PET++: Improving Localisation, Diagnosis and Quantification in Clinical and Medical PET Imaging with Randomised Optimisation
Project News Publications Team Contact Software


Positron Emission Tomography (PET) is a pillar of modern diagnostic imaging, allowing non-invasive, sensitive and specific detection of functional changes in several disease types. In endocrinology, the precise localisation of small functioning tumours of the pituitary or adrenal glands is crucial for planning curative surgery or radiotherapy. While PET imaging shows good promise for this task, initial studies suggest significant room for improvement, with improved PET imaging and subsequent more accurate localisation opening up the possibility for more adapted therapies. In dementia, the accurate quantification of PET images is key for the early detection of disease. Improved PET imaging may allow for earlier detection of dementia while asymptomatic and increased sensitivity to assess and monitor treatment once appropriate drugs have been found.

In this project mathematicians team up with researchers and clinicians from Addenbrooke's Hospital Cambridge, Dementias Platform UK (DPUK), GE Healthcare and University College London (UCL) for improved diagnosis and localization for tumours in endocrinology and earlier diagnosis of dementia with improved PET imaging. In particular, we investigate modern PET reconstruction approaches based on advanced mathematical methods to increase the PET image resolution and contrast, while keeping computational complexity low, thereby directly benefiting clinical workflow.


12/2020Our paper Robust Image Reconstruction with Misaligned Structural Information got published by IEEE Access. This is joint work with Leon Bungert (Erlangen, Germany).
11/2020Presentation at the IEEE Medical Imaging Conference, on [Improving a stochastic algorithm for regularized PET image reconstruction] (poster, 1.2MB).
10/2020Presentation at the bi-weekly meetings of Institute of Nuclear Medicine, UCL [Regularized PET reconstruction with a stochastic algorithm] (slides, 1MB).
09/2020Our paper on Accelerating Variance-Reduced Stochastic Gradient Methods got accepted by Mathematical Programming.
07/2020New review paper out on Multi-modality imaging with structure-promoting regularisers. If you are new to image reconstruction from multi-modality imaging data, this is the best place to start!
06/2020Team member C. Delplancke participated to CCPi/SyneRBI vHackathon algorithm, during which Stochastic Primal-Dual Hybrid Gradient algorithm was implemented in CCPi Core Imaging Library (CIL), and to SyneRBI/STIR vHackathon PET scanner support.
04/2020New paper out on an Analysis of Stochastic Gradient Descent in Continuous Time.
04/2020New paper out on making structure-promoting regularizers robust to misalignment: Robust Image Reconstruction with Misaligned Structural Information.
04/2020Our paper on SIRF: Synergistic Image Reconstruction Framework got published in Computer Physics Communication. SIRF is a powerful software toolbox for PET-MR reconstruction.
03/2020The "collaborative computational project" (CCP) on Synergistic Reconstruction for Biomedical Imaging was funded by EPSRC. Its aim is to further develop a software framework called SIRF which we use for PET-MR image reconstruction.
10/2019New paper out on Accelerating Variance-Reduced Stochastic Gradient Methods.
10/2019Great news! Claire Delplancke will be joining our team in Bath in January 2020.
08/2019We are hosting a CCP PET-MR hackathon in Bath on September 23-24, 2019. For more infos see link.
08/2019Our paper on Faster PET reconstruction with non-smooth priors by randomization and preconditioning just got accepted in Physics in Medicine & Biology.
08/2019We are excited for Jonas Latz joining our team in Cambridge in January 2020.
03/2019EPSRC announced the funding of our project EP/S026045/1 on Improving Localisation, Diagnosis and Quantification in Clinical and Medical PET Imaging with Randomised Optimisation to start September 2019.


