Volume 7 Number 4 March 2017


Performance Analysis of Color Images Using Thresholding Techniques for Image Segmentation in Ziehl- Neelsen Sputum Slide Images
D. Nithiyapriya, I. Laurence Aroquiaraj

Abstract- Image segmentation is significant in investigation of clinical images. In this way, viable and accurate techniques are expected to get right diagnosis of quantitative clinical samples. This paper examines the comparison among color thresholding and extensive thresholding methods for Ziehl-Neelsen TB bacilli slide images in sputum samples. Color Thresholding utilizes RGB as input image while Global Thresholding used YCBr as input image. Also we applied Bayesian segmentation to predict the probability of a pixel depicting a ‘TB entity’ by make use of knowledge of Ziehl-Neelsen stain colours and shape size analysis. This technique is testing on N number of different images. The sensitivity and specificity of all tested classifiers is above 99% for the identification of bacillus objects represented by features. Our results may be used to reduce technician involvement in screening for TB, and would be particularly useful in laboratories in countries with a high burden of TB, where, typically, ZN rather than Auramine staining of sputum smears is the method of choice. The experiment results demonstrated the efficiency of the proposed system.


Performance and Classification Evaluation of J48 Algorithm and Kendall’s Based J48 Algorithm (KNJ48)
N. Saravanan, V. Gayathri

Abstract- We have been using the most popular algorithm J48 for classification of data. The J48 algorithm is used to classify different applications and perform accurate results of the classification. J48 algorithm is one of the best machine learning algorithms to examine the data categorically and continuously. When it is used for instance purpose, it occupies more memory space and depletes the performance and accuracy in classifying medical data. Our proposed method is to measure the improved performance and produce higher rate of accuracy. For this research, the dengue dataset was collected from various government hospitals in Krishnagiri District. To measure the entropy of information and to identify the dataset and to increase the accuracy of J48 algorithm, the entropy of J48 is modified with Kendall’s Rank Correlation Coefficient algorithm (KNJ48) to improve the accuracy of classification and performance time. Thus, it is modified as Kendall’s New Rank Correlation Coefficient J48 algorithm (KNJ48) for better performance.


Kidney Stone Detection using Contrast Limited Adaptive Histogram Equalization (CLAHE) on CT Scan Images
L. Prisilla, I. Laurance Aroqiaraj

Abstract- In modern days need for health services are increasing and the demand for making the Computer-aided medical diagnosis is increasing. The development of imaging techniques, diagnosis using Computed Tomography (CT) images has become widespread because of its low cost, reliable and noninvasive procedure. Feature extraction, analysis, and pattern recognition techniques for these images are used for finding the abnormality like a tumor, cyst, stone etc. Kidney-Urine-Belly Computed Tomography (KUB CT) investigation is an imaging modality that has the potential to improve kidney stone screening and prognosis. This research focuses on efficient computer-aided medical diagnosis from KUB CT kidney images using Contrast Limited Adaptive Histogram Equalization (CLAHE). This image is first removed from Computer-aided medical diagnosis integrates computer science, image processing, pattern recognition and AI techniques and its performance depends on some factors like segmentation, feature selection, reference database size, computational efficiency, etc. Computer-Aided Diagnosis (CAD) is a technique in medicine that helps doctors in the interpretation of medical images. Imaging techniques in mammography, Computerized Tomography (CT), Magnetic Resonance Imaging (MRI), Xray and Ultrasound (US) diagnostics yields lots of information, which the radiologist has to analyze and evaluate comprehensively in short time. The advance developments made in the field of information technology and medical imaging, there has been a tremendous need for the ability to create speckle noise using Non-Sub sampled CT after which objective diagnosis is made using our proposed technique. The experiment results demonstrated that the work has 87.5% accuracy, which predicts the program’s potential is diagnostic effective for kidney stone detection.


Markov Decision on Data Backup Scheduling for Big Data
K. Madasamy, C. Elango, M. Ramaswami

Abstract- Surfing over the data deluge is an unavoidable phenomena of the day, managing and protecting the massive amount of data is a vital task. Data backup operation is one of the most essential areas of Information Technology. In this article, we study a backup processing model using Markov Decision Process (MDP) for dealing with Big data. As backup storage operation of huge voluminous of data is a tedious task, we scheduled this process optimally using the versatile tool Markov Decision Process (MDP). A Numerical example is provided to illustrate the suggested data processing and backup model a viable one.


Classification of Gastric Carcinoma using Elman Neural network and Autoencoder
P. K. Shanthakumaar, K. Thangavel, D. Arul Pon Daniel

Abstract- In medical diagnosis, Breath analysis is one of the non-invasive methods of gaining information of the clinical state of the individual through the exhale breath. Biomarkers play a potential significance in disease diagnosis, thus identification and qualification of the diagnosis is the driving force of the analysis of the exhale breath. Carcinoma a type of cancer that occur on the skin or the tissue lining the organs such as liver, kidney, etc, Gastric Carcinoma occurs in the inner lining of the stomach. The proposed work is to earlier detection of gastric carcinoma at the earlier stage, since it has no significant symptoms. In experimental results Autoencoder produce better results compared with Elman Neural Network in term of statistical accuracy and Mean Square Error.