Dr. Hui Xue - National Heart, Lung, and Blood Institute, National - CMIC/WEISS joint seminar series

Speaker: Dr. Hui Xue - National Heart, Lung, and Blood Institute, National Institute of Healthcare

Title: Cardiac Imaging Future with AI powered Automation, Quantification and Clinical deployment

Abstract:

Deep learning techniques are penetrating the field of cardiac imaging and causing rapid progress in clinical imaging pipeline. In this presentation, I will provide an in-depth review of our AI R&D in the domain of cardiac MR imaging. The presentation will introduce deep learning solutions for  automated reporting, quantitative mapping and semi-supervised imaging AI. Open-source software stack was developed to enable inline AI inference on imaging devices, which leads to wide global deployment of our deep learning models. The semi-supervised learning was developed to enable training with both small labeled datasets and the large unlabeled datasets. This further encourages continuous data curation and open new opportunities. In this way, an effective imaging AI flywheel is up and running.

Bio:

Dr. Hui Xue is an active contributor to imaging AI application development. He is currently a scientist at the National Heart Lung and Blood Institute, NIH, USA. The AI imaging applications he developed had been deployed at more than forty hospitals globally and used daily to help cardiologists. He is leading cardiac AI developing team and working as technical leader from product definition, data collection, model iteration to final delivery of deep learning based solutions to end-users. Dr. Xue is also a top contributor and maintainer of the well-known open-source MR imaging computing framework, Gadgetron. He had contributed more than 300K line of code to GitHub and in particular developed generic MR reconstruction package for linear and non-linear reconstruction and InlineAI package capable of running latest AI models directly on MR scanners. Dr. Xue earned his PhD in Computing and Medical Imaging from Imperial College London in 2008. He got the MSc in Biomedical Engineering and BSc in ECE from Xi’an JiaoTong University, China in 2003 and 2001. From 2008 to 2013, he worked as a research scientist in Siemens Corporate Research, Princeton, USA. His main achievements included, as one of the key developers, the release of several product features, such as HeartFreeze Inline Motion Correction and MyoMaps. He had authored over 60 scientific publications and holds 14 US and international patents. He is a fellow of The Society for Cardiovascular Magnetic Resonance (SCMR) since Jan 2019.

Chair: Rhodri Davies