Back to Projects

AI-enabled Imaging

Optimising image-metrics for cancer-treatment decisions by bootstrap and AI-methods for PET image reconstruction

Project ID: 2020_039

1st Supervisor: Andrew Reader, King’s College London
2nd Supervisor: Jorge Cardoso, King’s College London
Clinical Champion: Sally Barrington, King’s College London

Aim of the PhD Porject:

To develop new PET image reconstruction methods to improve image-derived metrics of critical importance to disease diagnosis, treatment planning and treatment monitoring.

Project Description / Background:

PET imaging is a powerful functional imaging technique which is in widespread use for diagnosing and monitoring cancer, infection, neurological and cardiovascular diseases [1, 2]. PET images are used for assisting in disease diagnosis and lesion characterisation, cancer staging and treatment response.  The images that are reported by clinicians use quantitative metrics, with particular reliance on standardised uptake values (SUVs) to provide reports that inform how patients are treated at various time points, given that more than one treatment pathway is often available.

Importantly, PET scans are conducted during or after early stages of treatment to assess efficacy, and decide whether (or not) a treatment plan is effective.  PET-guided therapy has become standard practice, continuing or even reducing treatment where effective or switching treatment if not. These are important decisions, as cancer therapies are often accompanied by side-effects which should be limited if at all possible, by avoiding unnecessary treatment, or avoiding inappropriate treatment escalation. On the other hand, discontinuation or pressing on with ineffective treatment can allow cancer to progress with detriment to patients who are sub-optimally treated.

SUVs [3] are commonly used to characterise lesions, assess cancer aggressiveness and monitor treatment.  SUVs are also used to measure uptake in reference regions such as the liver to establish whether patients have had a complete metabolic response.  There is increasing interest in using SUV-derived metrics to measure metabolic tumour burden to develop risk-adapted (before treatment) plans.

Most often the maximum SUV (SUVmax) or recently the “peak SUV” is measured in a region of interest (ROI). However, PET images are reconstructed from relatively noisy, count-limited data, and it is necessary to compensate for noise by various empirical and heuristic methods. Given that the maximum value in a region will be very dependent on the noise level, or the amount of smoothing applied, it is without question that noisy images, or inappropriately smoothed images, can have a notable impact on these clinical metrics and therefore patient management.

There is therefore a need for robust and objective ways to reconstruct PET images which compensate not only for noise, but also biases arising from the empirical or arbitrarily chosen levels of image smoothing, thereby mitigating against variability in SUV measures.

We have recently proposed a new reconstruction methodology which exploits a simple statistical method known as bootstrapping [4]. In previous work, bootstrapping has been used to find out the level of noise in PET images, which is a post hoc analysis of what has already happened in the reconstruction. The new method, in contrast, exploits bootstrapping during the iterative reconstruction process, seeking optimal levels of smoothing of the reconstruction in accordance with the noise level of the data. The outcome is a fixed reconstructed image with an appropriate level of detail, justified only by the measured data. Arbitrary smoothing is avoided, and robustness to differing levels of noise in the raw data has already been demonstrated.

We anticipate that using bootstrap-optimised regularised image reconstruction of clinical PET data will result in more robust SUV-metrics being available for clinical decision making for patient benefit.

Beyond this, AI-advanced versions of the method will be explored, using deep generators to outperform the bootstrap in generating independent noisy data replicates. It is, nonetheless, important to note that the core method to be assessed, based on bootstrapping, is independent of AI. This is potentially advantageous, with expected ready application across different clinical centres and different software platforms and scanners, through the method’s independence from training data (which can restrict utility to dataset sizes and types comparable to those used in the training corpus).

Expected academic background of the candidate: mathematics, physics, and/or computational science.


  1. Muehllehner, G. and J.S. Karp, Positron emission tomography. Physics in Medicine and Biology, 2006. 51(13): p. R117-R137.
  2. Jones, T. and E.A. Rabiner, The development, past achievements, and future directions of brain PET. Journal of Cerebral Blood Flow and Metabolism, 2012. 32(7): p. 1426-1454.
  3. Kinahan, P.E. and J.W. Fletcher, Positron emission tomography-computed tomography standardized uptake values in clinical practice and assessing response to therapy. Semin Ultrasound CT MR, 2010. 31(6): p. 496-505.
  4. Reader, A.J. and S. Ellis, Bootstrap-optimised regularised image reconstruction. Accepted for presentation at the IEEE NSS and Medical Imaging Conference 2019, 2019.
  5. Reader, A.J. and J. Verhaeghe, 4D image reconstruction for emission tomography. Phys Med Biol, 2014.
  6. Mehranian, A., et al., Multi-modal synergistic PET and MR reconstruction using mutually weighted quadratic priors. Magn Reson Med, 2018.

Back to Projects