Hasan Koyuncu

Artificial neural network based on rotation forest for biomedical pattern classification

The novel classifier system based on ensemble classifier is proposed in this paper. Rotation forest algorithm based on principal component algorithm was used as ensemble classifier method. In presented classifier system, artificial neural network was used as base classifier in this ensemble classifier system. Rotation forest structure has been generally realized with decision trees in literature. But, multilayer perceptron neural network was utilized as base classifier in rotation forest structure in our study. However, principal component analysis was used for obtaining different feature sets from original data set. The proposed RF-ANN structure was applied to Wisconsin breast cancer data taken form UCI Database. The obtained results were compared with the results of neural network optimized particle swarm optimization (PSO-ANN). The realized experimental studies were represented that RF-ANN structure was successful than PSO-ANN structure. RF-ANN classified breast cancer dataset with 98.05% classification accuracy using 9 classifiers.

A hybrid tool on denoising and enhancement of abdominal CT images before organ & tumour segmentation

Most of abdominal CT images include Gaussian noise, and CT scans form a blurry vision because of the internal fat tissue inside of abdomen. These two handicaps (noise and fat tissue) constitute an impediment in front of an accurate abdominal organ & tumour segmentation. Also segmentation techniques generally fall into error on segmentation of close grayscale regions. Therefore, denoising and enhancement parts are crucial for better segmentation results on CT images. In this paper, we form a tool including three efficient algorithms for the purpose of image enhancement before abdominal organ & tumour segmentation. At first, the denoising process is realized by Block Matching and 3D Filtering (BM3D) algorithm for elimination of Gaussian noise stated in arterial phase CT images. At second, Fast Linking Spiking Cortical Model (FL-SCM) is used for removing the internal fat tissue. At last, Otsu algorithm is processed to remove the redundant parts within the image. In experiments, Peak Signal-to-Noise Ratio (PSNR) and Structural Similarity (SSIM) index are used to evaluate the performance of proposed method, and a visual comparison is presented. According to results, it is seen that proposed tool obtains the best PSNR and SSIM values in comparison with two steps of pipeline (FL-SCM and BM3D & FL-SCM). Consequently, BM3D & FL-SCM & Otsu (BFO) ensures a clean abdomen particularly for segmentation of liver, spleen, pancreas, adrenal tumours, aorta, ribs, spinal cord and kidneys.

A New Breakpoint in Hybrid Particle Swarm-Neural Network Architecture: Individual Boundary Adjustment

Neural Network (NN) is an effective classifier, but it generally uses the Backpropagation type algorithms which are insufficient because of trapping to local minimum of error rate. For elimination of this handicap, stochastic optimization algorithms are used to update the parameters of NN. Particle Swarm Optimization (PSO) is one of these providing a robust coherence with NN. In realized studies about Hybrid PSO-NN, position and velocity boundaries of weight and bias are chosen equal or set free in space which leave the performance of PSO-NN in suspense. In this paper, the limitations of weight velocity (wv), weight position (wp), bias velocity (bv) and bias position (bp) are diversely changed and their effects on the output of hybrid structure are examined. Concerning this, the formed structure is called as Bounded PSO-NN on account of adjusting the optimum operating conditions (intervals). On performance evaluation, proposed method is tested on binary and multiclass pattern classification by using six medical datasets: Wisconsin Breast Cancer (WBC), Pima Indian Diabetes (PID), Bupa Liver Disorders (BLD), Heart Statlog (HS), Breast Tissue (BT) and Dermatology Data (DD). Upon analyzing the results, it was revealed that Bounded PSO-NN has a faster processing time than general PSO-NNs in which set-free and wpi=bpi and wvi=bvi conditions are settled. The superiority in terms of processing time is about 199s (set-free) and 307s (wpi=bpi and wvi=bvi) for training, about 16ms (set-free) and 9ms (wpi=bpi and wvi=bvi) for test. In terms of classification performance, PSO-NN (set-free condition), PSO-NN (wpi=bpi & wvi=bvi) and PSO-NN with individual boundary adjustment (bounded PSO-NN) respectively achieves to accuracy rates as 69.84%, 95.31% and 97.22% on WBC, 47.01%, 76.69% and 77.73% on PID, 55.36%, 67.54% and 73.91% on BLD, 64.82%, 81.48% and 85.56% on HS, 75%, 92.31% and 100% on BT, 27.47%, 92.31% and 100% on DD. In the light of experiments, it is seen that Bounded PSO-NN is better than general PSO-NNs for obtaining the optimum results. Consequently, the importance of limitations is clarified and it is proven that each limitation must be adjusted individually, not be set free or not be chosen equal.

