Abstract: |
| Cardiovascular disease remains the leading cause of death worldwide, underscoring the urgent need for precise and efficient assessment of heart function to improve health outcomes. However, current clinical workflows primarily rely on 2D imaging, such as cardiac magnetic resonance (CMR), for cardiac evaluation. Accurately reconstructing real-world 3D cardiac anatomy and motion from these 2D images remains a major challenge. Moreover, the highly sensitive nature of medical imaging data makes it difficult to access and share widely. In this talk, I will explore how advanced artificial intelligence techniques and deep learning frameworks can help overcome these limitations. I will present our latest research findings, including recent work accepted by IEEE Transactions on Medical Imaging and MICCAI. The presentation will cover innovative methods such as multi-view 2D CMR fusion for 3D shape and motion estimation, and the use of diffusion models to generate dynamic medical imaging data, for example, echocardiogram videos. I will demonstrate how these approaches can significantly improve the accuracy and robustness of medical image analysis. |
Biography: |
| Dr. Qingjie Meng is an Assistant Professor in the School of Computer Science at the University of Birmingham and an Honorary Research Associate at Imperial College London. She received her PhD in Computing from Imperial College, where she also conducted postdoctoral research. Her research interests lie at Medical AI. Her current work focuses on 2D-to-3D cardiac shape reconstruction and motion tracking, generative models, transfer learning and representation learning. She has published more than 20 papers in leading journals and conferences, including IEEE Transactions on Medical Imaging, IEEE Transactions on Image Processing, CVPR, and MICCAI. She is the Area Chair of MICCAI 2025, main organiser of BioImage Computing workshop at ECCV 2024 and the ASMUS workshop at MICCAI 2024-2025. |