University of Leeds Featured PhD Programmes
Engineering and Physical Sciences Research Council Featured PhD Programmes
King’s College London Featured PhD Programmes

(EPSRC DTP) Image translation using multichannel generative adversarial networks for automatic tumour delineation


Project Description

Inter-observer error in tumour delineation is a significant source of uncertainty in the cancer patient pathway. Presently, the individual practice of the observing clinician dictates the treatment plan. While mandatory peer review helps to reduce this variation, it is time consuming and unstandardised. The variable nature of tumour location and physical anatomy means consistency becomes very difficult to achieve when performing delineations by hand. This project seeks to explore the application of various machine learning techniques to automate tumour volume delineation for radiotherapy of glioblastoma, aiming to shift the paradigm of radiotherapy treatment planning towards unbiased computational methods driven by Multi-Channel Generative Adversarial Networks (MCGADs) [1].

The ideal network should consist of a generator which will be trained to take a CT and/or PET image [2] of a brain with a Glioblastoma Multiforme (GBM) type tumour and manipulate the image such that it removes the tumour from the scan. The model will learn through the use of tumour free brain scan images. Through the analysis of areas with statistically significant variation during translation, a tumour can be identified. Simultaneously, a discriminator should be trained to differentiate between synthesised and real data. Ideally, the dynamic model could also be trained in reverse to allow the manipulation of images in order to create a tumour in the image from a healthy scan to be used as a training tool. While the model will clearly have clinical relevance, its applications will not be confined to oncology. The architecture should be able to modify data such that the high-level features are preserved; this form of image synthesis is applicable to any form of image and potentially adaptable to any form of input.

Entry Requirements:
Applications are invited from UK/EU nationals only. Applicants must have obtained, or be about to obtain, at least an upper second class honours degree (or equivalent) in a relevant subject.

For information on how to apply for this project, please visit the Faculty of Biology, Medicine and Health Doctoral Academy website (https://www.bmh.manchester.ac.uk/study/research/apply/)

Funding Notes

EPSRC DTP studentship with funding for a duration of 3.5 years to commence in September 2020. The studentship covers UK/EU tuition fees and an annual minimum stipend £15,285 per annum. Due to funding restrictions, the studentship is open to UK and EU nationals with 3 years residency in the UK.

As an equal opportunities institution we welcome applicants from all sections of the community regardless of gender, ethnicity, disability, sexual orientation and transgender status. All appointments are made on merit.

References

[1] Varghese Alex; Mohammed Safwan K. P.; Sai Saketh Chennamsetty; Ganapathy Krishnamurthi. “Generative adversarial networks for brain lesion detection”, Proceedings Volume 10133, Medical Imaging 2017: Image Processing; 101330G (2017)

[2] Bi L., Kim J., Kumar A., Feng D., Fulham M. (2017) “Synthesis of Positron Emission Tomography (PET) Images via Multi-channel Generative Adversarial Networks (GANs)”. In: Cardoso M. et al. (eds) Molecular Imaging, Reconstruction and Analysis of Moving Body Organs, and Stroke Imaging and Treatment. RAMBO 2017, CMMI 2017, SWITCH 2017. Lecture Notes in Computer Science, vol 10555. Springer, Cham


Email Now

Insert previous message below for editing? 
You haven’t included a message. Providing a specific message means universities will take your enquiry more seriously and helps them provide the information you need.
Why not add a message here
* required field
Send a copy to me for my own records.

Your enquiry has been emailed successfully





FindAPhD. Copyright 2005-2020
All rights reserved.