XClose

UCL EPSRC Centre for Doctoral Training in Intelligent Integrated Imaging in Healthcare

Home
Menu

Modelling and Accounting for Delineation Uncertainties and Anatomical Changes in Proton Beam Therapy

Now Closed

4-year_phd_studentship.png

26 May 2020

 

4-Year PhD Studentship: 
Modelling and Accounting for Delineation Uncertainties and Anatomical Changes in Proton Beam Therapy using Machine Learning

Primary Supervisor: Dr Jamie McClelland, (UCL, Centre for Medical Image Computing, Department of Medical Physics and Biomedical Engineering)
Secondary Supervisor: Professor Andy Nisbet (UCL, Department of Medical Physics and Biomedical Engineering)
Subsidiary Supervisors: Dr Sarah Gulliford (University College London Hospital)

Project summary
This is a 4-year PhD studentship funded by the UCL Department of Medical Physics and Biomedical Engineering (MPBE). The project brings together expertise from UCL and University College London Hospital. The funding covers an annual tax-free stipend (at least £17009) and tuition fees. Home and EU candidates are eligible for this scholarship. The successful candidate will be enrolled onto the UCL CDT in Intelligent Integrated Imaging in Healthcare (i4health) and benefit from the being part of a cohort of PhD students as well as participation in the activities and events organised by the centre.
Background
Proton therapy is an advanced form of radiotherapy that can potentially reduce the dose to healthy tissue when compared to standard photon radiotherapy.
Proton therapy plans are based on a CT scan of the patient acquired 1-2 weeks prior to the start of treatment. The tumour and Organs at Risk (OARs) are delineated on the planning scan, and these are used to generate a treatment plan which aims to deliver a sufficiently high dose to the tumour while ensuring that the OARs do not receive too much dose. However, there are a number of uncertainties that may cause the delivered dose to differ from the planned dose including proton range uncertainty and setup error.   Additionally, there is often some level of uncertainty in the delineation of structures where individual voxels in the planning CT are assigned as belonging to a specific organs or structure, e.g. the tumour, the spinal cord, the parotid gland. Proton therapy is commonly delivered in daily fractions over 5-8 weeks. During this time the patient can undergo anatomical changes such as weight loss, which can change the size and shape of the individual structures as well as their spatial relationship with each other.   Cone-Beam CT (CBCT) scans can be acquired daily or weekly to assess these changes, and if they are deemed too large the treatment can be re-planned. In order to maximise the benefits of proton therapy it is essential to produce treatment plans that account for treatment uncertainties and anatomical changes.
Research Aims
The project will utilise state-of-the-art machine learning methods to model the uncertainties in the organ at risk (OAR) delineations and day to day anatomical changes to patients receiving proton therapy for head and neck cancer treatments. Different models will be developed for these two sources of uncertainty. Deep-learning based approaches will be used to simultaneously generate high-quality automatic delineations of organs at risk and to estimate and quantify the different types of uncertainty in the delineations (uncertainties in the training data, e.g. variability in the manual delineations, uncertainty in the model parameters). Statistical shape/deformation models and deep-learning based approaches will be investigated to model the variability in the day to day changes in OAR using routine imaging from patients acquired during the course of treatment.  These predictive models can then be used to forecast uncertainty in shape and position of OAR for a new patient over the course of proton therapy and ultimately guide robust proton treatment planning to take account of these changes.
Requirements
Applicants are expected to have a first degree in Computer Science or Biomedical Engineering or relevant Physical Sciences based subject passed at 2:1 level (UK system or equivalent) or above. Good working knowledge of Python and/or MATLAB is desirable. Some experience with machine learning, medical image analysis, or radiotherapy is also desirable.

Deadline May 15th 2020
To Apply: Please send a CV and Covering Letter detailing why you want to apply for this studentship and why you believe you are suitable for the studentship to Dr Jamie McClelland j.mcclelland@ucl.ac.uk