Application of Wavelet Thresholding De-noising in DSA

Digital subtraction angiography (DSA) is a type of medical imaging technique used in interventional radiology, which can eliminate the interferential background and give prominence to blood vessels by computer processing. By analyzing the characteristics of DSA medical images, this paper proposes a new solution for this particular image de-noising, which is using improved soft thresholding. In order to obtain the different level information, we decompose the images of angiogram and subtraction by 2-D discrete wavelet transform. Then we use our new thresholding de-noising approach which can remove more noise and reserve more details. At last, we use wavelet reconstruction to get the new DSA medical image which can display the blood vessels much more clearly. By contrast, the efficiency of our method is better than median filtering and Wiener filtering approaches.