thyroid gland, and accurate diagnosis of Hashimoto's thyroiditis would be helpful to better manage the disease process and predict thyroid failure. Most of the published computer-based techniques that use ultrasound thyroid images for Hashimoto's thyroiditis diagnosis are limited by lack of procedure standardization because individual investigators use various initial ultrasound settings. This article presents a computer-aided diagnostic technique that uses grayscale features and classifiers to provide a more objective and reproducible classification of normal and Hashimoto's thyroiditis-affected cases. In this paradigm, we extracted grayscale features based on entropy, Gabor wavelet, moments, image texture, and higher order spectra from the 100 normal and 100 Hashimoto's thyroiditis-affected ultrasound thyroid images. Significant features were selected using t-test. The resulting feature vectors were used to build the following three classifiers using tenfold stratified cross validation technique: support vector machine, k-nearest neighbor, and radial basis probabilistic neural network. Our results show that a combination of 12 features coupled with support vector machine classifier with the polynomial kernel of order 1 and linear kernel gives the highest accuracy of 80%, sensitivity of 76%, specificity of 84%, and positive predictive value of 83.3% for the detection of Hashimoto's thyroiditis. The proposed computer-aided diagnostic system uses novel features that have not yet been explored for Hashimoto's thyroiditis diagnosis. Even though the accuracy is only 80%, the presented preliminary results are encouraging to warrant analysis of more such powerful features on larger database
Computed tomography carotid wall plaque characterization using a combination of discrete wavelet transform and texture features:A pilot study
No full text
Evolutionary algorithm based classifier parameter tuning for automatic diabetic retinopathy grading: A hybrid feature extraction approach
No full textAutomated detection of optic disk in retinal fundus images using intuitionistic fuzzy histon segmentation
No full textApplication of higher-order spectra for automated grading of diabetic maculopathy
No full textDiabetic macular edema (DME) is one of the most common causes of visual loss among diabetes mellitus patients. Early detection and successive treatment may improve the visual acuity. DME is mainly graded into non-clinically significant macular edema (NCSME) and clinically significant macular edema according to the location of hard exudates in the macula region. DME can be identified by manual examination of fundus images. It is laborious and resource intensive. Hence, in this work, automated grading of DME is proposed using higher-order spectra (HOS) of Radon transform projections of the fundus images. We have used third-order cumulants and bispectrum magnitude, in this work, as features, and compared their performance. They can capture subtle changes in the fundus image. Spectral regression discriminant analysis (SRDA) reduces feature dimension, and minimum redundancy maximum relevance method is used to rank the significant SRDA components. Ranked features are fed to various supervised classifiers, viz. Naive Bayes, AdaBoost and support vector machine, to discriminate No DME, NCSME and clinically significant macular edema classes. The performance of our system is evaluated using the publicly available MESSIDOR dataset (300 images) and also verified with a local dataset (300 images). Our results show that HOS cumulants and bispectrum magnitude obtained an average accuracy of 95.56 and 94.39 % for MESSIDOR dataset and 95.93 and 93.33 % for local dataset, respectively
SU₂ Nonstandard Bases: Case of Mutually Unbiased Bases
Full text linkThis paper deals with bases in a finite-dimensional Hilbert space. Such a space can be realized as a subspace of the representation space of SU₂ corresponding to an irreducible representation of SU₂. The representation theory of SU₂ is reconsidered via the use of two truncated deformed oscillators. This leads to replacement of the familiar scheme {j²,jz} by a scheme {j²,vra}, where the two-parameter operator vra is defined in the universal enveloping algebra of the Lie algebra su₂. The eigenvectors of the commuting set of operators {j²,vra} are adapted to a tower of chains SO₃⊃C₂j₊₁ (2j∈N∗), where C₂j₊₁ is the cyclic group of order 2j+1. In the case where 2j+1 is prime, the corresponding eigenvectors generate a complete set of mutually unbiased bases. Some useful relations on generalized quadratic Gauss sums are exposed in three appendicesThe senior author (M.R.K.) acknowledges Philippe Langevin for useful correspondence. The authors thank Hubert de Guise, Michel Planat, and Metod Saniga for interesting discussions. They are indebted to Bruce Berndt and Ron Evans for providing them with an alternative proof of the result in Appendix C. Thanks are due to the Referees for useful and constructive suggestions
SU₂ Nonstandard Bases: Case of Mutually Unbiased Bases
No full textThis paper deals with bases in a finite-dimensional Hilbert space. Such a space can be realized as a subspace of the representation space of SU₂ corresponding to an irreducible representation of SU₂. The representation theory of SU₂ is reconsidered via the use of two truncated deformed oscillators. This leads to replacement of the familiar scheme {j²,jz} by a scheme {j²,vra}, where the two-parameter operator vra is defined in the universal enveloping algebra of the Lie algebra su₂. The eigenvectors of the commuting set of operators {j²,vra} are adapted to a tower of chains SO₃⊃C₂j₊₁ (2j∈N∗), where C₂j₊₁ is the cyclic group of order 2j+1. In the case where 2j+1 is prime, the corresponding eigenvectors generate a complete set of mutually unbiased bases. Some useful relations on generalized quadratic Gauss sums are exposed in three appendicesThe senior author (M.R.K.) acknowledges Philippe Langevin for useful correspondence. The authors thank Hubert de Guise, Michel Planat, and Metod Saniga for interesting discussions. They are indebted to Bruce Berndt and Ron Evans for providing them with an alternative proof of the result in Appendix C. Thanks are due to the Referees for useful and constructive suggestions
An integrated index for identification of fatty liver disease using radon transform and discrete cosine transform features in ultrasound images
No full textAlcoholic and non-alcoholic fatty liver disease is one of the leading causes of chronic liver diseases and mortality in Western countries and Asia. Ultrasound image assessment is most commonly and widely used to identify the Non-Alcoholic Fatty Liver Disease (NAFLD). It is one of the faster and safer non-invasive methods of NAFLD diagnosis available in imaging modalities. The diagnosis of NAFLD using biopsies is expensive, invasive, and causes anxiety to the patients. The advent of advanced image processing and data mining techniques have helped to develop faster, efficient, objective, and accurate decision support system for fatty liver disease using ultrasound images. This paper proposes a novel feature extraction models based on Radon Transform (RT) and Discrete Cosine Transform (DCT). First, Radon Transform (RT) is performed on the ultrasound images for every 1 degree to capture the low frequency details. Then 2D-DCT is applied on the Radon transformed image to obtain the frequency features (DCT coefficients). Further the 2D-DCT frequency coefficients (features) obtained are converted to 1D coefficients vector in zigzag fashion. This 1D array of DCT coefficients are subjected to Locality Sensitive Discriminant Analysis (LSDA) to reduce the number of features. Then these features are ranked using minimum Redundancy and Maximum Relevance (mRMR) ranking method. Finally, highly ranked minimum numbers of features are fused using Decision Tree (DT), k-Nearest Neighbour (k-NN), Probabilistic Neural Network (PNN), Support Vector Machine (SVM), Fuzzy Sugeno (FS) and AdaBoost classifiers to get the highest classification performance. In this work, we have obtained an average accuracy, sensitivity and specificity of 100% in the detection of NAFLD using FS classifier. Also, we have devised an integrated index named as Fatty Liver Disease Index (FLDI) by fusing two significant LSDA components to distinguish normal and FLD class with single number
