Machine learning can significantly aid in identifying risk factors from large biomedical datasets. In this seminar, we will discuss how we integrate machine learning with conventional epidemiological methods to identify adverse and protective risk factors for disease outcomes. We will present results from our study on predicting risk factors for cancer. Following this, we will share findings from a subsequent study conducted in collaboration with expert gynaecological oncologists and consumer members, aimed at discovering risk factors for ovarian cancer to enable earlier detection and inform new prevention strategies for this cancer, which currently has a poor prognosis due to late-stage diagnosis.
Dr Madakkatel is a Research Associate in the Nutritional and Genetic Epidemiology Research Group, specialising in the application of Machine Learning in Epidemiology and Public Health. His research focuses on feature selection and risk factor discovery, aiming to uncover critical insights that can inform public health interventions. He is passionate about both developing new methodologies and applying existing ones to extract valuable information from health data.
Having completed his Diploma and Bachelor’s degree in Computer Engineering, he pursued further education and received his M.Sc. degree in Information Technology, with a focus on Informatics. In 2021, he obtained a PhD in Data Science with a focus on Epidemiology/Public Health.
From a background in molecular biology research (PhD in Genetics and postdoctoral experience in genetics/physiology), Amanda transitioned from the ‘wet lab’ to join Professor Elina Hyppönen's Nutritional and Genetic Epidemiology Research Group, where she works on large cohort data projects identifying risk factors for conditions including cancer.
As a Research Fellow, and Project Manager of an MRFF-funded ovarian cancer project, Dr Lumsden works closely with consumer members, researchers, and gynaecological oncologists, and is passionate about finding ways to identify women at risk of ovarian cancer to help facilitate earlier detection, and discovering new risk factors that can inform on strategies to prevent its incidence.
