Detection of Hard Exudates in Colour Fundus Images Using Fuzzy Support Vector Machine-Based Expert System.

Diabetic retinopathy is a major cause of vision loss in diabetic patients. Currently, there is a need for making decisions using intelligent computer algorithms when screening a large volume of data. This paper presents an expert decision-making system designed using a fuzzy support vector machine (FSVM) classifier to detect hard exudates in fundus images. The optic discs in the colour fundus images are segmented to avoid false alarms using morphological operations and based on circular Hough transform. To discriminate between the exudates and the non-exudates pixels, colour and texture features are extracted from the images. These features are given as input to the FSVM classifier. The classifier analysed 200 retinal images collected from diabetic retinopathy screening programmes. The tests made on the retinal images show that the proposed detection system has better discriminating power than the conventional support vector machine. With the best combination of FSVM and features sets, the area under the receiver operating characteristic curve reached 0.9606, which corresponds to a sensitivity of 94.1% with a specificity of 90.0%. The results suggest that detecting hard exudates using FSVM contribute to computer-assisted detection of diabetic retinopathy and as a decision support system for ophthalmologists.

Computer-aided detection of breast cancer on mammograms: a swarm intelligence optimized wavelet neural network approach.

Breast cancer is the second leading cause of cancer death in women. Accurate early detection can effectively reduce the mortality rate caused by breast cancer. Masses and microcalcification clusters are an important early signs of breast cancer. However, it is often difficult to distinguish abnormalities from normal breast tissues because of their subtle appearance and ambiguous margins. Computer aided diagnosis (CAD) helps the radiologist in detecting the abnormalities in an efficient way. This paper investigates a new classification approach for detection of breast abnormalities in digital mammograms using Particle Swarm Optimized Wavelet Neural Network (PSOWNN). The proposed abnormality detection algorithm is based on extracting Laws Texture Energy Measures from the mammograms and classifying the suspicious regions by applying a pattern classifier. The method is applied to real clinical database of 216 mammograms collected from mammogram screening centers. The detection performance of the CAD system is analyzed using Receiver Operating Characteristic (ROC) curve. This curve indicates the trade-offs between sensitivity and specificity that is available from a diagnostic system, and thus describes the inherent discrimination capacity of the proposed system. The result shows that the area under the ROC curve of the proposed algorithm is 0.96853 with a sensitivity 94.167% of and specificity of 92.105%.

A swarm optimized neural network system for classification of microcalcification in mammograms.

Early detection of microcalcification clusters in breast tissue will significantly increase the survival rate of the patients. Radiologists use mammography for breast cancer diagnosis at early stage. It is a very challenging and difficult task for radiologists to correctly classify the abnormal regions in the breast tissue, because mammograms are noisy images. To improve the accuracy rate of detection of breast cancer, a novel intelligent computer aided classifier is used, which detects the presence of microcalcification clusters. In this paper, an innovative approach for detection of microcalcification in digital mammograms using Swarm Optimization Neural Network (SONN) is used. Prior to classification Laws texture features are extracted from the image to capture descriptive texture information. These features are used to extract texture energy measures from the Region of Interest (ROI) containing microcalcification (MC). A feedforward neural network is used for detection of abnormal regions in breast tissue is optimally designed using Particle Swarm Optimization algorithm. The proposed intelligent classifier is evaluated based on the MIAS database where 51 malignant, 63 benign and 208 normal images are utilized. The approach has also been tested on 216 real time clinical images having abnormalities which showed that the results are statistically significant. With the proposed methodology, the area under the ROC curve (A ( z )) reached 0.9761 for MIAS database and 0.9138 for real clinical images. The classification results prove that the proposed swarm optimally tuned neural network highly contribute to computer-aided diagnosis of breast cancer.

An improved decision support system for detection of lesions in mammograms using Differential Evolution Optimized Wavelet Neural Network.

In this paper, a computerized scheme for automatic detection of cancerous lesion in mammograms is examined. Breast lesions in mammograms are an area with an abnormality or alteration in the breast tissues. Diagnosis of these lesions at the early stage is a very difficult task as the cancerous lesions are embedded in normal breast tissue structures. This paper proposes a supervised machine learning algorithm - Differential Evolution Optimized Wavelet Neural Network (DEOWNN) for detection of tumor masses in mammograms. Differential Evolution (DE) is a population based optimization algorithm based on the principle of natural evolution, which optimizes real parameters and real valued functions. By utilizing the DE algorithm, the parameters of the Wavelet Neural Network (WNN) are optimized. To increase the detection accuracy a feature extraction methodology is used to extract the texture features of the abnormal breast tissues and normal breast tissues prior to classification. Then DEOWNN classifier is applied at the end to determine whether the given input data is normal or abnormal. The performance of the computerized decision support system is evaluated using a mini database from Mammographic Image Analysis Society (MIAS). The detection performance is evaluated using Receiver Operating Characteristic (ROC) curves. The result shows that the proposed algorithm has a sensitivity of 96.9% and specificity of 92.9%.