"AI breakthrough for fast and cheaper CMR scans" -- Oxford RDM news page, 7 July 2021
"AI replaces contrast dyes for needle-free CMR" -- NIHR Oxford BRC news page, 7 July 2021
"AI replaces contrast dye for fast, cheaper ..." -- Oxford University Hospitals news, 8 July 2021
"New AI heart scanner will cut NHS backlog ..." -- The Telegraph, 7 August 2021
SCMR Newsletter, 29 July 2021
"AI breakthrough for faster, cheaper and injection-free heart scans" -- BHF news page
BBC Radio 4 Today Interview, on how new AI technologies can help with NHS backlog, 9 August 2021
Times Radio Interview, on AI and robotics in healthcare, 10 August 2021
PhD, MEng, BEng, BSc
British Heart Foundation CRE Intermediate Transition Fellow
Deep Learning in Cardiovascular Magnetic Resonance
I am a deep learning (machine learning) scientist, with expertise in CMR, and cross-domain knowledge of cardiovascular diseases, MR physics and scan protocols. I work on the interpretation and enhancement of gadolinium-free native CMR modalities, particularly quantitative T1-mapping, using novel artificial intelligence approaches. My recent research focus has been on AI Virtual Native Enhancement imaging, where we develop AI techniques that could serve as "virtual contrast dye" to replace intravenous contrast dye. This work has been funded by the John Fell Fund and British Heart Foundation Centre of Research Excellence.
I also work actively on CMR T1-mapping standardisation. T1-mapping offers a unique opportunity for myocardial tissue characterization. However, for clinical sites attempting to implement T1 mapping, it is often unclear how to validate the methods correctly before using them for clinical diagnosis. We have developed a robust and practicable phantom quality assurance approach to assure the accurate installation of CMR T1-mapping methods. The programme aims to translate T1-mapping into widespread use in a real-life setting.
Zhang Q. et al, (2021), Circulation
Endogenous T1ρ cardiovascular magnetic resonance in hypertrophic cardiomyopathy.
Thompson EW. et al, (2021), J Cardiovasc Magn Reson, 23
Towards Replacing Late Gadolinium Enhancement with Artificial Intelligence Virtual Native Enhancement for Gadolinium-Free Cardiovascular Magnetic Resonance Tissue Characterization in Hypertrophic Cardiomyopathy.
Zhang Q. et al, (2021), Circulation
Deep neural network ensemble for on-the-fly quality control-driven segmentation of cardiac MRI T1 mapping.
Hann E. et al, (2021), Med Image Anal, 71
Cardiac stress T1-mapping response and extracellular volume stability of MOLLI-based T1-mapping methods.
Burrage MK. et al, (2021), Sci Rep, 11
Cardiovascular magnetic resonance stress and rest T1-mapping using regadenoson for detection of ischemic heart disease compared to healthy controls.
Burrage MK. et al, (2021), Int J Cardiol
Standardization of T1-mapping in cardiovascular magnetic resonance using clustered structuring for benchmarking normal ranges.
Popescu IA. et al, (2021), Int J Cardiol, 326, 220 - 225
Quality assurance of quantitative cardiac T1-mapping in multicenter clinical trials - A T1 phantom program from the hypertrophic cardiomyopathy registry (HCMR) study.
Zhang Q. et al, (2021), Int J Cardiol
1. Zhang Q, Piechnik SK, Ferreira VM, Hann E, Popescu IA: “Enhancement of Medical Images”, Oxford University Innovation, PCT/GB2020/052117, published 11 March 2021 (Publication number WO/2021/044153)
2. Zhang Q, Piechnik SK, Ferreira VM, Werys K, Popescu IA: “Validation of Quantitative Magnetic Resonance Imaging Protocols”, Oxford University Innovation, PCT/GB2020/051189, published 26 Nov 2020 (Publication number WO/2020/234570)
3. Hann E, Piechnik SK, Popescu IA, Zhang Q, Werys K, Ferreira VM: “Method and Apparatus for Quality Prediction”, Oxford University Innovation, PCT/GB2020/050249, published 13 Aug 2020 (Publication number WO/2020/161481)