Leah Gerrard, Data Science and Knowledge Discovery Lab
Image: Leah Gerrard
What is the most rewarding aspect of your research?
The most rewarding aspect of my research is being able to benefit patient care and influence policy decision making for cancer patients in Australia. The potential for my research to have real-world impact motivates me to continue this important work.
What are the real-world applications of your research?
This research is exploring the cohorts and pathways for patients with cancer. Now that large, linked, complex health care datasets are becoming available, there is potential to leverage these datasets to better understand patients and their pathways. However, patients are very heterogeneous and may have very different pathways before cancer diagnosis and through treatment and follow-up care. By using deep learning to identify cancer subtypes and predict patient outcomes, it will provide information on different progression patterns for cancer patients which can assist in personalised care and targeting health policy so that it is more effective.
What ideas do you (or your team) have for future research?
We would like to continue to improve deep learning methods for patient subtyping and outcome prediction and apply these types of methods to different diseases in addition to cancer - such as diabetes, cardiovascular disease and stroke.
Are you involved in collaborative research?
This research is being done in collaboration with the Australian Government Department of Health. Having this collaboration provides us with expertise in developing research questions that will help answer current real-world health challenges. We also benefit from being able to access and analyse large linked datasets, which are essential in better understanding the patient journey through the healthcare system.
What inspired you to undertake a PhD in computer science?
Having completed my undergraduate degree in Medical Science, I had a keen interest in health research. I hadn’t initially considered computer science as a research path, but seeing the amount of health data being collected and stored sparked my interest in methods capable of analysing these huge amounts of information. Machine learning and AI are well equipped at learning from these large datasets and hold a lot of promise in improving healthcare in the future. Being able to apply machine learning methods to try and solve health problems has been a great fit for me as I am enjoying learning new skills in data science while being able to utilise my existing knowledge of health.
