A NOVEL APPROACH TO MASS ABNORMALITY DETECTION IN MAMMOGRAPHIC IMAGES

摘要

Masses are important indication of breast cancer. Mass abnormality detection is a very difficult task amongst mammographic image analysis. In this paper we propose a novel approach to feature extraction and classification for mass abnormality detection in digital mammograms. Fractal Hausdorff dimension is used to characterise the texture feature of mammographic images. It has been shown that fractal dimension correlates strongly with human observers' subjective rankings of image texture. Support vector machine (SVM), a novel type of learning machine based on statistical learning theory, is trained through supervised learning to detect mass abnormality. The proposed method distinguishes mass abnormality from normal background tissue, achieving 91.1% correct classification. The method presented here provides a promising tool for mass abnormality detection in mammographic image analysis.