A novel kernel Wasserstein distance on Gaussian measures: An application of identifying dental artifacts in head and neck computed tomography. (May 2020)

• Record Type: Journal Article
• Authors: Oh, Jung Hun; Pouryahya, Maryam; Iyer, Aditi; Apte, Aditya P.; Deasy, Joseph O.; Tannenbaum, Allen
• Is Part Of: Computers in biology and medicine. Volume 120(2020)
• Publication Date: 2020-05
• Subjects: Kernel Wasserstein distance -- Kernel Kullback–Leibler divergence -- Kernel trick -- Reproducing kernel Hilbert space; Medicine -- Data processing -- Periodicals; Biology -- Data processing -- Periodicals; 610.285
• DOI: 10.1016/j.compbiomed.2020.103731
• Languages: English
• ISSNs: 0010-4825

Abstract

Abstract: The Wasserstein distance is a powerful metric based on the theory of optimal mass transport. It gives a natural measure of the distance between two distributions with a wide range of applications. In contrast to a number of the common divergences on distributions such as Kullback–Leibler or Jensen–Shannon, it is (weakly) continuous, and thus ideal for analyzing corrupted and noisy data. Until recently, however, no kernel methods for dealing with nonlinear data have been proposed via the Wasserstein distance. In this work, we develop a novel method to compute the L 2 -Wasserstein distance in reproducing kernel Hilbert spaces (RKHS) called kernel L 2 -Wasserstein distance, which is implemented using the kernel trick . The latter is a general method in machine learning employed to handle data in a nonlinear manner. We evaluate the proposed approach in identifying computed tomography (CT) slices with dental artifacts in head and neck cancer, performing unsupervised hierarchical clustering on the resulting Wasserstein distance matrix that is computed on imaging texture features extracted from each CT slice. We further compare the performance of kernel Wasserstein distance with alternatives including kernel Kullback–Leibler divergence we previously developed. Our experiments show that the kernel approach outperforms classical non-kernel approaches in identifying CT slices with artifacts. Highlights: A novel L2-Wasserstein distance in reproducing kernel Hilbert spaces was proposed. The resultant distance matrix was integrated with a hierarchical clustering method. The approach was used to identify computed tomography slices with dental artifacts.