Rough Set Approach in Ultrasound Biomicroscopy Glaucoma Analysis

2005 IEEE Instrumentationand Measurement Technology Conference Proceedings, 2005

In closed angled Glaucoma, fluid pressure in the eye increases because of inadequate fluid flow between the iris and the cornea. One important technique to assess patients at risk of glaucoma is to analyze ultrasound images of the eye to detect the structural changes. We propose an algorithm to automatically identify important features in the ultrasound images. Tests were performed by comparison of results with eighty images of glaucoma patients against the landmarks identified by a technologist. A success rate of 97% was achieved. This work will help improving the efficiency of clinical interpretation of ultrasound images of the eye.

Image processing is a special form of signal processing which provides valuable information toward human image. The most important image processing step in human picture recognition system consists of Edge enhancement process. In this paper we propose a new idea for edge enhancement using smoothing filters based on Rough sets. Rough set theory is an important tool (it is a mathematical tool) that process uncertainty and non-integrity[1][10]. This theory is efficient in analysing and processing inconsistent information, and then detecting connotative information. Rough set theory is to classify the pixel in each part then noise pixels can be separated and removed, which is to providèbetter' input for other automated image processing techniques. Smoothing is often used to reduce noise within an image or to produce a less pixel image. Edges are the representations of the discontinuities of image intensity functions. For processing these discontinuities in an image a good edge enhancement technique is essential. This paper proposes a new idea for enhancement of a medical image using smoothing techniques and Image noise is mostly unwanted and manifested in the pixels of an image and the main application of image averaging is noise removal. In this paper, we deal with Rough set based medical image smoothening filters in order to improve the quality of the image and as well as to help to solve various complex image processing tasks in the future. This paper focuses on an approach which tries to combine the advantages of the various smoothing filters techniques.

International Journal of Hybrid Intelligent Systems, 2007

The objective of this paper is to introduce a rough neural intelligent approach for rule generation and image classification. Hybridization of intelligent computing techniques has been applied to see their ability and accuracy to classify breast cancer images into two outcomes: malignant cancer or benign cancer. Algorithms based on fuzzy image processing are first applied to enhance the contrast of the whole original image; to extract the region of interest and to enhance the edges surrounding that region. Then, we extract features characterizing the underlying texture of the regions of interest by using the gray-level co-occurrence matrix. Then, the rough set approach to attribute reduction and rule generation is presented. Finally, rough neural network is designed for discrimination of different regions of interest to test whether they represent malignant cancer or benign cancer. Rough neural network is built from rough neurons, each of which can be viewed as a pair of sub-neurons, corresponding to the lower and upper bounds. To evaluate performance of the presented rough neural approach, we run tests over different mammogram images. The experimental results show that the overall classification accuracy offered by rough neural approach is high compared with other intelligent techniques.

International Journal of Biomedical Imaging

Horizontal and vertical cup to disc ratios are the most crucial parameters used clinically to detect glaucoma or monitor its progress and are manually evaluated from retinal fundus images of the optic nerve head. Due to the rarity of the glaucoma experts as well as the increasing in glaucoma’s population, an automatically calculated horizontal and vertical cup to disc ratios (HCDR and VCDR, resp.) can be useful for glaucoma screening. We report on two algorithms to calculate the HCDR and VCDR. In the algorithms, level set and inpainting techniques were developed for segmenting the disc, while thresholding using Type-II fuzzy approach was developed for segmenting the cup. The results from the algorithms were verified using the manual markings of images from a dataset of glaucomatous images (retinal fundus images for glaucoma analysis (RIGA dataset)) by six ophthalmologists. The algorithm’s accuracy for HCDR and VCDR combined was 74.2%. Only the accuracy of manual markings by one opht...

Pattern Recognition Letters, 2003

We present applications of rough set methods for feature selection in pattern recognition. We emphasize the role of the basic constructs of rough set approach in feature selection, namely reducts and their approximations, including dynamic reducts. In the overview of methods for feature selection we discuss feature selection criteria, including the rough set based methods. Our algorithm for feature selection is based on an application of a rough set method to the result of principal components analysis (PCA) used for feature projection and reduction. Finally, the paper presents numerical results of face and mammogram recognition experiments using neural network, with feature selection based on proposed PCA and rough set methods.

Journal of Biomedical Imaging, 2009

Journal of Intelligent Systems, 2004

In this paper, we present an enhanced rough set approach for attribute reduction and generating classification rules from digital mammogram datasets. For this purpose, the presented approach is described in a hierarchical fashion. First, the preprocessing phase is adopted to enhance the contrast and edges of the mammogram images; moreover image processing segmentation algorithm is used to extract the region of interest. In the next phase, five texture features from the co-occurrence matrix are extracted and represented in attribute vector, and the reducts with minimal number of attributes are extracted. In the third phase, the decision rules within the generated reduct sets are generated. In the last phase, the classifier model was built and quadratic distances similarly function is used for matching process. To evaluate the validity of the rules based on the approximation quality of the attributes, we introduce a statistical test to evaluate the significance of the rules. The experimental results show that the classification algorithm performs well, reaching over 93% in accuracy with less number of rules compared with a well-known decision trees and neural network classifier models.

Physics Procedia, 2015

This paper addresses the use of computer-aided diagnosis (CAD) system for the cataract classification based on ultrasound technique. Ultrasound A-scan signals were acquired in 220 porcine lenses. B-mode and Nakagami images were constructed. Ninety-seven parameters were extracted from acoustical, spectral and image textural analyses and were subjected to feature selection by Principal Component Analysis (PCA). Bayes, K Nearest-Neighbors (KNN), Fisher Linear Discriminant (FLD) and Support Vector Machine (SVM) classifiers were tested. The classification of healthy and cataractous lenses shows a good performance for the four classifiers (F-measure ≥92.68%) with SVM showing the highest performance (90.62%) for initial versus severe cataract classification.

An Accurate, Fast and Noise-Adaptive segmentation of Brain MR Images for clinical Analysis is a challenging problem. An improved Hybrid Clustering Algorithm is presented here, which integrates the concept of recently popularized Rough Sets and that of Fuzzy Sets. The concept of lower and upper approximations of rough sets is incorporated to handle uncertainty, vagueness, and incompleteness in class definition. For making the segmentation robust to Noise and intensity inhomogeneity, the images are proposed to be pre-processed with a neighbourhood averaging spatial filter. To accelerate the segmentation process, a novel Suppressed Rough Fuzzy C-Means model is presented in which a membership suppression mechanism has been implemented, which creates competition among clusters to speed-up the clustering process. The effectiveness of the presented algorithm along with comparison with other related algorithm has been demonstrated on a set of MR and CT scan images. The results using MRI data show that our method provides better results compared to standard Fuzzy C-Means based algorithms and other modified similar techniques.