Cervix cancer is a growing problem in worldwide. Thanks for continuous improvements in cancer treatments, such as radiotherapy, many women survive their disease but they can also experience side effects from their treatments. For example, patients can experience considerable gastro-intestinal side effects, bladder inflammation as well as fatigue and weakness.
The cervix and uterus can display large motion between treatment days. Therefore, we treat with a large safety margin to ensure we cover the disease. However, this also means that we potentially include a large amount of normal tissue in the treatment volume. This results in the side effects a patient experiences, particularly bowel toxicity.
One way to reduce these side effects is with more precise targeting of an individual patients’ disease, but this can be difficult visually on medical images, particularly CT. MRI offers more possibilities, with a wider range of image contrasts available and the possibility of functional image sequences. However, it is difficult to interpret all this information qualitatively.
This project aims to apply advanced image characterisation using a radiomics approach to better characterise a patient’s disease on multi-parametric MRI. The project will combine image processing and recognition techniques, statistics and machine learning along with mpMRI data and clinical patient and tumour characteristics.
Briefly, the student will:
Set-up a radiomics pipeline using pyradiomics
Investigate the stability of features for reproducibility and repeatability on the available mpMRI images
Propose a suitable methodology to capture radiomics information from multiple mpMRI images from one time-point
Investigate radiomics differences between normal cervix/uterus tissue and disease (utilising contours drawn by a clinician to separate these tissues)
Investigate how radiomics parameters change over treatment using the additional imaging time-points available for each patient. This may provide information on how a patient’s disease is responding to treatment.
Candidates are expected to hold (or be about to obtain) a minimum upper second class honours degree (or equivalent) in physics, computer science or a related discipline. Experience with programming, especially python is essential, and previous experience working with medical imagine data is beneficial.
For international students we also offer a unique 4 year PhD programme that gives you the opportunity to undertake an accredited Teaching Certificate whilst carrying out an independent research project across a range of biological, medical and health sciences. For more information please visit
http://www.internationalphd.manchester.ac.uk
