Taranjit Kaur

Optimized Multi Threshold Brain Tumor Image Segmentation Using Two Dimensional Minimum Cross Entropy Based on Co-occurrence Matrix

The present chapter proposes an automatic segmentation method that performs multilevel image thresholding by using the spatial information encoded in the gray level co-occurrence matrix (GLCM). The 2D local cross entropy approach that has been designed by extending the one dimensional (1-D) cross entropy thresholding method to a two dimensional (2D) one using the GLCM, serves as a fitness function. The use of conventional exhaustive search based implementations for multilevel thresholding are computationally expensive. Under such conditions evolutionary algorithm like particle swarm optimization (PSO) has been used. The effectiveness of this method was tested on brain tumor MR images and comparison was done with seven other level set based segmentation algorithms, using three different measures (1) Jaccard, (2) Dice and (3) Root mean square error (RMSE). The results demonstrate that average metric values were equal to 0.881902, 0.936394 and 0.070123 for proposed approach, which were significantly better than existing techniques.

A novel feature selection method for brain tumor MR image classification based on the Fisher criterion and parameter-free Bat optimization

The present paper proposes a novel feature selection technique for the MR brain tumor image classification that aims to choose the optimal feature subset with maximum discriminatory ability in the minimum amount of time. It is based on the fusion of the Fisher and the parameter-free Bat (PFree Bat) optimization algorithm. As the conventional Bat algorithm is bad at exploration, a modification is proposed that guides the Bat by the pulse frequency, global best and the local best position. This improved version of Bat referred to as the PFree Bat algorithm eliminates the velocity equation and directly updates the Bat position. Subsequently, this method in conjunction with the Fisher criteria has been used to select the best set of features for brain tumor classification. The chosen features are then fed to the commonly used least square (LS) support vector machine (SVM) classifier to categorize the area of interest into the high or low grade. For the evaluation of the proposed attribute selection method, tenfold cross-validation has been conducted on a set of 95 ROIs taken from the BRATS 2012 dataset. On an extensive comparison with the other hybrid approaches, the proposed approach brought about the 100% recognition rate in the smallest amount of time. Furthermore, an integrated index is proposed that uniquely identifies the best performing algorithm, taking into account the accuracy, number of features and the computational time. For the fair comparison, the performance of the proposed method has also been examined on breast cancer dataset taken from UCI repository. The obtained results validate that the designed algorithm has better average accuracy than existing state-of-the-art works.

A joint intensity and edge magnitude-based multilevel thresholding algorithm for the automatic segmentation of pathological MR brain images

Multilevel thresholding is one of the most popular image segmentation techniques due to its simplicity and accuracy. Most of the thresholding approaches use either the histogram of an image or information from the grey-level co-occurrence matrix (GLCM) to compute the threshold. The medical images like MRI usually have vague boundaries and poor contrast. So, segmenting these images using solely histogram or texture attributes of GLCM proves to be insufficient. This paper proposes a novel multilevel thresholding approach for automatic segmentation of tumour lesions from magnetic resonance images. The proposed technique exploits both intensity and edge magnitude information present in image histogram and GLCM to compute the multiple thresholds. Subsequently, using both attributes, a hybrid fitness function has been formulated which can capture the variations in intensity and the edge magnitude present in different tumour groups effectively. Mutation-based particle swarm optimization (MPSO) technique has been used to optimize the fitness function so as to mitigate the problem of high computational complexity existing in the exhaustive search methods. Moreover, MPSO has better exploration capabilities as compared to conventional particle swarm optimization. The performance of the devised technique has been evaluated and compared with two other intensity- and texture-based approaches using three different measures: Jaccard, Dice and misclassification error. To compute these quantitative metrics, experiments were conducted on a series of images, including low-grade glioma tumour volumes taken from brain tumour image segmentation benchmark 2012 and 2015 data sets and real clinical tumour images. Experimental results show that the proposed approach outperforms the other competing algorithms by achieving an average value equal to 0.752, 0.854, 0.0052; 0.648, 0.762, 0.0177; 0.710, 0.813, 0.0148 and 0.886, 0.937, 0.0037 for four different data sets.

