VISUALIZATION OF RISK IN BREAST CANCER USING INTELLIGENT TECHNIQUES

Abstract

Breast cancer is the leading malignancy affecting women in the world. The present study emphasizes to develop a method for the detection of cancer tumors/lesions in breast at an early stage by various image analysis techniques and different models using Statistical Analysis. Health models are used to assess the risk factors for breast cancer using retrospective data in which, a multivariate method Binary Logistic Regression was used to understand the influence of pre menopause and post menopause and socioeconomic parameters for the occurrence of breast cancer. The second, a Fuzzy Expert System is used to resolve the prognostic decision making in the seriousness of the breast cancer. The cost associated for screening varies from country to country and also the availability of experts to read the Images and diagnose is a concern. To overcome such problems, the study has developed a non-invasive method through the aide of computer to automatically detect the Region of Interest (ROI) in the Mammogram by applying Mandelbrot s Fractal Dimension on Mammogram to differentiate into different types conditions of Breast cancer. The cell line study was used observe Morphological changes of individual cells after different concentrations of Raloxifene treatment on MCF-7 cell line using Fractal Dimension. Wavelet Analysis was used to diagnose the Normal and Calcified mammographic images. The Images were subjected to 2D-Wavelet (Haar) in MatLab environment for segregation. Irrespective of modern diagnosing method, treatment, etc. in modern medical science, there are plenty of newly devised methodologies and techniques for the timely detection of breast cancer. Most of these techniques make use of highly advanced technologies such as medical image processing. The present study is an attempt to highlight the available breast cancer detection techniques based on image processing which is non-invasive tool and in addition to enumerate the risk factors associated with breast cancer using retrospective data.