This project is no longer listed on FindAPhD.com and may not be available.
Click here to search FindAPhD.com for PhD studentship opportunities(MCRC Non-Clinical) Risk prediction and early detection of breast cancer using Automated Whole Breast Ultrasound (ABUS) and mammographic images
Prof Sue Astley, Dr Martin Fergie, Dr Anthony Maxwell
No more applications being accepted
Competition Funded PhD Project (European/UK Students Only)
About the Project
Automated Whole Breast Ultrasound (ABUS) has recently been developed as a sensitive method for detecting early signs of cancer in women with breasts that contain a high proportion of radio-dense fibroglandular tissue, in whom standard x-ray mammography is ineffective. As personalised breast screening develops, it is expected that ABUS will be used in a significant proportion of women, but compared with x-ray mammography, few automated methods are available to support radiologists interpreting these images.
In this project we will employ state-of-the art machine learning methods to analyse both the complex three-dimensional data sets produced by ABUS and corresponding mammograms, with the aim of identifying women with a high risk of developing breast cancer (for whom preventive interventions would be appropriate), and those with early signs of the disease.
Data will be obtained from the Institut Jules Bordet in Belgium, which has internationally recognised expertise in ABUS and an existing database of over 3000 images, along with corresponding x-ray mammograms.
We have previously demonstrated that convolutional neural networks can be successfully trained to learn high risk patterns in two-dimensional x-ray mammograms with performance comparable to human experts.
In this project we will extend the methodology to three dimensional datasets, and develop deep learning methods to:
1. learn the relationship between high risk mammographic patterns and high risk ABUS appearances using paired data;
2. locate regions of ABUS images corresponding to subtle abnormalities using datasets with expert annotations; and
3. map abnormalities that are visible in ABUS but were not detected mammographically back to the corresponding locations in the mammograms to enable focused analysis of local mammographic patterns associated with a high probability of cancer.
This technology will facilitate both the development of Computer Aided Detection (CAD) for ABUS, and earlier detection of abnormalities in x-ray mammography.
Entry Requirement
Candidates must hold, or be about to obtain, a minimum upper second class (or equivalent) undergraduate degree in relevant subject. A related master’s degree would be an advantage.
In this project we will employ state-of-the art machine learning methods to analyse both the complex three-dimensional data sets produced by ABUS and corresponding mammograms, with the aim of identifying women with a high risk of developing breast cancer (for whom preventive interventions would be appropriate), and those with early signs of the disease.
Data will be obtained from the Institut Jules Bordet in Belgium, which has internationally recognised expertise in ABUS and an existing database of over 3000 images, along with corresponding x-ray mammograms.
We have previously demonstrated that convolutional neural networks can be successfully trained to learn high risk patterns in two-dimensional x-ray mammograms with performance comparable to human experts.
In this project we will extend the methodology to three dimensional datasets, and develop deep learning methods to:
1. learn the relationship between high risk mammographic patterns and high risk ABUS appearances using paired data;
2. locate regions of ABUS images corresponding to subtle abnormalities using datasets with expert annotations; and
3. map abnormalities that are visible in ABUS but were not detected mammographically back to the corresponding locations in the mammograms to enable focused analysis of local mammographic patterns associated with a high probability of cancer.
This technology will facilitate both the development of Computer Aided Detection (CAD) for ABUS, and earlier detection of abnormalities in x-ray mammography.
Entry Requirement
Candidates must hold, or be about to obtain, a minimum upper second class (or equivalent) undergraduate degree in relevant subject. A related master’s degree would be an advantage.
Funding Notes
The Studentship will cover an annual stipend (currently at £19,000 per annum), running expenses and PhD tuition fees at UK/EU rates. Where international student fees are payable, please provide evidence within your application of how the shortfall will be covered (approximately £17,000 per annum).
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.
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.
Open Days
Continue with Facebook