ZW.
Research

From stochastic processes to useful AI systems.

My group develops theory and methods for backward stochastic processes in generative machine learning, then applies them to medical imaging, computational imaging, and inverse problems in engineering and the natural sciences.

01

Generative models

Diffusion models, latent-variable learning, score matching, and stochastic flow inference.

02

Inverse problems

Physics-constrained generation and uncertainty-aware reconstruction for scientific imaging.

03

Healthcare AI

Cross-modality translation, image analysis, multimodal foundation models, and clinical decision support.

Selected publications

2026

  1. Jingbo Wang, Kegeng Tang, Zihao Wang, Shaogang Ren. SCDM: Scalable Causal Discovery in Nonlinear Temporal Systems with Meta-Learning. Neural Information Processing Systems (NeurIPS), 2026.
    NeurIPS 2026 presentation
  2. Kegeng Tang, Jingbo Wang, Shaogang Ren, Zihao Wang (Work as PI). CT-ΔBench: A Benchmark for Longitudinal 3D Medical Imaging Difference Reporting with Vision-Language Models. Conference on Language Modeling (COLM), 2026.
    arXiv 2608.11534 COLM 2026 presentation Hugging Face dataset
  3. Zihao Wang (Work as PI). Backward SDE-Based Diffusion for Physics-Constrained Generation. International Conference on Machine Learning (ICML), 2026.
    Project website Paper PDF Coding agent skill
  4. Zihao Wang (Work as PI), Yu Gan, Ona Wu. BlochShift: Physics-guided diffusion with Bloch-based drift for MRI image translation. IEEE IVMSP, 2026.
  5. Y. Liu, S. Chan, J. Chu, Y. Zhang, C. Yang, Zihao Wang, Y. R. Gel, Y. Chen. Multi-Modal Enhanced Graph Transfer Learning for Digital Finance Fraud Detection. Proceedings of the ACM Web Conference (WWW), 8709-8712, 2026.
  6. Zihao Wang (Work as PI), Yuzhou Chen, Shaogang Ren. Adaptive Domain Shift in Diffusion Models for Cross-Modality Image Translation. International Conference on Learning Representations (ICLR), 2026.
    Paper on OpenReview Source code on GitHub

2025

  1. Zihao Wang, C. Vandersteen, C. Raffaelli, N. Guevara, H. Delingette. Multi-Energy Quasi-Symplectic Langevin Inference for Latent Disentangled Learning. IEEE Transactions on Image Processing, 34:7037-7049, 2025.
  2. Zihao Wang, A. M. Kulpanowski, W. A. Copen, B. L. Hancock, E. S. Rosenthal, et al., O. Wu. Automated Detection of Severe Cerebral Edema Using Explainable Deep Transfer Learning after Hypoxic Ischemic Brain Injury. Resuscitation, 214:110652, 2025.
  3. Zihao Wang, Yuzhou Chen, Shaogang Ren. Stochastic Flow Inference for Medical Image Digital Twins. ACM/IEEE CHASE, 353-357, 2025.

2024 and earlier highlights

  1. Zihao Wang, Y. Yang, Y. Chen, T. Yuan, M. Sermesant, H. Delingette, O. Wu. Mutual Information Guided Diffusion Model: with application in zero-shot-learning cross-modality image translation. IEEE Transactions on Medical Imaging, 2024.
  2. Zihao Wang, T. Demarcy, C. Vandersteen, D. Gnansia, C. Raffaelli, N. Guevara, H. Delingette. Bayesian Logistic Shape Model Inference: application to cochlea image segmentation. Medical Image Analysis, 2022.
  3. Zihao Wang, C. Vandersteen, T. Demarcy, D. Gnansia, C. Raffaelli, N. Guevara, H. Delingette. Inner-ear Augmented Metal Artifact Reduction with Simulation-based 3D Generative Adversarial Networks. Computerized Medical Imaging and Graphics, 2021.
  4. Zihao Wang, C. Vandersteen, C. Raffaelli, N. Guevara, H. Delingette. One-shot Learning Landmarks Detection. DGM4MICCAI, 2021. People’s Choice Award.

For the complete and continuously updated list, please visit Google Scholar. Research code is maintained by Laplace Lab.

Open research

Collaboration and advising

I work with students and collaborators who are excited about rigorous AI methods and their use in scientific and clinical settings. If your interests overlap, please get in touch.