Cardiac CT Image Enhancement for 3D Heart Registration and Visualization

Cardiac CT imaging has been established as one of the most valuable tools in diagnosis and treatment of cardio vascular diseases. Modern CT imaging devices provide a significant amount of data with high precision. However, resulting CT images can often suffer from significant quality degradations, especially in the preferable case of low level of radiation. Such degradations can complicate the registration and visualization processes, thus making it harder for medical doctors to analyze the acquired data. The goal of this paper is to propose a method for the enhancement of input CT data based on Multiscale Wavelet Decomposition and Spatial Pixel Profiling in 2D slices. Proposed methods are used as preprocessing steps on input CT data, preparing the data for segmentation and registration using Active Contour Models segmentation method. This paper displays the benefits of such approach for the purpose of 3D reconstruction.