Canon Medical/Royal Academy of Engineering Research Chair in Healthcare AI

In the UK alone, currently 7 million people live with cardiovascular disease and this number will increase as the population ages. Under-resourced and under-staffed healthcare systems are struggling with the rising caseload and the large volumes of information being generated. Currently, excitement in Artificial Intelligence (AI) for healthcare is high, because of its potential to help stem this information overload and reduce healthcare costs.

The AI paradigm fuelling this excitement heavily depends on well-curated training data and is largely seen as a black box. In contrast we will:

  1. learn from complex, multimodal, healthcare records with minimal supervision; and
  2. focus on problems underpinning learning causal data representations optimised to provide a transparent base for the desired diagnoses and predictions. We will then translate these techniques to automated estimation of cardiac biomarkers, disease diagnosis, and most ambitiously, cardiac episode prediction, thus opening roads to preventive care.

News

  • Tutorial: Diffusion Models for Medical Imaging MICCAI Tutorial, MICCAI, October 2023.

  • Workshop: Co-organizing “Medical Applications with Disentanglements”, Workshop, MICCAI, 2022. link

  • Workshop: Co-organizing “Domain Adaptation and Representation Transfer”, Workshop, MICCAI, 2022. link

  • Invited talks: Big AI in healthcare, Upenn (May 2021), DKFZ (March 2022)

  • Keynote: Algorithms to do more with less?, International Conference on Image and Vision Engineering IMPROVE 2021.

  • Tutorial: DREAM: Disentangled Representations for Efficient Algorithms for Medical data, MICCAI Tutorial, MICCAI, September 2021. This year, the tutorial has an associated paper.

  • Keynote: Doing More with Less by Disentangled Data Representations, Keynote DART, MICCAI, September 2021.

  • Tutorial: DREAM: Disentangled Representations for Efficient Algorithms for Medical data, MICCAI Tutorial, MICCAI, October 2020.

  • Keynote: Doing More with Less by Disentangled Data Representations, Keynote DART, MICCAI, October 2020.

  • Event: Sotos invited to present at The Global AI Summit, organized by the Saudi Data and Artificial Intelligence Authority, March 2020. [postponed due to COVID-19]

  • News: How can AI help? Reading the language of medicine, Canon Medical Blog, October 2019.

  • Event: Public Event in Japan, British Embassy. AI and Healthcare, October 2019.

  • Keynote: Disentangled Representation Learning in Medical Imaging, Keynote STACOM, MICCAI, October 2019.

  • Tutorial: Simulation and Synthesis, Tutorial at International Conference on Acoustics Speech and Signal Processing (ICASSP), May 2019.

  • News: Interview in Greek Huffington Post, September 2019.

  • News: How can AI help? Reading the language of medicine, Canon Medical Blog, (link)

  • Event: Intervention at Future-proofing society – how digital health and social care can empower and transform lives, Scottish Parliament and IET Event.

  • News: Prof. Tsaftaris featured in School of Engineering News.

  • News: Prof. Tsaftaris featured in RAEng News.

  • Kick-off: We started a new project.

Publications

1. Learning to Segment from Scribbles using Multi-scale Adversarial Attention Gates
G. Valvano, A. Leo, S.A. Tsaftaris
IEEE Transactions on Medical Imaging, vol. 40, no. 8, pp. 1990-2001, Aug. 2021
Preprint Code Website

2. Pseudo-healthy synthesis with pathology disentanglement and adversarial learning
T. Xia, A. Chartsias, S.A. Tsaftaris
Medical Image Analysis, vol. 64, 2020
Full Text Preprint Code

3. Diffusion Models for Causal Discovery via Topological Ordering
P. Sanchez, X. Liu, A.Q. O'Neil, S.A. Tsaftaris
ICLR 2023
Preprint Code

4. Indication as Prior Knowledge for Multimodal Disease Classification in Chest Radiographs with Transformers
G. Jacenków; A.Q. O’Neil, S.A. Tsaftaris
ISBI 2022
Full Text Preprint Code

5. CTR: Contrastive Training Recognition Classifier for Few-Shot Open-World Recognition
N. Dionelis, S. A. Tsaftaris, M. Y. Vaighan
ICPR 2022

