Wavelet Fusion in DSA Based on Dynamic Fuzzy Data Model

Abstract

Digital subtraction angiography (DSA) is a type of medical imaging technique which can eliminate the interferential background and give prominence to blood vessels by computer processing. By analyzing the characters of DSA medical image sequence, we find angiograms are taken by the same instrument in same scene at different time, so the DSA serial images are present different parts of blood vessels. This paper proposes a novel method for this particular image fusion, which is using discrete wavelet transform improved by dynamic fuzzy data model. Firstly, we decompose each of DSA images of by 2-D discrete wavelet transform. Then we construct a membership function based on dynamic fuzzy data model to optimize the performance of wavelet coefficients selection in order to combine all series of DSA images in different level. At last, we use wavelet reconstruction to synthesize one DSA medical image which could contain more integrated accurate detail information of blood vessels. By contrast, the efficiency of our method is better than weighted average, Laplacian pyramid and traditional wavelet transform method. It is helpful to medical imaging aid diagnosis.