The Motion Modelling and Analysis Group (MMAG) is an academic research group based within the School of Biomedical Engineering and Imaging Sciences of King's College London. The group is based at St. Thomas' Hospital in central London.
The MMAG has worked extensively in the past on the imaging, modelling and estimation of repetitive motion. Organ motion due to respiration and the beating of the heart is approximately periodic, so measurements of this motion made from imaging data can be used to construct mathematical models of the periodic motion and its variability. Such models can be applied to update guidance information in image-guided interventions or to correct acquired imaging data for the corrupting effects of motion. The group has worked on applying such models in a range of applications, but predominantly in the context of image-guided cardiac interventions and PET-MR imaging.
Currently, the main research theme of the group is the analysis of motion with the aim of extracting clinically useful biomarkers. Here, work focuses on cardiac cycle motion, in which changes in heart function will manifest themselves as subtle changes in the observed motion of the heart as it beats. The group works on developing novel machine learning techniques to learn motion features that are useful for a range of clinical prediction tasks.
The group is also involved in developing machine learning solutions to other problems, such as radiotherapy and vocal tract imaging.
- 20 September 2021: New MICCAI workshop papers now have Arxiv papers online:
- Ines Machado et al - "Quality-aware Cine Cardiac MRI Reconstruction and Analysis from Undersampled k-space Data", Proceedings MICCAI STACOM (Arxiv paper)
- Jorge Mariscal-Harana et al, "Improved AI-based Segmentation of Apical and Basal Slices From Clinical Cine CMR", Proceedings MICCAI STACOM, 2021. (Arxiv paper)
- Tareen Dawood et al, "Uncertainty-Aware Training for Cardiac Resynchronisation Therapy Response Prediction", Proceedings MICCAI STACOM, 2021. (Arxiv paper)
- 24 June 2021: New MICCAI 2021 paper by Esther Puyol-Antón now available on Arxiv: "Fairness in Cardiac MR Image Analysis: An Investigation of Bias Due to Data Imbalance in Deep Learning Based Segmentation" (Arxiv paper)
- 22 April 2021: New journal paper by Chris Arthurs et al - "Active training of physics-informed neural networks to aggregate and interpolate parametric solutions to the Navier-Stokes equations", Journal of Computational Physics, 2021. (open access paper)
- 22 March 2021: New journal paper by Laia Humbert-Vidan et al - "Comparison of machine learning methods for prediction of osteoradionecrosis incidence in patients with head and neck cancer", British Journal of Radiology, 2021. (paper)
- 16 November 2020: New journal paper published by Matthieu Ruthven et al - "Deep-learning-based segmentation of the vocal tract and articulators in real-time magnetic resonance images of speech", Computer Methods and Programs in Biomedicine, 2020. (paper)
- 8 September 2020: New MICCAI workshop papers now have Arxiv papers online:
- Nick Byrne et al - "A Persistent Homology-based Topological Loss Function for Multi-class CNN Segmentation of Cardiac MRI", Proceedings MICCAI STACOM (Arxiv paper)
- Bram Ruijsink et al, "Quality-aware Semi-supervised Learning for CMR Segmentation", Proceedings MICCAI STACOM (Arxiv paper)
- Zhen Yuan et al, "Deep Learning for Automatic Spleen Length Measurement in Sickle Cell Disease Patients", Proceedings MICCAI ASMUS, 2020. (Arxiv paper)
- 7 September 2020: New journal paper by James Clough et al - "A Topological Loss Function for Deep-Learning based Image Segmentation using Persistent Homology", IEEE Transactions on Pattern Analysis and Machine Intelligence, 2020. (paper)
- 24 August 2020: New journal paper by Esther Puyol-Antón - "Automated Quantification of Myocardial Tissue Characteristics From Native T1 Mapping Using Neural Networks With Uncertainty-based Quality-control", Journal of Cardiovascular Magnetic Resonance, 22:60, 2020. (open access paper)
- 15 July 2020: New TMI paper by Ilkay Oksuz: Deep Learning Based Detection and Correction of Cardiac MR Motion Artefacts During Reconstruction for High-Quality Segmentation - see project summary
- 25 June 2020: Accepted MICCAI 2020 paper by Esther Puyol-Antón now has a preprint available online: Interpretable Deep Models for Cardiac Resynchronisation Therapy Response Prediction