Magnetic resonance imaging (MRI) is an attractive alternative to X-ray fluoroscopy for the guidance of cardiac catheterisation procedures. MRI enables visualisation of soft tissues and interventional devices but also provide a comprehensive assessment of cardiac function and flow. However, this approach remains lengthy (several hours) which currently limits its wider clinical use. The aim of this project is thus to develop novel fast free-breathing accelerated acquisition schemes combined with AI-based image reconstruction to shorten overall procedural time. Specially, highly accelerated acquisition schemes based on non-cartesian trajectory and deep-learning-based image reconstruction will be explored for artefact removal and noise reduction. Novel smart MR-respiratory sensors will be developed for real-time breathing motion monitoring and improved motion correction. Finally, the use of deep learning models will be explored for real-time image quality assessment and automated imaging parameter adjustments. These techniques will be evaluated in phantom, and patients undergoing MRI-guided cardiac catheterisation procedures.
Back to projects
AI-based next-generation imaging for fast interventional MRI-guided cardiac catheterisation procedures