Scout Particle Swarm Optimization

Particle Swarm Optimization is a robust optimization algorithm proved itself in various technical areas like training of classifiers, image classification and function optimization, etc. It simulates the foraging behaviour of bird swarms. While doing that, it uses velocity and position metrics for directing its particles to food. Concerning this, it has various advantages like high convergence, speedy process capability and a few parameters to be adjusted. But it has a significant disadvantage restricting the performance. This handicap is regeneration of the particle which couldn’t improve itself along iterations. Moreover, Artificial Bee Colony Optimization (ABC) is a valuable optimization algorithm imitating the foraging behaviour like PSO. However, ABC uses honey bees grouped as employed bees, onlooker bees and scout bees. The employed bee and onlooker bee phases do the same work with velocity and position concepts in PSO. But, scout bee phase regenerates the useless particles in order to achieve higher performance by upgrading diversity. Therefore, it’s seen that addition of scout bee phase into PSO looks like a smart idea. So, in this study, Scout PSO (ScPSO) algorithm is designed which is more effective and useful than PSO. For performance analysis of ScPSO, it was used in training of NN classifier. Furthermore, ScPSO-NN is compared with NN and PSO-NN methods on medical pattern classification. For this purpose, Wisconsin Breast Cancer-Original (WBC), Pima Indian Diabetes (PID), Heart Statlog (HS) and Bupa Liver Disorders (BLD) datasets are used and test process is realized by 10-fold cross validation method. As a result, ScPSO-NN achieves classification accuracies as 97.51% (WBC), 78.13% (PID), 86.30% (HS) and 75.07% (BLD).

An Efficient Pipeline for Abdomen Segmentation in CT Images

Computed tomography (CT) scans usually include some disadvantages due to the nature of the imaging procedure, and these handicaps prevent accurate abdomen segmentation. Discontinuous abdomen edges, bed section of CT, patient information, closeness between the edges of the abdomen and CT, poor contrast, and a narrow histogram can be regarded as the most important handicaps that occur in abdominal CT scans. Currently, one or more handicaps can arise and prevent technicians obtaining abdomen images through simple segmentation techniques. In other words, CT scans can include the bed section of CT, a patient’s diagnostic information, low-quality abdomen edges, low-level contrast, and narrow histogram, all in one scan. These phenomena constitute a challenge, and an efficient pipeline that is unaffected by handicaps is required. In addition, analysis such as segmentation, feature selection, and classification has meaning for a real-time diagnosis system in cases where the abdomen section is directly used with a specific size. A statistical pipeline is designed in this study that is unaffected by the handicaps mentioned above. Intensity-based approaches, morphological processes, and histogram-based procedures are utilized to design an efficient structure. Performance evaluation is realized in experiments on 58 CT images (16 training, 16 test, and 26 validation) that include the abdomen and one or more disadvantage(s). The first part of the data (16 training images) is used to detect the pipeline’s optimum parameters, while the second and third parts are utilized to evaluate and to confirm the segmentation performance. The segmentation results are presented as the means of six performance metrics. Thus, the proposed method achieves remarkable average rates for training/test/validation of 98.95/99.36/99.57% (jaccard), 99.47/99.67/99.79% (dice), 100/99.91/99.91% (sensitivity), 98.47/99.23/99.85% (specificity), 99.38/99.63/99.87% (classification accuracy), and 98.98/99.45/99.66% (precision). In summary, a statistical pipeline performing the task of abdomen segmentation is achieved that is not affected by the disadvantages, and the most detailed abdomen segmentation study is performed for the use before organ and tumor segmentation, feature extraction, and classification.

A Novel Pipeline for Adrenal Tumour Segmentation

BACKGROUND AND OBJECTIVE Adrenal tumours occur on adrenal glands surrounded by organs and osteoid. These tumours can be categorized as either functional, non-functional, malign, or benign. Depending on their appearance in the abdomen, adrenal tumours can arise from one adrenal gland (unilateral) or from both adrenal glands (bilateral) and can connect with other organs, including the liver, spleen, pancreas, etc. This connection phenomenon constitutes the most important handicap against adrenal tumour segmentation. Size change, variety of shape, diverse location, and low contrast (similar grey values between the various tissues) are other disadvantages compounding segmentation difficulty. Few studies have considered adrenal tumour segmentation, and no significant improvement has been achieved for unilateral, bilateral, adherent, or noncohesive tumour segmentation. There is also no recognised segmentation pipeline or method for adrenal tumours including different shape, size, or location information. METHODS This study proposes an adrenal tumour segmentation (ATUS) pipeline designed to eliminate the above disadvantages for adrenal tumour segmentation. ATUS incorporates a number of image methods, including contrast limited adaptive histogram equalization, split and merge based on quadtree decomposition, mean shift segmentation, large grey level eliminator, and region growing. RESULTS Performance assessment of ATUS was realised on 32 arterial and portal phase computed tomography images using six metrics: dice, jaccard, sensitivity, specificity, accuracy, and structural similarity index. ATUS achieved remarkable segmentation performance, and was not affected by the discussed handicaps, on particularly adherence to other organs, with success rates of 83.06%, 71.44%, 86.44%, 99.66%, 99.43%, and 98.51% for the metrics, respectively, for images including sufficient contrast uptake. CONCLUSIONS The proposed ATUS system realises detailed adrenal tumour segmentation, and avoids known disadvantages preventing accurate segmentation.

