Dr Noorul Amin

With a background in applying machine learning and data analytics to biological challenges, Noorul is working to improve how AMR-related data is collected, integrated and analysed within and across sectors. His research will lay the groundwork for the development of predictive models to help industry make data-driven decisions regarding AMR risks and management.

Q: How did you become interested in AMR research?

Noorul: I’m more drawn to biological problems, where data science can make a bigger impact. That type of research gives me a sense of purpose.

After earning a bachelor’s degree in computer science and a master’s in computer engineering with a focus on data science and big data analytics, I moved to Australia for a PhD at La Trobe University. I used machine learning to classify non-coding RNAs and uncover their potential significance in diseases like cancer. After my PhD, I joined WEHI in Melbourne, where I worked on acute myeloid leukemia. I had a few breaks between my degrees, when I worked as a back-end developer in industry.

Given my background in data science and bioinformatics, working with SAAFE felt like a logical next step. It wasn’t a drastic change – more like a natural transition that allowed me to continue applying my expertise in data science, machine learning and predictive modelling to biological challenges.

Q: You’re working on the ‘Digital Transformation to Prevent Pathogen Resistance and Improve Food Security’ project in the SAAFE Analytics Program. How does it leverage your data science expertise?

Noorul: This project has several components. Engagement with industry is critical. We need to build trust and collaborate with SAAFE partners, stakeholders and government departments so we can understand their needs.

First, we’re focusing on the governance around the data. What are the sensitivities and privacy requirements of each sector? How do we protect the data and ensure that it’s trustworthy? We’re developing a data code, a framework that sets out clear policies and protocols so sectors can work together more easily.

Then we need to define the data landscape. We’re conducting workshops with people in each sector to understand their data environment. What kinds of data do they store and collect? How is it used – does it move around or stay in one place?

Then we need to develop data vocabularies to standardise terminology across the sectors. Each industry uses different naming conventions and metrics, which makes it difficult to integrate their data. For example, the water and horticulture sectors use different units to measure antimicrobial use, even though they’re interconnected.

Unfortunately, microbes don’t care which sector you’re in. So we need to take a holistic approach to understand AMR and how it spreads.

Q: How will you use artificial intelligence?

Noorul: Once we’ve completed the foundational aspects of the project, we’ll integrate data from real sources. Then we can use machine learning, a subset of AI, to analyse patterns in the data. It will help us make predictions – such as what microbial species has the potential to cause the next pandemic, what sectors are at risk, or where an outbreak might occur.

Understanding these patterns is important, and it’s not possible by looking at isolated lab samples. We need a data-driven approach.

Q: Is there a particular aspect of your work that you find especially challenging or exciting?

Noorul: The most exciting – and challenging – part is applying AI and machine learning to revolutionise AMR monitoring in Australia.

Microbes are very smart. Superbugs develop resistance in ways that are difficult to understand. Unlike humans, bacteria can transfer genes laterally through conjugation and horizontal gene transfer. This makes AMR unpredictable and complex.

As a data scientist, I think it's very exciting to use AI and machine learning to outsmart the superbugs. Without these tools, we wouldn't be able to stay ahead of AMR.