PhD defence
Deep learning for vascular segmentation and tissue characterization in CT images
- X. Zhang
- Date
- Wednesday 7 January 2026
- Time
- Location
-
Academy Building
Rapenburg 73
2311 GJ Leiden
Supervisor(s)
- Prof.dr. B.P.F. Lelieveldt
- dr. J. Dijkstra
- dr. A. Broersen
Summary
My PhD research focuses on developing intelligent methods for medical image analysis, with applications in cardiovascular analysis and liver structure modeling. In my early work, I developed machine learning models that predict the spatial orientation of lipid-rich and calcified plaques from CT angiography, enabling non-invasive detection and characterization of coronary plaques.
The main part of my research addresses automatic segmentation of liver vessels and functional anatomy (Couinaud segmentation), which are essential for surgical planning and treatment evaluation. I proposed a graph-attention guided diffusion model that ensures continuous and complete vessel segmentation, overcoming the fragmentation problems common in deep learning methods. In another approach, I introduced Top-K Maximum Intensity Projection (MIP) priors to combine information from multiple 2D views for better 3D vessel segmentation. Separately, I developed a graph reasoning-based method for Couinaud segmentation, which is vessel-prior-free and divides the liver into functional segments relevant for clinical use.
Together, these developments enable more accurate and consistent medical image analysis, improving visualization of organ structures and supporting safer surgical planning and patent care.
PhD dissertations
Approximately one week after the defence, PhD dissertations by Leiden PhD students are available digitally through the Leiden Repository, that offers free access to these PhD dissertations. Please note that in some cases a dissertation may be under embargo temporarily and access to its full-text version will only be granted later.
Press enquiries (journalists only)
General information
Beadle's Office
pedel@bb.leidenuniv.nl
+31 71 527 7211