Classification of adrenal lesions by bounded PSO-NN

Adrenal glands are the organs at which vitally important hormones are released. In adrenal glands, different kind of benign and malign lesions can arise. Herein, Dynamic Computed Tomography (dynamic CT) is the most used scan type for definition of lesion types. On the events that dynamic CT underwhelms, biopsy process is performed which is difficultly implemented because of the location of adrenal glands. During biopsy process, different complications can happen since adrenals glands are surrounded by spleen, lung, etc. At this point, a decision support system is needed for helping to medical experts. In this study, a Region of Interest (ROI) is defined that includes adrenal lesions. After that, feature extraction is realized by using Gray-Level Co-Occurance Matrix (GLCM) and the second-order statistics. At classification part, Neural Network (NN) and a novel approach including NN (Bounded PSO-NN) are evaluated by utilizing from three performance metrics. As a result, its confirmed that Bounded PSO-NN classifies the malign and benign patterns more accurately which obtained by analysis taken from ROI.

Elimination of white Gaussian noise in arterial phase CT images to bring adrenal tumours into the forefront

Dynamic Contrast-Enhanced Computed Tomography (DCE-CT) is applied to observe adrenal tumours in detail by utilising from the contrast matter, which generally brings the tumour into the forefront. However, DCE-CT images are generally influenced by noises that occur as the result of the trade-off between radiation doses vs. noise. Herein, this situation constitutes a challenge in the achievement of accurate tumour segmentation. In CT images, most of the noises are similar to Gaussian Noise. In this study, arterial phase CT images containing adrenal tumours are utilised, and elimination of Gaussian Noise is realised by fourteen different techniques reported in literature for the achievement of the best denoising process. In this study, the Block Matching and 3D Filtering (BM3D) algorithm typically achieve reliable Peak Signal-to-Noise Ratios (PSNR) and resolves challenges of similar techniques when addressing different levels of noise. Furthermore, BM3D obtains the best mean PSNR values among the first five techniques. BM3D outperforms to other techniques by obtaining better Total Statistical Success (TSS), CPU time and computation cost. Consequently, it prepares clearer arterial phase CT images for the next step (segmentation of adrenal tumours).

Optic Disc Segmentation with Kapur-ScPSO Based Cascade Multithresholding

The detection of significant retinal regions (segmentation) constitutes an indispensible need for computer aided diagnosis of retinal based diseases. At this point, image segmentation algorithm is wanted to be quick in order to spare time for feature selection and classification parts. In this paper, we deal with the fast and accurate segmentation process of optic discs in retinal images. For this purpose, a cascade multithresholding (CMT) process is proposed by a novel optimization algorithm (Scout Particle Swarm Optimization) and an efficient cost function (Kapur).

ScPSO-Based Multithresholding Modalities for Suspicious Region Detection on Mammograms

Abstract Multithresholding techniques typically use histogram information for finding the optimal thresholds on image enhancement, segmentation, etc. In optimization-based modalities, a robust optimization algorithm is needed to realize an effective multithresholding. At this point, the robustness of an optimization algorithm is revealed by the obtained output of cost function (objective values) to be maximized or minimized. In this chapter, we handle the design of novel and stochastic multithresholding modalities that are proposed by an effective optimization algorithm named Scout Particle Swarm Optimization (ScPSO). For the aim of best gray-level distribution (optimum multithresholding), Otsu and Kapur functions are preferred in optimization algorithm owing to their popularity and efficiency. In experiments, Otsu-ScPSO and Kapur-ScPSO are compared together beside the comparison of four optimization algorithms (Particle Swarm Optimization, Genetic Algorithm, Bacterial Foraging Algorithm, and Modified Bacterial Foraging Algorithm), which previously proved themselves in segmentation of well-known benchmark images. According to results, its seen that ScPSO-based modalities achieve better objective values and standard deviations than other optimization algorithms. In terms of being the best optimization-based approaches, Otsu-ScPSO and Kapur-ScPSO are compared by using computation time, standard deviation, PSNR, and SSIM metrics. For a real-time application, Otsu-ScPSO and Kapur-ScPSO are compared for the purpose of suspicious region detection (mass segmentation) on 15 mammogram images taken from mini-MIAS database. Consequently, its revealed that the segmentation of breast masses and the multithresholding of benchmark images are better performed using the optimum thresholds found by Kapur-ScPSO.