AI could help us more accurately screen for breast cancer – new research

Sasun Bughdaryan/Unsplash Carolyn Nickson, University of Sydney; The University of Melbourne and Bruce Mann, The University of Melbourne

At least 20,000 Australian women are diagnosed with breast cancer each year. And more than 3,300 die from the disease.

To save women’s lives, we need to detect breast cancer early. Breast screening, which halves women’s risk of dying from breast cancer, is key to that.

A new Australian study published today in The Lancet Digital Health suggests AI could help improve how we screen for breast cancer.

How do we currently screen for breast cancer?

Since 1992, Australia has offered free breast X-rays, known as mammograms, every two years to women aged between 50 and 74. Just over half of eligible women participate.

Of the women found to have cancer, about 25% are diagnosed between the biennial screens. These “interval cancers” are often aggressive and, unfortunately, more likely to be fatal.

In some cases, a more sensitive screening test may have detected them earlier.

The role of AI

Australia’s BreastScreen program was established in response to several major clinical trials conducted between the 1960s and 1980s. The screening technology used by the program has not substantially changed since then.

Researchers are now exploring risk-adjusted screening, which tailors screening to women based on their risk, as a way to detect more cancers earlier. This may include programs offering different technologies for women at higher risk of developing breast cancer.

Currently, we generally assess cancer risk via questionnaires that help identify if a woman has any risk factors associated with breast cancer.

One risk factor is breast density which refers to how much glandular tissue is in the breast. As well as being a risk factor for breast cancer, the higher a woman’s breast density, the harder it is to detect cancer on a mammogram.

We can also use one-off genetic testing to identify women with a higher lifetime risk of developing breast cancer. This involves looking for high-risk gene mutations such as BRCA1 and BRCA2, which are associated with increased breast and ovarian cancer risk. Genetic testing can also help us estimate a person’s lifetime risk of developing breast cancer.

More recently, researchers have been investigating artificial intelligence (AI) as a new approach to assess breast cancer risk. A new Australian study, published in The Lancet Digital Health today, focused on a specific AI tool known as BRAIx.

What did the study involve? And what did it find?

This study used an AI tool, known as BRAIx, trained using BreastScreen Australia data to help radiologists assess mammograms.

The study assessed how well BRAIx predicted women’s risk of developing breast cancer in the next four years, among women who had a clear mammogram.

Of the 95,823 Australian women assessed, 1.1% (1,098) had developed breast cancer in the four years after they received a clear mammogram. Of the 4,430 Swedish women assessed, 6.9% had developed breast cancer within two years of a clear screen.

The study findings show that BRAIx scores were very useful for identifying women who were more likely to develop cancer one to two years after having a clear screen. Findings from the Australian dataset suggest BRAIx scores identified cancers found three to four years later, but with less accuracy.

These findings suggest BRAIx could help identify women who might benefit from additional tests. This may include an MRI (which uses a magnetic field to produce images of organs and tissue) or contrast-enhanced mammography (which uses an iodine dye to improve the visibility of a regular mammogram).

These findings reinforce a 2024 Swedish study that used an AI-based risk assessment to select women for additional testing. The researchers referred 7% of women to have a follow-up MRI, and 6.5% of were found to have cancers missed by mammograms.

Does the study have any limitations?

As with most studies, yes. Here are two.

• it’s difficult to compare BRAIx to genetic testing. This is because BRAIx is trained to find missed or emerging cancers over a four year period. In contrast, genetic testing identifies a person’s risk of developing cancer over their lifetime

• it might not use the best breast density data. This study found BRAIx more accurately predicts breast cancer risk compared to assessments based on breast density. But this breast density data was collected using a different tool to those used by the Breastscreen program. So this finding should be interpreted carefully.

So, where to from here?

The study adds to a growing body of evidence that AI risk assessment could help breast screening programs find cancers earlier.

BRAIx is now being trialled as part of the BreastScreen Victoria program, to help read mammograms. And other states are already using and evaluating different AI tools for reading mammograms.

So it may be time for Australia to conduct a national, independent review of these new tools. As part of a more risk-adjusted approach to breast screening, they could save lives.

Carolyn Nickson, Principal Research Fellow, Cancer Elimination Collaboration, University of Sydney; The University of Melbourne and Bruce Mann, Professor of Surgery, Specialist Breast Surgeon, The University of Melbourne

This article is republished from The Conversation under a Creative Commons license. Read the original article.

