Within the construction industry, the issue of silicosis has been in the spotlight in recent years, specifically the danger it poses to many workers on building sites, with many urgently calling for better diagnostic tools.
Silicosis, often called a ‘silent lung disease’, is a chronic, potentially fatal, lung condition caused by inhaling silica dust, a common mineral found in sand, concrete, and stone.
In July 2024, the cutting and use of engineered stone was banned due to the risk of silicosis. But for any workers in mining or on tunnelling construction projects, like Sydney Metro West, Melbourne’s Metro Tunnel or Brisbane’s Cross River Rail, the dangers persist.
Fortunately, there’s now hope on the horizon thanks to groundbreaking research at the University of New South Wales (UNSW), where a team of scientists has developed a revolutionary rapid breath test that could catch the disease far earlier than current diagnostic methods.
The new diagnostic tool can analyse a person’s breath for signs of silicosis, even at its very earliest stage, within a matter of minutes. The test combines mass spectrometry—a scientific technique that analyses molecules—and AI to rapidly detect silicosis from breath samples, providing a non-invasive diagnostic tool for at-risk workers.
If the UNSW research lives up to its early promise, it could be a game changer in the working conditions and health of construction workers. Most importantly, it could provide a potentially life-saving shift in how occupational lung diseases are managed.
The research project has been underway over the past two years, led by Professor William Alexander Donald and Conjoint Professor Deborah Yates. Merryn Baker, a PhD candidate in the UNSW’s School of Chemistry, is one of the lead researchers.
“The main issue about silicosis is that it’s usually diagnosed at a late stage because the technologies we currently use, like X-rays, CT scans and lung function testing, can only pick up the disease much later once it’s visible or has caused significant lung damage,” Baker says.
“Once it’s chronic, it can have a huge impact on a person’s life and there’s only limited treatment options, or it could be fatal. So the need to diagnose it early, to limit exposure and prevent progression, is crucial.
“With the work we’ve been doing, the potential to now have earlier diagnoses and intervention can make a huge difference to the health outcomes, treatment options and long-term wellbeing.”
Unlike X-rays and CT scans that need to be conducted in a clinical setting, this new test could be deployed for mass worker screenings onsite and provide results within minutes. It also does not involve doses of radiation as X-rays do.
“Our method is very sensitive, and it can detect molecules down to parts per trillion, so it has an amazing capacity to pick up things that are at very small concentrations, which is what we can see with disease biomarkers in the breath,” Baker explains. “This is a disease where early detection is vital.”
The new detection device itself is surprisingly simple. “It’s a benchtop instrument that is only about one metre squared,” Baker explains. “It only takes one or two breaths into a plastic bag, with the contents then analysed by the machine. It takes one minute to run the sample and less than a minute for the AI to analyse the data. It’s all very fast.”
The technology can also be deployed on worksites. “There are some logistical considerations, like requiring nitrogen gas, but there’s no specific need for it to be in a clinical setting,” Baker says.
Historically, diagnosis of silicosis has lagged behind exposure by decades. According to Lung Foundation Australia, many cases of respiratory disease only surface long after workers have retired.
The journey to this new breakthrough has been full of challenges, Baker admits.
“When you’re working with AI and machine learning, you often don’t know if it’s going to work. The molecules we’re detecting are at such low concentrations that it’s hard to know if we’re picking up meaningful data until we run the AI algorithm. So when we saw it working with such high accuracy, it was a really great surprise.”
As the research project has another year to run as more testing is completed, there’s more work to be done before the device becomes widely available. Regulatory approval for medical devices—especially those involving AI—can take lengthy periods of time.
“In terms of instrumentation, the mass spectrometry devices are already available, but we need to do larger studies and get strong correlations between biomarkers and disease for approval, but we are getting closer,” Baker says. She remains optimistic, however, the device could be available within the next 12 to 18 months.
Baker reveals it’s been the important real-world impact the end result could have that has kept the UNSW team highly motivated by their project. “This test could help prevent a disease that has caused so much harm in people’s lives,” she says. “If we can help construction workers stay healthy and avoid the devastating consequences of silicosis, then that’s a massive achievement.”
