G. R. Sinha

An Adaptive K-means Clustering Algorithm for Breast Image Segmentation

Breast cancer is one of the major causes of death among women. Small clusters of micro calcifications appearing as collection of white spots on mammograms show an early warning of breast cancer. Early detection performed on X-ray mammography is the key to improve breast cancer diagnosis. In order to increase radiologist’s diagnostic performance, several computer-aided diagnosis (CAD) schemes have been developed to improve the detection of primary identification of this disease. In this paper, an attempt is made to develop an adaptive k-means clustering algorithm for breast image segmentation for the detection of micro calcifications and also a computer based decision system for early detection of breast cancer. The method was tested over several images of image databases taken from BSR APPOLO for cancer research and diagnosis, India. The algorithm works faster so that any radiologist can take a clear decision about the appearance of micro calcifications by visual inspection of digital mammograms and detection accuracy has also improved as compared to some

Early Detection of Breast Cancer using Self Similar Fractal Method

Breast cancer is one of the major causes of death among women. Small clusters of micro calcifications appearing as collection of white spots on mammograms show an early warning of breast cancer. Early detection performed on X-ray mammography is the key to improve breast cancer diagnosis. Image segmentation consists in finding the characteristic entities of an image, either by their contours (edges) or by the region they lie in. Our aim in this paper is to present a method for medical image enhancement based on the well established concept of fractal derivatives and selecting image processing techniques like segmentation of an image with self similar properties. The concept of a fractal is most often associated with geometrical objects satisfying two criteria: self-similarity and fractional dimensionality. The method was tested over several images of image databases taken from BSR APPOLO for cancer research and diagnosis, India.

Mammography Feature Analysis and Mass Detection in Breast Cancer Images

This paper introduces a novel approach for accomplishing mammographic feature analysis through detection of tumor, in terms of their size and shape with experimental work for early breast tumor detection. The objective is to detect the abnormal tumor/tissue inside breast tissues using three stages: Preprocessing, Segmentation and post processing stage. By using preprocessing noise are remove and than segmentation is applied to detect the mass, after that post processing is applied to find out the benign and malignant tissue with the affected area in the cancers breast image. Size of tumor is also detected in these steps. The occurrences of cancer nodules are identified clearly. Compared with an expert observer reading the Mammography, our algorithm achieves 96.5% sensitivity, 89% specificity, 95.6% accuracy value.

Isolated Handwritten Devnagri Numeral Recognition Using HMM

This paper describes a complete system for the recognition of isolated handwritten Devnagri numerals using Hidden-Markov model (HMM). The HMM has the property that its states are not defined as a priory information, but are determined automatically based on a database of handwritten numerals images. In this work the image database consist of 500 images of handwritten Devnagri characters from 50 different writers. Before extracting the features, the images are normalized using image isometrics such as translation, rotation and scaling. An automatic system trained 400 images of image database and numeral model form with multivariate Gaussian state conditional distribution. A separate set of 100 characters was used to test the system. The recognition accuracy for individual numerals varies from 30% to 100% for N=3 and 80% to 100% for N=5.