Program Information
Noise Suppression in Multi-Material Decomposition of Dual Energy CT Using Boundary Discrimination Regularization
Y Xue , Y Jiang , C Yang*, T Niu , Zhejiang University, Hangzhou, Zhejiang
Presentations
WE-DE-605-2 (Wednesday, August 2, 2017) 10:15 AM - 12:15 PM Room: 605
Purpose: The dual-energy CT (DECT) plays an increasingly important role in in clinical practice because of its material differentiation capability, such as liver-fat quantification, lung cancer disease and etc. Nevertheless, the magnified noise severely degrades the decomposed image quality, which limits the quantitative DECT application. In this work, we propose a multi-material decomposition (MMD) noise suppression method using boundary discrimination regularization to tackle the above problem.
Methods: Volume and mass conservation and three-material in each pixel assumption are introduced to as constrain condition. A least-square estimation using smooth regularization is used to suppress the noise. We design the regularization term based on the fact that the pixel values are relatively uniform inside the same material region and abrupt around the material boundary. Thus, the noise suppression inside a material is relatively stronger than that in the neighborhood of material boundary. The method is evaluated on the contrast rod slice of Catphan©600 phantom with the scanned energy of 75kVp and 125kVp.
Results: The proposed method decreases the overall STD of different material image by 74% as compared with that using direct inversion.
Conclusion: We propose a multi-material decomposition noise suppression method using a boundary discrimination regularization. The proposed method applied a boundary discrimination regularization term to suppress the noise and maintain the boundary at the same time. The proposed multi-material images are faithfully decomposed from the dual-energy measurements. It is thus practical to be implemented in clinical applications.
Funding Support, Disclosures, and Conflict of Interest: Zhejiang Provincial Natural Science Foundation of China (Grant No. LR16F010001) National High-tech R&D Program for Young Scientists by the Ministry of Science and Technology of China (863 Program, 2015AA020917)
Contact Email: