Philippa Grace McCabe - PhD Researcher
Research Project
My thesis is studying the diagnosis and predictive modelling of osteoarthritis through statistical and machine learning methods. Osteoarthritis (OA) is a degenerative bone disease that affects joints as a whole. OA is one of the most common diseases affecting people in old age with the prevalence in people 65 years and older range from 12% to 30%. The disease is also the most common form of arthritis to cause pain and mobility limitations. When diagnosising OA x-rays are taken. The scale to classify OA is the Kellgren-Lawrence scale grading form 0-4 depending on severity. A clinician usually analyses and classifies images for diagnosis. By using both humans and machines there is the potential for more reliable diagnoses in the future.
Research
Interests
I am interested in machine learning approaches and how interpretability can be maximised, allowing for results that can be easily understood and used with confidence. My previous work has looked at multitask learning and how it can be used to aid in drug design and discovery by making better use of QSARs.
Outputs
I have been working on mutlitask learning for drug design and discovery. I gave a talk at IJCNN about this.
My thesis is studying the diagnosis and predictive modelling of osteoarthritis through statistical and machine learning methods. Osteoarthritis (OA) is a degenerative bone disease that affects joints as a whole. I will be giving a talk at IDEAL on this research.
Publications
2019
P.G. McCabe, I. Olier, S. Ortega-Martorell, I. Jarman, V. Baltzopoulos, P. Lisboa. “Comparative Analysis for Computer-Based Decision Support: Case Study of Knee Osteoarthritis”. 20th International Conference on Intelligent Data Engineering and Automated Learning (IDEAL). Manchester, UK. Nov 2019. Paper Here
P. G. McCabe, S. Ortega-Martorell, I. Olier. “Benchmarking multi-task learning in predictive models for drug discovery”. International Joint Conference on Neural Networks (IJCNN). Budapest, Hungary. July 2019. Paper Here
Presentations
2020
Open Research Week Joint Research Cafe UoL, Edge Hill and LJMU. OA: Open Access and Osteoarthritis. February 2021. Oral Presentation
2019
IDEAL 2019 Conference, November 2019. Comparative Analysis for Computer-Based Decision Support: Case Study of Knee Osteoarthritis.
Seminar of the Department of Applied Mathematics, October 2019. Machine Learning and Healthcare: A quick-ish look at the bits and bobs I fill my time with.
IJCNN 2019 conference, Budapest, Hungary. July 2019. Oral presentation. Benchmarking multi-task learning in predictive models for drug discovery.
Joint Research Cafe University of Liverpool and LJMU, Liverpool. July 2019. Oral presentation.
LJMU Faculty Research Week, May 2019. Poster presentation. Diagnostic and Predictive Modelling in Osteoarthritis Using Machine Learning.
LJMU Doctoral Academy Conference, May 2019. Poster presentation. Diagnostic and Predictive Modelling in Osteoarthritis Using Machine Learning.
Biography
I attained my BSc in Mathematics in 2017 and my MSc in Data Science in 2018 at Liverpool John Moores University (LJMU).
I am currently in the final year of my applied mathematics PhD at LJMU.
My PhD is working with the OActive project to use statistical and machine learning tools to develop a diagnostic and predicitve model for osteoarthritis.
I work with the Department of Applied Mathematics, the Machine Learning Group at LJMU and the School of Sport and Exercise Science. My supervisors are Prof Paulo Lisboa, Prof Bill Baltzopoulos and Dr Ivan Olier-Caparasso.
Skills
I am proficient in using the R programming language, which I have evidenced in the use of machine learning methods for various research projects. I have used MATLAB and currently in the process of learning Python.
Previous Degrees
2018 - Data Science MSc, Liverpool John Moores University. Distinction.
2017 - Mathematics BSc (Hons), Liverpool John Moores University. 1st Class.
Outreach Activities
I have assisted in lab tutorials in the Department of Applied Mathematics at LJMU, specifically helping students use and better understand R/RStudio. From the feedback I have received, the students saw this as a positive experience and felt it did help them. Since September 2020 I have been employed as a teaching support officer, assisting with Engineering Mathematics at LJMU.
Funding
I am funded by the OActive project, funding scheme: H2020-SC1-PM-17-2017, with contributions by Liverpool John Moores University.