ENHANCEMENT OF SOLUBILITY AND DISSOLUTION RATE OF SIMVASTATIN BY USING SOLID DISPERSION TECHNIQUE ALONG WITH DIFFERENT COMBINATION OF POLYMERS

The solubility and dissolution rate of simvastatin, a drug used for the treatment of hyperlipidaemia. Simvastatin is a selective competitive inhibitor of HMG Co A reductase. However its absolute bioavailability is 5%. To increase the solubility of drug solid dispersion was prepared. Solid dispersion preliminary solubility analysis was carried out for the selection of the carrier and solid dispersion was prepared with Hydroxy Propyl Methyl Cellulose (HPMC) and Methyl Cellulose (MC). These solid dispersions were analyzed for the solubility and in-vitro dissolution profile solid dispersion of drug with polymer has shown enhanced solubility with improved dissolution rate. Further FTIR, X-Ray studies were carried out. Solid dispersion prepared with polymer in 1:5 ratios shows the presence of amorphous form confirmed by the characterization study. The study also shows that dissolution rate of simvastatin can be enhanced to considerable extent by solid dispersion technique with Polymer. Keywords: Solubility enhancement, Solid dispersion, Low aqueous solubility

Image Denoising using Alexander Fractional Hybrid Filter

In this paper, a hybrid filter based on the concept of fractional calculus and Alexander polynomial is proposed. The hybrid filtering mask is constructed by convolving the designed Alexander fractional differential and integral masks. The hybrid mask shows high robustness for images corrupted with Gaussian, salt & pepper, and speckle noises. For the experimentation, the standard and real world noisy images are used. The qualitative comparison shows that the proposed hybrid filter has better denoising with high edge preserving capability as compared to the other existing filters. Quantitatively, the performance of the proposed hybrid filter is also evaluated by the measures such as peak signal to noise ratio (PSNR), normalized cross-correlation (NK), average difference (AD), structural content (SC), maximum difference (MD) and normalized absolute error (NAE) on standard set of images. The average values of these metrics for Gaussian noise with maximum standard deviation σ=25 are PSNR = 32.729, NK = 0.8190,...

A novel fully automatic multilevel thresholding technique based on optimized intuitionistic fuzzy sets and tsallis entropy for MR brain tumor image segmentation

In the present paper, a hybrid multilevel thresholding technique that combines intuitionistic fuzzy sets and tsallis entropy has been proposed for the automatic delineation of the tumor from magnetic resonance images having vague boundaries and poor contrast. This novel technique takes into account both the image histogram and the uncertainty information for the computation of multiple thresholds. The benefit of the methodology is that it provides fast and improved segmentation for the complex tumorous images with imprecise gray levels. To further boost the computational speed, the mutation based particle swarm optimization is used that selects the most optimal threshold combination. The accuracy of the proposed segmentation approach has been validated on simulated, real low-grade glioma tumor volumes taken from MICCAI brain tumor segmentation (BRATS) challenge 2012 dataset and the clinical tumor images, so as to corroborate its generality and novelty. The designed technique achieves an average Dice overlap equal to 0.82010, 0.78610 and 0.94170 for three datasets. Further, a comparative analysis has also been made between the eight existing multilevel thresholding implementations so as to show the superiority of the designed technique. In comparison, the results indicate a mean improvement in Dice by an amount equal to 4.00% (p < 0.005), 9.60% (p < 0.005) and 3.58% (p < 0.005), respectively in contrast to the fuzzy tsallis approach.

Quantitative metric for MR brain tumour grade classification using sample space density measure of analytic intrinsic mode function representation

In the present study, a new feature named as density measure is proposed for classification of the glioma brain tumour magnetic resonance (MR) image into low and high-grade categories. The proposed feature is derived using improved complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN) and Hilbert transformation technique. The proposed feature uses the difference signal created by mapping of the fluid attenuation inversion recovery segmented region onto T1 and T1-contrast-enhanced MR images. This difference signal is decomposed into various intrinsic mode functions (IMFs) using improved CEEMDAN algorithm that effectively captures the texture variations existing in both tumour groups. Then, the Hilbert transformation of resulting IMF is computed that provides the analytic signal representation, thereby giving a better visualisation of the texture difference. The proposed feature is calculated from this analytic signal representation at 97% confidence level. Subsequently, this feature is utilised to calculate a quantitative metric for tumour classification. The proposed metric has been validated on 120 tumorous images taken from brain tumour image segmentation benchmark (BRATS) 2012 dataset and the images taken from Harvard Medical School repository. The results illustrate that proposed metric yields an overall classification accuracy of 100% which is better than existing state-of-art works.