Cheng Ouyang

Cheng Ouyang

Lecturer (Assistant Professor) in Healthcare Imaging at Department of Engineering Science, University of Oxford

Cheng Ouyang

I am a Lecturer (Assistant Professor) in Healthcare Imaging at the Department of Engineering Science, University of Oxford. My research centres on data-efficient, robust, and user-friendly machine learning approaches for medical image/signal computing. Prior to joining Oxford, I worked as a postdoctoral researcher on machine learning for cardiovascular science at the Institute of Clinical Sciences, Imperial College London. Before that, I obtained my PhD in Computing Research from the Department of Computing, Imperial College London.

Research Interest

Keywords: Medical image computing, Machine learning, Robustness, Uncertainty, Human-machine collaboration, Medical ultrasound, MRI

My research centres on machine learning and computer vision approaches for medical data interpretation and analysis. Topics of our research include but are not limited to the following:

  • Multimodal machine learning for healthcare data (e.g., image, signal, and language)
  • Domain generalization and uncertainty modeling for easy-to-deploy and trustworthy machine learning
  • Few-/zero-shot learning for improving data efficiency on long-tailed medical data
  • Machine learning for medical scientific discoveries (e.g., cardiovascular science)
  • Applications of the above learning approaches in addressing clinical and medical research challenges

We are also part of the Oxford Biomedical Image Analysis Cluster (BioMedIA).

Opening

We are looking for highly motivated students to work on machine learning for medical image/data computing. Please check the Vacancies page for more details.

News

[Sep. 2024] Our papers on multi-modal self-supervised learning for fine-grained medical vision tasks and on robust and generalizable diffusion models for inverse problems were accepted for NeurIPS 2024. Congrats to Che and Weitong!
[Jun. 2024] Our papers on topology correction for medical image segmentation and on foundation model for brain MRI segmentation were accepted for MICCAI 2024. Congrats to Liu and Xinru!
[May. 2024] Our paper on zero-shot ECG classification was accepted for ICML 2024; paper on federated LDCT denoising was accepted for MIDL 2024. Congrats to Che and Xuhang!
[Apr. 2024] Started my lectureship at Department of Engineering Science, University of Oxford.
[Apr. 2024] We are calling for participants for the CMRxRecon2024 Challenge and UNSURE 2024 Workshop at MICCAI 2024.
[Dec. 2023] Awarded the Imperial College London Seeds for Success Fund, for the topic of trustworthy machine learning for MRI reconstruction (project lead, 2024 cohort).
[Nov. 2023] Our paper about test-time adaptation for MRI super-resolution was accepted for Med-NeurIPS 2023. Congrats to Weitong!
[Oct. 2023] Awarded as a gold distinguished reviewer for IEEE Transactions on Medical Imaging.
[Aug. 2023] An invitated talk Generalizable and data-efficient medical image computing is given to the School of Computer Science, University of Birmingham.
[Aug. 2023] Awarded the Dame Julia Higgins Postdoc Collaborative Research Fund.
[June. 2023] Our paper about topology-aware medical image segmentation was accepted for MICCAI 2023. Congrats to Liu.
[Sep. 2022] We won the Fetal Tissue Annotation and Segmentation Challenge (FeTA), MICCAI 2022.

Contact

Email: <my_firstname> dot <my_lastname> at eng dot ox dot ac dot uk
Address: Old Road Campus, Roosevelt Dr, Headington, Oxford OX3 7DQ