IMAGE RESTORATION ON FUSION OF MAMMOGRAPHS AND MRI BREAST IMAGES USING DUAL TREE COMPLEX WAVELET TRANSFORM

Abstract

The developments in the medical field is remarkable in recent days due to the innovation in medicines, diagnosis of diseases at an early stage and computer aided applications for identifying the exact location of the affected part. The disease is also increasing day by day along with the medical developments. In those diseases, the most life threatening diseases is the cancer. In olden days, the cancer is mostly of blood cancer but there are different types of cancers like blood, stomach, colon, cervical and breast cancer. Among these cancers, the less life threatening is the breast cancer, which can be cured by medicine at early stage and surgery in the final stage of the cancer. This work is to improve the diagnosis of breast cancer by fusion method. The breast cancer is diagnosed using MRI scan and X-ray. These image modalities have different information of the breast. The previous work used the registration approach on these modalities to improve the information of the cancer from the images using intensity gradient approach and scalar gradient approach. In this proposed work, the images will be fuse to extract both the lesion and inter bone of the breast. The fusion process will perform by pre-processing steps like gray-scale conversion, and noise removal process. The pre-processed image will be fuse by using the dual tree complex wavelet transform. It has both the MRI and X-ray information of the breast. The geometric correction is applied to the fused image for restoration. This fused image is extracted with more information/property. The fused image will be evaluated with the help of following metrics like entropy, mean and variance to ensure the information and image quality is preserved in the fusion approach.