معرفی
Dr. Michael Roberts is a Principal Research Associate (equivalent to Reader/Associate Professor) at the University of Cambridge's Department of Applied Mathematics and Theoretical Physics (DAMTP) and Department of Medicine. He leads the BloodCounts! consortium and the algorithm development team for the global AIX-COVNET collaboration. His interdisciplinary research focuses on variational methods for image processing, machine learning for medical imaging, and addressing reproducibility challenges in AI-driven science.
Education:
- PhD in Mathematics, University of Liverpool (2019)
- MSc in Mathematics with Honors, Durham University (2015)
Research Interests:
- Image segmentation and registration
- Machine learning applications in medical imaging (e.g., lung diseases, cardiovascular imaging)
- Scientific integrity and reproducibility in AI
- Collaborations with clinicians, computer scientists, and applied mathematicians
Key Projects:
- BloodCounts!: Developing AI tools for blood cell analysis
- AIX-COVNET: Global collaboration for AI-driven COVID-19 imaging analysis
Publications Highlight Trends:
His work emphasizes improving AI reliability in medical contexts, addressing challenges like image quality assessment, artifact correction, and mitigating biases in machine learning models. Recent studies include OCT plaque analysis, chest radiograph quality control, and consensus-based guidelines for machine learning research (ReFORMS).
Grants and Labs:
Leads major interdisciplinary initiatives and collaborates with institutions like AstraZeneca. Active in Cambridge Image Analysis (CIA) group.
