20/03/2020

VC International Scholarship in Combination of Deep Learning and Traditional Methods for Lung Tumour Image Segmentation

Save to favorites

  • ORGANISATION NAME
    Cardiff University
  • ORGANISATION COUNTRY
    United Kingdom
  • DEADLINE DATE
    01/06/2020
  • RESEARCH FIELD
    Formal sciences
    Natural sciences

Outline

According to World Cancer Report by WHO, Lung cancer is the most common cancer type worldwide in terms of both incidence (2.1 million new cases in 2018) and mortality (1.8 million deaths in 2018). Early diagnosis and treatment can significantly increase the rate of survival. Brachytherapy is one of the most effective treatments for lung cancer, where a sealed radioactive source is put near or inside the tumour. The success of brachytherapy is heavily dependent on accurately outlining the target tumours and the surrounding organs at risk in the MRI/CT images. Usually, the contours are drawn manually by physicians with the aid of treatment planning systems. However, even an experienced physician can find it difficult to distinguish between tumours and normal tissues, because tumours are heterogeneous and cohesive to the surrounding tissues while their pathological features are highly similar to that of the surrounding tissues. Therefore, manually drawing contours can be tedious, time-consuming, and error-prone. Some automatic image segmentation methods exist, but the accuracy is not good enough for practical applications. In this project, we will combine traditional methods with deep learning to develop new image segmentation methods with high accuracy.

RESEARCH CHALLENGE

This is an interdisciplinary research project. The main research work is related to computer science, but medical knowledge is also used in this project. The PhD student will develop novel techniques about deep learning, computer vision, and medical imaging. S/he will also have an opportunity to explore the state-of-the-art AI techniques for lung cancer treatment.

RESEARCH ENVIRONMENT

We have a long-standing, strong and dynamic research culture which is organised into specialist research sectors. Specialise in the internationally recognised Visual Computing sector, work with exceptional computing resources equipment and alongside our research staff who are leaders in their areas of expertise. Our staff publish widely, edit and review for international journals, act as external examiners both within the UK and internationally, provide consultancy and secure research funding. Students join the Doctoral Academy Programme to develop their research and professional skills, and receive support to attend workshops and conferences to maximise exposure to the wider research context.

What is funded

Scholarship covers:

- overseas tuition fees;

- annual stipend (£15,285 for 2020/21);

- research consumables, training, conference travel.

Students earn additional income supporting teaching.

Duration

3 years, subject to satisfactory progress

Eligibility

Applicants must:

- be nationals of (or permanently domiciled in) the world’s Least Developed and Other Low Income Countries based on the DAC list of ODA Recipients 2020. See (View Website)

- meet the academic criteria;

- be liable to pay overseas tuition fees;

- not have been awarded another scholarship covering both tuition fees and stipend.

If you have either a tuition-fee only or stipend only award, you are eligible but will not receive double-funding.

How to Apply

Visit the website to apply for qualification Doctor of Philosophy in Computer Science & Informatics, mode of study Full-time, with start date 1 October 2020. In the research proposal section of your application, specify the project title and supervisors of this project. In the funding section, enter " VC International Scholarship".

 

 

 

Disclaimer:

The responsibility for the funding offers published on this website, including the funding description, lies entirely with the publishing institutions. The application is handled uniquely by the employer, who is also fully responsible for the recruitment and selection processes.