Jiayang Shi

I am currently a PostDoc researcher at Research Institute for Mathematics and Computer Science (CWI), Amsterdam, the Netherlands, supervised by Tristan van Leeuwen and Joost Batenburg.

Prior to this, I worked as a PhD student at Leiden University from 2021 to 2025, supervised by Daan Pelt and Joost Batenburg. I earned my bachelor's degree in 2013 from Hangzhou, China, and completed my master's in 2021 in Karlsruhe, Germany.

Email  /  Google Scholar  /  GitHub  /  LinkedIn

Research

DM4CT is a comprehensive benchmark for diffusion-based methods in CT reconstruction. It systematically compares ten state-of-the-art diffusion models alongside modern learning-based approaches (e.g., Gaussian Splatting and INRs) and classical reconstruction algorithms across diverse CT datasets. The benchmark provides detailed insights into reconstruction quality, efficiency, and practical trade-offs. In addition, we introduce a new real-world high-resolution synchrotron CT dataset to facilitate realistic evaluation of CT reconstruction methods.

A self-supervised approach designed for edge illumination X-ray phase contrast micro-CT to enhance the resolution of anisotropic volumes.

A artifact reduction method for CT images based on deep learning. Three CNNs reduce artifacts in a subsequent manner, targeting different types of artifacts. This method is seamlessly integrated into existing CT pipelines.

A novel Bayesian framework for joint reconstruction of multiple objects from sparse-view CT scans using Implicit Neural Representations (INRs) to improve reconstruction quality. By capturing shared patterns across multiple objects with latent variables, our method enhances the reconstruction of each object, increases robustness to noise, and accelerates the learning process.

A Benchmark Low-dose Industrial CT Dataset Tailored for Deep Learning: This work introduces a dataset specifically designed for industrial CT applications, emphasizing deep learning approaches. It features a comprehensive analysis comparing the effectiveness and efficiency of 2D, 2.5D, and 3D training methodologies in the context of denoising and image enhancement.

A self-supervised approach utilizing off-axis training to improve the resolution of through-plane CT scans. This method is trained using in-plane images and applied to through-plane images, offering flexibility with any resolution and overlap.

Survey paper on diffusion models for tabular data.

Unsupervised cross-domain graph level anomaly detection.

Graph neural network for log anomaly detection.

A novel method to reconstruct a spectral central view and its aligned disparity map from spatio-spectrally coded light fields using multi-task deep learning.

Teaching Assistant

Supervised Thesis

Donghang Lyu (master student): Res-Swin: Effective Combination of ResNet and Swin Transformer for LDCT Denoising
Omar Elkilany (master student@TUM & WayGate): Beam Hardening Correction using Implicit Neural Representations
Jan Zuromski (master student): Classifier Guided Inversion for Image Reconstruction