Friday June 17th, 2022

Join our one-day conference to learn about how SmartHeart researchers have been using machine learning methods to make MR cardiac imaging faster, smarter and more efficient.

Participation is free but registration is required.

Location: Imperial College London, South Kensington




Registration, refreshments and networking

Introduction and overview of the SmartHeart Project
Daniel Rueckert, Imperial College London & TU Munich



MR fingerprinting: towards a virtual biopsy for cardiovascular MRI
Claudia Prieto, King’s College London

3D multi-parametric whole heart MRI: simple, fast and efficient
René Botnar, King’s College London

AI-enabled cardiac imaging quality control
Julia Schnabel, King’s College London & TU Munich

Pathways from MSK to Cardiac MRI Reconstruction
Kerstin Hammernik, Imperial College London & TU Munich


Coffee & Poster Session



AI-enabled cardiac functional quantification
Andy King, King’s College London

Modelling the Shapes for 100,000 Hearts
Wenjia Bai, Imperial College London

Learning left ventricular motion patterns from cardiac MR image sequences
Alison Noble, University of Oxford



Combining Modelling and Imaging of the Heart
Blanca Rodriguez, University of Oxford





An industry perspective on AI in CMR
Speaker TBC, Siemens Healthineers



Barts Clinical talk (title TBC)
Speaker TBC

Kings Clinical talk (title TBC)
Reza Razavi, Kings College London

AI Research at HeartFlow: learning from CCTA data and beyond
Matt Sinclair, HeartFlow



The Potential of AI in Cardiac MR: a clinical perspective
Rhodri Davies, University College London

AI has the potential to transform clinical practice in cardiac MRI. Programmes such as SmartHeart and the availability of standardised cardiac MR images have made this a fertile area of research. Progress in AI research is propagating into clinical practice and most CMR analysis tools now have AI at their core. I will give a clinical perspective on the use of AI for cardiac MR and explore its potential, focussing on AI for precision measurement, showing the profound effect it has on clinical management. I will explore some of the challenges in translating AI algorithms into clinical use, and look ahead at other potential applications and opportunities for AI in clinical work.


Poster Session & Drinks

Keynote Speakers

Rhodri Davies is an Associate Clinical Professor in Cardiology and Machine Learning at the Institute of Cardiovascular Science, University College London and Barts Heart Centre. Having originally trained in computer science, he obtained a PhD in statistical shape analysis for medical imaging before complete clinical training. He combines his role at UCL with a post as consultant cardiologist at the University Hospital of Wales, specialising in cardiac MRI and echocardiography.  His research interests are in applying AI to cardiac imaging to gain insight into health and disease and improve patient care.


Self-Supervised Learning for Cardiac MR Image Segmentation by Anatomical Position Prediction Wenjia Bai, Imperial College London

A cooperative training method to improve model cross-domain performance using single-domain data only Chen Chen, Imperial College London

Realistic Adversarial Data Augmentation Chen Chen, Imperial College London

AtrialJSQnet: A New framework for joint segmentation and quantification of left atrium and scars incorporating spatial and shape information Lei Li, University of Oxford

Atrial Scar Quantification via Multi-scale CNN in the Graph-cuts Framework Lei Li, University of Oxford

Multiscale Graph Convolutional Networks for Cardiac Motion Analysis Ping Lu, University of Oxford

A Deep Learning-based Integrated Framework for Quality-aware Undersampled Cine Cardiac MRI Reconstruction and Analysis Ines Machado, Kings College London

Self-Supervision with Superpixels: Training Few-shot Medical Image Segmentation without Annotation Cheng Ouyang, Imperial College London

Data Efficient Unsupervised Domain Adaptation for Cross-Modality Image Segmentation Cheng Ouyang, Imperial College London

Free-running 3D T1 and T2 myocardial mapping and cine MRI in 3 minutes using low-rank non-rigid motion-corrected reconstruction Andrew Phair, Kings College London

Fairness in Cardiac MR Image Analysis: An Investigation of Bias Due to Data Imbalance in Deep Learning Based Segmentation Esther Puyol-Anton, Kings College London

Interpretable Deep Models for Cardiac Resynchronisation Therapy Response Prediction Esther Puyol-Anton, Kings College London

MR physics-based data augmentation for cardiac MR segmentation in different domains Devran Ugurlu, Kings College London

Joint Motion Correction and Super Resolution for Cardiac Segmentation Shuo Wang, Imperial College London

Deep Generative Model-based Quality Control for Cardiac MRI Segmentation Shuo Wang, Imperial College London




Friday June 17th, 2022
09:00 – 17:00

Room 3.01 (level 3)
Royal School of Mines
Imperial College London
South Kensington, London, SW7 2AZ

Please note the following before registering:

  • Participation is free but registration is required.
  • In line with COVID-19 safety precautions, if you feel sick or are experiencing COVID-like symptoms prior to the meeting, we ask that you do not participate in-person.
  • Spaces are limited, so if you are no longer able to attend, please let us know so that your place can be offered to someone else.