Surprisingly Useful AI Article Enhancements
Monash University researchers have adapted a consumer-focused Ai tool normally used to predict buyer behaviour to identify potential safety signals and repurposing opportunities for medicines.
The tool, Magnum Opus, uses association discovery algorithms to uncover patterns in large datasets. While it is commonly applied to retail and consumer analytics, researchers have redeployed it to analyse prescription records and detect links between medicines and health outcomes.
Analysing Pharmaceutical Data
The study, published in Clinical Pharmacology and Therapeutics, was conducted by the Centre for Medicine Use and Safety (CMUS) within the Monash Institute of Pharmaceutical Sciences and the Department of Data Science and Ai at the Faculty of Information Technology.
Researchers examined a 10 per cent sample of data from the Pharmaceutical Benefits Scheme covering over 300 million prescriptions between 2014 and 2024. The team focused on patients with three conditions – coronary heart disease, type 2 diabetes, and epilepsy – and reviewed all medicines prescribed before diagnosis.
Patterns emerged across the cohorts. In coronary heart disease, cholesterol-lowering and blood-thinning medicines appeared as expected, given their use in managing overlapping risk factors. In type 2 diabetes, antipsychotics, diuretics, and statins were linked to increased risk, while some medicines prescribed for Parkinson’s disease and osteoporosis appeared to lower risk. For epilepsy, the study confirmed known associations between antidepressants, antipsychotics, and increased epilepsy risk, but also identified a common blood pressure medicine linked to lower risk.
Discovering Hidden Links
Dr George Tan, Research Fellow with CMUS and co-lead author, said the analysis shows how prescription records can reveal unexpected insights.
“Medicine use patterns tell a story. By studying prescriptions, we can not only see how conditions are being treated, but also discover surprising links that may point to new risks, new protections, or even new uses for existing medicines,” he said.
Professor Geoff Webb from the Faculty of Information Technology, senior author of the study, said adapting data mining methods from consumer analytics has clear potential in healthcare.
“This study shows how big data can generate early clues about medicine safety and effectiveness in much the same way it helps predict consumer behaviour,” Webb said.
“Ai is helping us move from reactive to proactive healthcare. By harnessing the same algorithms that predict what people might buy next, we can begin to anticipate which medicines may work best for which patients – and which might pose hidden risks. It’s about using Ai not just to analyse data, but to generate new medical knowledge that can ultimately improve patient care.”
Dr Lynn Miller, co-lead author, added that association discovery algorithms are not intended to replace clinical research but to guide it. “By adapting these methods to healthcare, we can uncover meaningful relationships across millions of prescriptions, not to replace clinical research, but to help guide where that research should look next.”
Early Stage Research
The researchers emphasised the work is exploratory and designed to generate hypotheses rather than definitive conclusions. Future studies will test the timing and biological plausibility of the associations and examine whether the findings are consistent across different datasets.
The full study can be accessed here.
Last Updated on October 13, 2025 by Nick Ross
