Dr. Xie Qiaoyun is a reformed purist.
Until recently, the geospatial scientist and lecturer at the University of Washington's College of Engineering hardly saw the value in tarnishing his plant research with something as nontraditional as artificial intelligence.
But now she's using AI to go beyond academic theory, analysing vast amounts of satellite imagery of vegetation hotspots around the world to precisely monitor the day-to-day effects of climate change.
“A few years ago, I was actively opposed to that idea,” Dr. Xie says. “I believed in traditional plant biology and physiology theory, and I still do, and I thought that traditional theory should be the main way, if not the only way, to study these things.
“But by the time I graduated with my PhD, everything changed and I realized how powerful AI could be in my field. It has literally been a game changer for my research.”
I think AI can really help in a complex field like climate change. It essentially avoids the assumptions that scientists make when they develop equations, and derives its analysis and questions from the data itself.”
Dr. Xie's research uses satellite imagery and field measurements, which she calls “eyes in the sky,” to track ecosystem dynamics. She looks at vegetation color, seasonal changes, leaf growth, ground cover, and plant-climate interactions to capture evidence of change. This information, at a scale enabled by AI, helps predict future impacts of climate change and inform mitigation strategies.
Since moving to UWA from the University of Technology Sydney in May last year, Dr Xie has been working with the Western Australian Department of Biodiversity, Conservation and Tourism to study the drought and heat tolerance of the state's jarrah forests, as well as partnering with the Western Australian Agricultural Research Collaboration to monitor the impacts of climate change on vast areas of rangeland. Dr Xie says the opportunity for meaningful industry engagement was a big attraction in his move to UWA.
“I don't just want to publish papers,” says Dr Xie, “I want my findings and conclusions to be used by decision makers, because this is a real issue: (climate) change is happening, it's happening faster than we think, and we need to act now to adapt to that change.”
Another UWA academic using AI to carry out impactful research is Jake Kendrick. Currently completing his PhD in Medical Physics at UWA, Kendrick is working on a project developing an automated segmentation framework for metastatic prostate lesions.
By Carrie Cox “In a nutshell, I trained an AI model to analyse hundreds of PET scans (an imaging test used to diagnose where disease is in the body) and automatically identify sites of disease,” he explains. “It would take a very long time for a radiologist to do the same thing and look at those scans individually and physically enclose the sites of disease. The model I developed was able to do it completely automatically, without the intervention of a radiologist, and show that tumour volume is indeed a predictor of overall survival.”
 imageDr. Qiaoyun Xie, Faculty of Engineering |
 imageBy: Julianne Sparke, Industry Engagement Manager |
 image: Jake Kendrick, PhD candidate |
Kendrick's research has been published in the European Journal of Nuclear Medicine and Molecular Imaging, and the research will be expanded and refined, ultimately leading to the development of a fully validated AI model suitable for clinical practice.
“I think that will happen within the next two years, and my goal is to make this model available to anyone through open-source software,” he says. That would allow anyone to segment scans, speeding up image analysis and making researchers' work that much easier to identify features that predict patient responses.
“Overall, what we want to do is predict in advance whether a patient will respond to a particular treatment and offer them the optimal treatment pathway, because once prostate cancer has spread to other parts of the body, the five-year survival rate is very low. We want to give patients the best chance of survival.”
The research conducted by Dr Xie and Mr Kendrick is supported by the UWA Data Institute, which was established two years ago to help transform data science research.
Industry Engagement Manager Julianne Sparke says the institute is rapidly transforming the data science landscape in Western Australia, providing industry and government with a clearer pathway to leverage UWA’s data science expertise.
“We're creating new opportunities to address the real-world challenge of transforming vast amounts of disparate data into meaningful knowledge,” Spark says. “The goal is to make the data speak so it can help you solve problems.”
Foundation sponsors of the UWA Data Institute are BCG X and IMDEX Limited.
Read the full Uniview Winter 2024 issue [Accessible PDF 12MB]
