About me
Nice to meet you here!
I am Xiaojian Xu, an enthusiastic researcher specializing in computational imaging. Currently, I am an AI Research Scientist with the Foundation AI team at GE HealthCare (GEHC), where I focus on developing large language models, foundational AI models, vision-language models, and multimodal models for medical imaging and healthcare applications.
Prior to joining GEHC, I was a postdoctoral researcher in the EECS department at the University of Michigan (UMich) from 2022 to 2024, working under the guidance of Prof. Jeffrey Fessler. I earned my Ph.D. in Computer Science in 2022 from Washington University in St. Louis (WashU), where I had the privilege of working with Prof. Ulugbek Kamilov in the Computational Imaging Group (CIG). Before my time at WashU, I completed my B.E. in Communication Engineering at the University of Electronic Science and Technology of China (UESTC) in 2014. I was admitted to the graduate program as an exam-exempt student and later earned my M.E. in Communication and Information Systems from UESTC in 2017.
My research interests span computational imaging, deep learning, VLMs, LLMs, generative AI, NLP, optimization theory, biomedical imaging, inverse problems, and computer vision. If you’re interested in our work, feel free to reach out!
Recent work
- 05-20-2025: Our work Shorter SPECT Scans Using Self-Supervised Coordinate Learning to Synthesize Skipped Projection Views is accepted to EJNMMI physics!
- 05-15-2025: Our work FlanS - A Foundation Model for Free-Form Language-based Segmentation in Medical Images is accepted to KDD 2025 (Research Track)!
- 05-07-2025: Our work Swap-Net: A Memory-Efficient 2.5D Network for Sparse-View 3D Cone Beam CT Reconstruction to ICF Applications is accepted to IEEE Transactions on Computational Imaging!
- 10-01-2024: Our work Provable Preconditioned Plug-and-Play Approach for Compressed Sensing MRI Reconstruction is accepted to IEEE Transactions on Computational Imaging!
- 09-25-2024: Our work Learning Image Priors through Patch-based Diffusion Models for Solving Inverse Problems is accepted to NeurIPS 2024!
- 09-16-2024: Our work Reconstructing Richtmyer-Meshkov instabilities from noisy radiographs using low dimensional features and attention-based neural networks is accepted to Optics Express!
Personal update
* 05-20-2024: I started a new position as an AI Research Scientist in the Foundation AI Team at GE HealthCare (GEHC).
* 08-22-2022: I started my postdoc research at UMich under the supervision of Prof. Jeffrey Fessler.
* 07-28-2022: I passed my dissertation defense on "Model-based Deep Learning for Computational Imaging".
* 05-24-2021: I started my research intern at Facebook Reality Labs Research (FRL).
* 05-04-2021: I passed my thesis proposal defense on "Model-based Deep Learning for Computational Imaging".
* 04-03-2020: I passed my oral qualification exam on "Computational Imaging: Leverage the Power of Deep Learning".
* 05-27-2019: I started my research intern at Mitsubishi Electric Research Laboratory (MERL).
* 02-01-2018: I joined the Computational Imaging Group (CIG) under the supervision of Prof. Ulugbek Kamilov.
* 08-27-2017: I joined the Computer Science and Engineering (CSE) department at WashU.
Contact information
Washington University in St. Louis
Department of Computer Science and Engineering
Email: xiaojian+lastname@wustl.edu