Read More........→March 10, 2026

Here’s why you might want to clean your headphones

Rina Wong (Fu), Curtin University

Whether it’s enjoying a podcast, listening to music or chatting on the phone, many of us spend hours a day using our headphones. One 2017 study of 4,185 Australians showed they used headphones on average 47–88 hours a month.

Health advice about headphones tends to focus on how loud sounds might affect our hearing. For example, to avoid hearing loss, the World Health Organization advises people to keep the volume at below 60% their device’s maximum and to use devices that monitor sound exposure and limit volume.

But apart from sound, what else is going in our ears? Using headphones – particularly in-ear versions such as earbuds – blocks the ear canal and puts the skin in contact with any dirt or bacteria they may be carrying.

Here’s what you need to know about keeping your ears clean and safe.

First, let’s take a look at your ear

Over-ear headphones cover the entire external ear – the elastic cartilage covered by skin that’s shaped to trap soundwaves. In-ear headphones (as well as hearing aids) are shaped to fit and cover the entrance to the external ear canal, which is called the concha.

Sound vibrations travel through the ear canal – which is S-shaped and a few centimetres long – to reach your ear drum.

Deeper parts of the ear canal produce earwax and oils. These help keep your skin healthy, hydrated and less vulnerable to infection.

Tiny hairs in the ear canal also help regulate temperature and keep foreign debris out. These hairs and earwax help trap and move small particles, shed skin and bacteria out of the ear canal.

Earwax is the ear’s self-cleaning method and we only tend to notice it when there’s too much.

Excessive buildup can block your hearing or even clog the mesh of your earpods. But don’t try to dig earwax out of your ears yourself. If you’re concerned, speak to a pharmacist or GP for advice.

How headphones can affect the ear’s bacteria

Healthy ear canals host a range of non-harmful microbes – mainly bacteria, but fungi and viruses too. They compete for space and nutrients, and this diversity makes it trickier for any potential pathogens (disease-causing microorganisms) to take hold.

But wearing headphones (and other in-ear devices such as hearing aids or ear plugs) may upset the balance between “good” and “bad” bacteria.

One 2024 study compared bacteria in the external ear canals of 50 people who used hearing aids and 80 who didn’t. The researchers found hearing-aid users – whose external ear canals are blocked for extended periods – had fewer types of bacteria than those who didn’t.

Another 2025 study looked at how using headphones (including over-ear, in-ear and on-ear) affected fungi and bacteria in the ear canal. It found using headphones was linked to a greater risk of ear infections, especially if people shared them.

This may because wearing headphones – especially in-ear devices – makes the external ear canal hotter and more humid. Trapped moisture is especially likely if you exercise and sweat while wearing headphones.

Higher humidity increases your risk of ear infection and discharge, including pus.

Wearing in-ear devices such as hearing aids or headphones for extended periods can also interfere with the ear’s natural “self-cleaning” function, aided by earwax.

So, what should I do?

Most of us need – or like – to wear headphones in our day-to-day routines. But for good ear health, it’s important to give your ears a break.

Allow your ear canals to “breathe” at different points throughout the day so they’re not constantly blocked and growing humid and hot.

You could also try bone conduction headphones. These don’t block the ear canal, because they transmit sound through your skull directly into the inner ear, without needing to block the ear canal. These can be expensive though. And while they allow our ears to breathe, high-intensity vibrations (high volume) can still damage hearing, so as with all headphones caution is required.

Other tips

Clean your devices regularly

Recommendations range from once a week to daily to after a physical workout.

For example, you can wipe them with a cloth or use a soft-bristled children’s toothbrush dampened with mildly soapy water. Blot dry with a paper towel and allow a few hours of drying before recharging or reuse.

But it’s best to follow your manufacturer’s guidelines. And don’t forget to clean the case and the body of your earbuds too.

Don’t use headphones when sick

If you have an ear infection, avoid using earphones as they may increase the temperature and humidity in your ear and slow recovery.

Watch for symptoms

If your ears become itchy, red or have discharge, stop using in-ear devices and seek medical advice.

Rina Wong (Fu), Research Fellow, Health Sciences, Curtin University

This article is republished from The Conversation under a Creative Commons license. Read the original article.

Read More........→March 10, 2026