A Semiparametric Gaussian Mixture Model for Chest CT-based 3D Blood Vessel Reconstruction

• DOI: 10.48550/arxiv.2403.19929
• Publication Date: 2024-03-28
• Authors: Qianhan Zeng; Jing Zhou; Ying Ji; Hansheng Wang
• Keywords: Semiparametric model

Computed tomography (CT) has been a powerful diagnostic tool since its emergence in the 1970s. Using CT data, three-dimensional (3D) structures of human internal organs and tissues, such as blood vessels, can be reconstructed using professional software. This 3D reconstruction is crucial for surgical operations serve vivid medical teaching example. However, traditional heavily relies on manual operations, which are time-consuming, subjective, require substantial experience. To address this problem, we develop novel semiparametric Gaussian mixture model tailored vessels. extends classical by enabling nonparametric variations component-wise parameters interest according to voxel positions. We kernel-based expectation-maximization algorithm estimating parameters, accompanied supporting asymptotic theory. Furthermore, propose regression method optimal bandwidth selection. Compared conventional cross-validation-based (CV) method, outperforms CV terms computational statistical efficiency. In application, methodology facilitates fully automated vessel with remarkable accuracy.