Researchers are developing a less invasive way of diagnosing endometriosis using a combination of ultrasound, magnetic resonance imaging (MRI) and artificial intelligence, that could be available within the next two years.
For decades, keyhole surgery has been considered the only way to properly diagnose the condition which affects one-in-nine Australian women and those assigned female at birth by the age of 44.
Endometriosis occurs when tissue that is similar to the lining of the uterus grows elsewhere in the body, causing painful symptoms and sometimes infertility.
Barriers in accessing specialist care have meant those with the condition end up waiting on average between seven and 12 years to receive a diagnosis.
The University of Adelaide's 'Imagendo' research team is using artificial technology to combine the diagnostic capabilities of pelvic ultrasounds and MRIs to better identify endometriosis lesions and develop ground-breaking algorithms for diagnosing the condition.
Gynaecologist and Fertility Specialist Louise Hull said if successful, the diagnosis model would be rolled out worldwide with an aim of having it available in clinics within the next two years.
"The ultrasound machines as well as the MRIs have just got better and better over time, but the techniques have also got better," she said.
"What's happened is there has been an interest in endometriosis and funding for endometriosis that we have just never had before.
"It means that we can have really well-qualified sonographers whose main role is to do endometriosis scanning and they can see lesions and ligaments outside the uterus which a standard gynaecological scan just wouldn't detect."
Last November the federal government added pelvic MRI scans for endometriosis and other conditions that affect fertility to the Medicare Benefits Scheme — prior to this MRIs were deemed too expensive to use as a diagnostic tool.
The European Society of Human Reproduction and Embryology (ESHRE), the National Institute for Health and Care Excellence (NICE) and the Royal Australian and New Zealand College of Obstetricians and Gynaecologists (RANZCOG) have all recently changed their criteria guidelines for diagnosing endometriosis.
"They have all agreed that very specialist scans for endometriosis either ultrasound scans or MRIs can often pick up endometriosis quite accurately, although a negative scan can't rule it out," Professor Hull said.
"We've be able to work with the Australian Institute of Machine Learning in a big collaboration where we can use lots and lots of scans, ultrasounds and MRIs and put all of the digital data together to add up the diagnostics to get much more accurate results for people, so they don't have to have surgery as a first step."
Protecting fertility
Professor Hull said research had shown about one third of people who had diagnostic surgery for endometriosis did not need the procedure.
"The second thing is — it's a pain trigger and if you have pain it will up-regulate all of the neural responses in your body which includes back pain, pain in the pelvis, stabbing pain, headaches and fatigue and it will trigger that again," she said.
In other cases the diagnostic surgery revealed more extensive endometriosis than expected.
"They would have to stop the surgery and plan another surgery a few weeks down the track and have bowel surgeons or prepare your body differently, so you would end up having two surgeries," Professor Hull said.
The improved diagnostic model also has the potential to reduce infertility among those with endometriosis by providing them with the opportunity to be diagnosed earlier and freeze their eggs at a younger age.
"If we know that you've got endo, people can actually protect their fertility when they're younger," Professor Hull said.
'Making a difference'
Senior Research Fellow and Imagendo Program Manager Jodie Avery said they had been using data from ultrasounds, MRIs and operation notes donated by women to create the diagnostic algorithm.
"We are really interested if anyone wants to donate their scans, particularly if they've had a transvaginal endometriosis scan, because they are the scans that we actually really need," she said.
The research team has also been recruiting women to test that the algorithm works.
"It's pretty amazing knowing that we are really going to be making a difference in the lives of women with endometriosis because they quite often have to go through many different surgeries just to find out if they've got the condition or not," Dr Avery said.
"If we can go in and say, 'here's a really easy, non-invasive way that doesn't cause any pain and doesn't cause taking too much time off work or education', it's just going to make such a difference to the lives of women."
The team is exploring a variety of ways that the model could be rolled out once the algorithm has been tested and approved for use.
It could be used on an online platform, where patients and doctors can upload the imaging and get a probability score around the likelihood of endometriosis or the algorithm could be embedded into ultrasounds for a real-time diagnosis.
Like many others, Belle Burgess long suspected she had endometriosis but it was a long wait to be officially diagnosed through laparoscopic surgery.
The 23-year-old said the availability of a less-invasive diagnostic method would be "life-changing" for people with endometriosis as it would provide them with a better understanding of the extent of their condition before going in for surgery.
"The surgery itself was quite an intense surgery, but having the aftermath of nerve damage and the recovery time affected my quality of life immensely, like I have never had before," she said.
"Being able to know that I have endometriosis before heading into such a big surgery and experience would be amazing so that I can prepare better for it, also to know how I can manage it before the surgery.
"And obviously saving the money, you know surgeries aren't cheap and I don't think they are affordable for a lot of people who have endometriosis. It's not accessible to majority of women."
Changing the 'established norm'
Currently in Australia, there is a lack of sonographers who are trained in identifying endometriosis through ultrasound.
"We have been a little bit slow on the uptake," Specialist Sonographer Alison Deslandes, who is also a PhD student with the study, said.
"In part when you've got an established norm which is that a pelvic ultrasound is done by doing a scan of the uterus and a scan of the ovaries, it takes a bit of time for people to change their thinking and move beyond that.
"We also have a problem where in Australia scans are mostly done through the private sector and funded through Medicare and Medicare pay a fairly minimal rebate really for a pelvic ultrasound and people are disincentivised to then extend the scan which is what they need to do to look for endometriosis."
Ms Deslandes said there was strong demand amongst sonographers to learn how to specifically scan for endometriosis, with an increasing number of referrals coming through for the scan.
"The learning curve is not that great to start to learn to scan for endometriosis and a few studies have found that within about 50 scans most people can get to a level of confidence doing that," she said.
