Dr. Qiaoyun Xie is a reformed purist.
Not long ago, the geospatial scientist and professor in UWA’s School of Engineering saw little point in sullying her plant research with something as unconventional as artificial intelligence.
Today, she’s using AI to go beyond academic theory: analyzing vast amounts of satellite imagery of global vegetation hotspots to precisely monitor the day-to-day impacts of climate change.
“A few years ago, I actively pushed back against this idea,” says Dr. Xie. “I was a strong believer in traditional theory of plant biology and physiology—and I still am—but I thought traditional theory should be the primary, if not the only, way to study these things.”
“But by the time I got my PhD, everything had changed and I saw how powerful AI could be in my field. It was a real game changer for my research.”
“I think AI is really useful in complex areas like climate change, where it bypasses the assumptions that we as scientists make when we set up equations and instead draws analyses and questions from the data itself.”
Dr. Xie’s research uses satellite imagery, or what she describes as her “eyes in the sky,” as well as on-the-ground measurements, to track ecosystem dynamics. She studies vegetation color, seasonal changes, leaf growth, soil cover, and plant interactions with climate to gather evidence of variability—information that, at a scale enabled by AI, can help predict future impacts of climate change and inform mitigation strategies.
Since joining UWA from the University of Technology Sydney in May, Dr Xie has been working with the Western Australian Department of Biodiversity, Conservation and Attractions to study drought and heat resilience in the state’s jarrah forests, as well as partnering with the WA Agricultural Research Collaboration to monitor the impact of climate change on our vast rangelands. She says the opportunity for meaningful industry engagement was a major part of the appeal of moving to UWA.
“I don’t want to just publish papers,” says Dr. Xie. “I want my findings and conclusions to be used by decision-makers, because the stakes are compelling. Climate change is happening and it’s happening faster than we think, so we need to act now to adapt to these changes.”
Jake Kendrick is another UWA academic using AI to produce impactful research. Currently completing his PhD in medical physics at UWA, Mr Kendrick’s current project focuses on developing an automated segmentation framework for metastatic prostate lesions.
By Carrie Cox “In a nutshell, I trained an AI model to analyze hundreds of positron emission tomography scans (an imaging test used to diagnose the presence of disease in the body) and automatically identify sites of disease,” he explains. “For a radiologist to do the same thing, which is to look at these scans individually and physically circle the sites of disease, would take a very, very long time. The model I developed was able to do this completely automatically without any intervention from the radiologist, and I was able to show that tumor volume was actually a predictor of overall survival.”
 Picture:Dr. Qiaoyun Xie, School of Engineering |  Picture: Julianne Sparke, Industry Engagement Manager |  Picture:Mr. Jake Kendrick, PhD Candidate |
Mr Kendrick’s research, published in the European Journal of Nuclear Medicine and Molecular Imaging, will now be extended and refined to ultimately develop a fully validated AI model suitable for clinical practice.
“I think we’ll see that in the next couple of years. My goal is for the model to be available to everyone through open source software,” he says. “That means we’ll be able to speed up the analysis of these images because anyone will be able to segment these scans, and that will make it much easier to do research to identify predictive features of patient response.”
“Overall, we want to be able to predict in advance whether patients will respond to certain treatments and thus offer them the optimal treatment pathways, because once prostate cancer has spread to other parts of the body, five-year survival rates are very low. We want to give patients the best possible chance of survival.”
The research by Dr Xie and Mr Kendrick was supported by the UWA Data Institute, established two years ago to help transform data science research.
Julianne Sparke, Industry Engagement Manager, says the Institute is rapidly transforming the data science landscape in Western Australia and providing industry and government with clearer pathways to harness UWA’s data science expertise.
“We are creating new opportunities to address real-world challenges by transforming vast amounts of data of different types into useful knowledge,” Sparke says. “It’s about making data ‘talk’ so it can help solve problems.”
The founding sponsors of the UWA Data Institute are BCG X and IMDEX Limited.
Read the full issue of Uniview’s Winter 2024 edition [Accessible PDF 12MB]