Head and Neck cancers are the 6th most common cancer in the world with an increasing incidence and worsening outcomes. In the UK alone, at least 12,000 new cases are diagnosed each year (32 per day) with over 4000 related deaths per year (11 per day).
Reasons for this poor prognosis include late diagnosis and lack of specific genetic signatures that would allow prediction of behaviour and guide treatment. Furthermore, head and neck cancer readily infiltrate adjacent structures (such as bones, muscle, blood and lymphatic vessels, nerves) and can spread to neck lymph nodes and beyond.
Machine learning is a branch of Artificial Intelligence (Ai) and is being widely applied in cancer research as a useful diagnostic tool aiming to improve cancer diagnosis and prediction. Use of machine learning has been shown to remove subjectivity in and variability in pathological analysis allowing standardisation and a quantitative output which can play a key role in informing treatment decisions. It has been shown to be important in obtaining ‘big data’ from whole slide images (WSI) which may not be otherwise analysed.
This project aims automated prediction of the behaviour of head and neck cancers using digital features from the initial biopsy (which all patients undergo). It will explore the digital footprint of a range of HNCs (oral, oropharyngeal, laryngeal etc.). Cases will be identified using the local pathology archives and scanned using a high resolution digital scanner. Supervised learning will be provided using annotations for relevant pathological features (cancer grade, stage, invasive front, immune response, perineural and vascular invasion) and linked with clinical features such as age, gender, smoking and alcohol use, metastasis, recurrence and 5-year survival. In addition, Convolutional neural network (CNN) which is a class of deep learning will also be used in parallel to detect any cellular, nuclear and/or morphological features not routinely analysed by pathologists.
This training will be used to develop and optimise novel AI algorithms in collaboration with the Tissue Imaging Analytics lab at University of Warwick for automated detection of known features and to identify novel digital biomarkers of prognosis. The developed algorithms will be applied on a large unseen cohort to determine validity. Statistical correlation will be carried out using hazard and survival analyses.
This is an exciting and innovative multidisciplinary project which has the potential to revolutionise treatment of head and neck cancer patients. The collaboration between the two supervisor has been ongoing for over two years with numerous publications and experience of using AI in head and neck cancers. The proposed project and its potential results can significantly reduce the diagnostic time and allow automated prediction of future cancer behaviour. The methods can also be readily translated and applied to clinical and diagnostic practice to benefit patients suffering from this devastating disease.
Interested candidates should in the first instance contact Dr Ali Khurram ([email protected]
How to Apply:
Please complete a University Postgraduate Research Application form available here: http://www.shef.ac.uk/postgraduate/research/apply
Please clearly state the prospective main supervisor in the respective box and select Dentistry as the department.
Interviews are due to take place on Monday 25th March 2019.
The Faculty of Medicine, Dentistry and Health has received an allocation of three EPSRC studentships for 2019 entry from the Doctoral Training Partnership grant that is awarded to the University of Sheffield to fund PhD studentships in the EPSRC remit. These studentships will be 42 months in duration, and include home fee, stipend at RCUK rates and a research training support grant (RTSG) of £4,500.
Home/EU students must have spent the 3 years immediately preceding the start of their course in the UK to receive the full funding.