Guanyu Yang

TL;DR: An algorithm is proposed that perform a temporal tracking of the vessel central axis in a 3-D dynamic sequence in multi-slice computed tomography (MSCT) based on geometric moments and a local cylindrical approximation.

TL;DR: This work proposes the better registration better segmentation (BRBS) framework with three main contributions that are experimentally shown to have substantial practical merit, and improves the authenticity in the registration-based generation program and proposes the knowledge consistency constraint strategy that constrains the registration network to learn according to the domain knowledge.

TL;DR: In this paper, a self-supervised learning method combined with radiomics is proposed to learn rich visual representation about tumors without human supervision, which can solve the poor predictive ability caused by tumor heterogeneity and inadequate labeled image data.

TL;DR: Experiments demonstrate that the proposed set of noise features used to identify the origin of one CT image with a detection rate of at least 94% and that it achieves better performance than sensor pattern noise (SPN) based strategy proposed for general public camera devices.

TL;DR: A robust approach termed CMPP (constrained minimal path propagation) is here proposed to extend a method derived to deal with curve-like structures such as vessel centerlines to extract the complete vessel lumen after the centerlines have been found.