FindAPhD Weekly PhD Newsletter | JOIN NOW FindAPhD Weekly PhD Newsletter | JOIN NOW

Confidence, predictivity and biomarker detection to enhance the usability of breast cancer computer aided diagnosis


   School of Electronics, Electrical Engineering and Computer Science

This project is no longer listed on FindAPhD.com and may not be available.

Click here to search FindAPhD.com for PhD studentship opportunities
  Dr Hui Wang  No more applications being accepted  Funded PhD Project (Students Worldwide)

About the Project

This PhD project will investigate the usability of computer aided diagnosis systems by designing a reliable, prediction-specific, confidence measure and developing a deep learning architecture that can learn medical biomarker models.

This PhD project is a continuation of an EU Horizon 2020 project on breast cancer image analysis. Computer-aided diagnosis (CAD) is a computerized procedure for medical image analysis to provide a second objective opinion in interpretation of medical images to aid medical decision-making. CAD has the potential to save time and/or as a tool to provide an objective second opinion, although this potential has not been fully realized due largely to the usability problem. Two reasons for the problem can be identified. One is that CAD systems lack a reliable confidence measure for its prediction. Machine learning models usually output a confidence score for any prediction, but the correlation between the confidence and the probability of the prediction being true is generally low; in other words, the predictivity of the confidence score is low. Therefore, the availability of the confidence score does not increase the usability of CAD. Another reason for the usability problem is that the machine learning model for a CAD system is typically not a model of the biomarker of a disease, thus the model may only work well in the closed world implied by the data used to train the model.

This project will tackle this usability problem with a deep learning framework. We will research how to measure confidence of individual predictions and how to measure predictivity of the confidence in order to obtain a highly predictive confidence measure through optimisation. We will also research how to build machine learning models for biomarkers. Based on the research findings we will develop a breast cancer CAD system and validate it in collaboration with partners in a Northern Ireland hospital.

Objectives:

  1. Design of a confidence measure. Contextual probability will be considered.
  2. Design of a predictivity measure, which can be incorporated into the loss function of deep learning.
  3. Development of deep learning architecture that can learn models for disease biomarkers.
  4. Demonstration of research findings in a research prototype for computer aided diagnosis.

Academic Requirements:

A minimum 2.1 honours degree or equivalent in Computer Science or Electrical and Electronic Engineering or relevant degree is required.

For further information about eligibility criteria please refer to the DfE Postgraduate Studentship Terms and Conditions 2021-22 at https://go.qub.ac.uk/dfeterms

Applicants should apply electronically through the Queen’s online application portal at: https://dap.qub.ac.uk/portal/

Further information available at: https://www.qub.ac.uk/schools/eeecs/Research/PhDStudy/

Funding Notes:

This three year studentship, for full-time PhD study, is potentially funded by the Department for the Economy (DfE) and commences on 1 October 2022. For UK domiciled students the value of an award includes the cost of approved tuition fees as well as maintenance support (Fees £4,500 pa and Stipend rate £15,609 pa - 2022-23 rates to be confirmed). To be considered eligible for a full DfE studentship award you must have been ordinarily resident in the United Kingdom for the full three year period before the first day of the first academic year of the course. 

For candidates who do not meet the above residency requirements, a small number of international studentships may be available from the School. These are expected to be highly competitive, and a selection process will determine the strongest candidates across a range of School projects, who may then be offered funding for their chosen project.


References

Decision Support and Information Management System for Breast Cancer (DESIREE). H2020 PHC-30-2015, 1 Feb 2016 – 31 July 2019. http://desiree-project.eu
Andrik Rampun, Hui Wang, Reyer Zwiggelaar, Bryan Scotney, Philip Morrow (2018). Confidence Analysis for Breast Mass Image Classification. IEEE International Conference on Image Processing 2018.
Andrik Rampun, Philip J. Morrow, Bryan W. Scotney, Hui Wang (2020). Breast density classification in mammograms: An investigation of encoding techniques in binary-based local patterns, Computers in Biology and Medicine, 122:103842
PhD saved successfully
View saved PhDs