6. OMASGAN: Out-of-distribution Minimum Anomaly Score GAN for Anomaly Detection
N. Dionelis, S. A. Tsaftaris, M. Y. Vaighan
SSPD 2021
Preprint

7. vMFNet: Compositionality Meets Domain-generalised Segmentation
X. Liu, S. Thermos, P. Sanchez, A.Q. O'Neil, S.A. Tsaftaris
MICCAI 2022
Full Text Preprint Code

8. What is Healthy? Generative Counterfactual Diffusion for Lesion Localization
P. Sanchez, A. Kascenas, X. Liu, A.Q. O'Neil, S.A. Tsaftaris
MICCAI Workshop 2022, Deep Generative Models
Full Text Preprint Code

9. HSIC-InfoGAN: Learning Unsupervised Disentangled Representations by Maximising Approximated Mutual Information
X. Liu, S. Thermos, P. Sanchez, A.Q. O'Neil, S.A. Tsaftaris
MICCAI Workshop 2022, Medical Applications with Disentanglements (MAD)
Preprint

10. Why patient data cannot be easily forgotten?
R. Su, X. Liu, S.A. Tsaftaris
MICCAI 2022
Full Text Preprint

11. Diffusion Causal Models for Counterfactual Estimation
P. Sanchez, S.A. Tsaftaris
CLeaR 2022
Preprint Code

12. Semi-supervised Meta-learning with Disentanglement for Domain-generalised Medical Image Segmentation
X. Liu, S. Thermos, A. O'Neil, S.A. Tsaftaris
MICCAI 2021
Preprint Code

13. Controllable cardiac synthesis via disentangled anatomy arithmetic
S. Thermos, X. Liu, A. O'Neil, S.A. Tsaftaris
MICCAI 2021
Preprint Code

14. Have you forgotten? A method to assess if machine learning models have forgotten data
X. Liu, S.A. Tsaftaris
MICCAI 2020
Preprint

15. Measuring the Biases and Effectiveness of Content-Style Disentanglement
X. Liu, S. Thermos, G. Valvano, A. Chartsias, A.Q. O'Neil, S.A. Tsaftaris
BMVC, 2021
Preprint Code

16. INSIDE: Steering Spatial Attention with Non-Imaging Information in CNNs
G. Jacenków, A.Q. O'Neil, B. Mohr, S.A. Tsaftaris
MICCAI 2020
Full Text Preprint Code

17. Disentangled Representations for Domain-generalized Cardiac Segmentation
X. Liu, S. Thermos, A. Chartsias, A.Q. O'Neil, S.A. Tsaftaris
MICCAI 2020, STACOM: Statistical Atlases and Computational Modelling of the Heart
Preprint Code

18. Max-Fusion U-Net for Multi-Modal Pathology Segmentation with Attention and Dynamic Resampling
H. Jiang, C. Wang, A. Chartsias, S.A. Tsaftaris
MICCAI 2020, MyoPS: Multi-sequence CMR based Mycardial Pathology Segmentation Challenge
Preprint Code

19. Semi-supervised Pathology Segmentation with Disentangled Representations
H. Jiang, A. Chartsias, X. Zhang, G. Panastasiou, S. Semple, M. Dweck, D. Semple, R. Dharmakumar, S.A. Tsaftaris
MICCAI 2020, DART: Domain Adaptation and Representation Transfer
Preprint Code

20. Consistent Brain Ageing Synthesis
T. Xia, A. Chartsias, S.A. Tsaftaris
MICCAI 2019
PDF Code

21. Multimodal cardiac segmentation using disentangled representations
A. Chartsias, G. Papanastasiou, C. Wang, C. Stirrat, S. Semple, D. Newby, R. Dharmakumar, S.A. Tsaftaris
MICCAI 2019, STACOM: Statistical Atlases and Computational Models of the Heart
PDF Code

22. Temporal Consistency Objectives Regularize the Learning of Disentangled Representations
G. Valvano, A. Chartsias, A. Leo, S.A. Tsaftaris
MICCAI 2019, DART: Domain Adaptation and Representation Transfer
Preprint Code

23. Conditioning Convolutional Segmentation Architectures with Non-Imaging Data
G. Jacenków, A. Chartsias, B. Mohr, S.A. Tsaftaris
MIDL 2019
Full Text

People

Funding

Generously supported by Canon Medical Research Europe, the Royal Academy of Engineering and the School of Engineering.