2021J. Budd, Y. van Gennip, J. Latz, Classification and image processing with a semi-discrete scheme for fidelity forced Allen–Cahn on graphs, to appear in GAMM Mitteilungen [arXiv]
2020 M. J. Ehrhardt, Multi-modality imaging with structure-promoting regularisers, to appear in Springer Handbook of Mathematical Models and Algorithms in Computer Vision and Imaging [preprint] [software@github]
E. Ovtchinnikov, R. Brown, C. Kolbitsch, E. Pasca, C. da Costa-Luis, A. G. Gillman, B. A. Thomas, N. Efthimiou, J. Mayer, P. Wadhwa, M. J. Ehrhardt, S. Ellis, J. S. Jørgensen, J. Matthews, C. Prieto, A. J. Reader, C. Tsoumpas, M. Turner, D. Atkinson, K. Thielemans, SIRF: Synergistic Image Reconstruction Framework, Computer Physics Communication 249, 107087 [print]
D. Driggs, M. J. Ehrhardt, C.-B. Schönlieb, Accelerating Variance-Reduced Stochastic Gradient Methods, Mathematical Programming [print] [preprint]
L. Bungert, M. J. Ehrhardt, Robust Image Reconstruction with Misaligned Structural Information, IEEE Access 8, 222944-222955, [preprint] [print] [software@github]
2019M. J. Ehrhardt, P. J. Markiewicz, C.-B. Schönlieb, Faster PET Reconstruction with Non-Smooth Priors by Randomization and Preconditioning, Physics in Medicine & Biology 64(22), 225019 [preprint] [IOP] [slides (<1MB)]
D. Kazantsev, E. Pasca, M. Basham, M. Turner, M. J. Ehrhardt, K. Thielemans, B. A. Thomas, E. Ovtchinnikov, P. J. Withers, A. W. Ashton, Versatile regularisation toolkit for iterative image reconstruction with proximal splitting algorithms, Proceedings of SPIE 11072 Fully 3D, 2019 [print]
2018A. Chambolle, M. J. Ehrhardt, P. Richtárik, C.-B. Schönlieb, Stochastic Primal-Dual Hybrid Gradient Algorithm with Arbitrary Sampling and Imaging Applications, SIAM Journal on Optimization 28(4), 2783-2808 [print] [preprint] [slides (<1MB)] [poster (2 MB)] [software@github] [software@ODL]
P. J. Markiewicz, M. J. Ehrhardt, K. Erlandsson, P. J. Noonan, A. Barnes, J. M Schott, D. Atkinson, S. R. Arridge, B. F. Hutton, S. Ourselin, NiftyPET: A high-throughput software platform for high quantitative accuracy and precision PET imaging and analysis, Neuroinformatics 16(1), 95–115 [print] [software]
2017M. J. Ehrhardt, P. Markiewicz, A. Chambolle, P. Richtárik, J. Schott, C.-B. Schönlieb, Faster PET Reconstruction with a Stochastic Primal-Dual Hybrid Gradient Method, Proceedings of SPIE [SPIE library] [preprint (2 MB)] [poster (2 MB)]


2021 E. B. Gutierrez, C. Delplancke, M. J. Ehrhardt, Convergence Properties of a Randomized Primal-Dual Algorithm with Applications to Parallel MRI, [preprint]
2020 J. Latz, Analysis of Stochastic Gradient Descent in Continuous Time, [preprint]


2019 Synergistic Reconstruction Symposium, Chester, UK. Faster PET Reconstruction with Non-Smooth Anatomical Priors by Randomization and Preconditioning [slides (2MB)]
Applied Mathematics Seminar, University of Leicester, UK. A Randomized Algorithm for Non-Smooth Optimization and Medical Imaging Applications [slides (1MB)]
Quantitative Imaging of Electrochemical Interfaces Workshop, Diamond Light Source, Harwell Campus, UK. 1 + 1 > 2? Getting More Out of Multi-Modality Imaging [slides (8MB)]
2nd IMA Conference On Inverse Problems From Theory To Application, London, UK. Faster PET Reconstruction with Non-Smooth Priors by Randomization [slides (7MB)] [paper, IOP] [preprint]


Project Leads

Medical and Clinical Project Team


PhD Students

Project Collaborators

Advisory Board


Carola-Bibiane Schönlieb
University of Cambridge, UK

Useful Links

© 2019-21 PET++ team - last updated: February 2021