University of Adelaide researchers hope to develop an artificial intelligence algorithm that can eventually diagnose endometriosis via ultrasound, without the need for invasive surgery.
One in nine Australian women will be diagnosed with endometriosis by the age of 44. The painful condition currently requires keyhole surgery to diagnose accurately, which means diagnosis is frequently delayed.
“Sometimes it takes up to seven years for women to get a diagnosis for endometriosis, because they have to wait for the surgery,” says Dr Jodie Avery, a senior research fellow at Adelaide’s Robinson Research Institute, and project manager of the newly funded study, called Imagendo.
“We’re looking at combining ultrasound scans with MRI scans using artificial intelligence, to create a new algorithm to predict whether a woman has endometriosis or not,” says Avery.
It’s common practice to have a transvaginal gynaecological ultrasound scan when seeking a diagnosis for endometriosis. MRI scans are less common, but still used. The researchers will collect both ultrasound and MRI data from patients, along with their surgical outcomes if they’ve gone on to have endometriosis surgery, to develop an algorithm that can predict the likelihood of endometriosis.
“After we create the algorithm, we’re going to test it using a diagnostic test accuracy study,” says Avery. The algorithm will be checked with two cohorts, each with 200 women.
The researchers want it to become an ordinary feature of ultrasound scans. “Hopefully we can design something that we can incorporate into ultrasound machines later on,” says Avery. “So when a woman has a pelvic ultrasound, it could predict whether she has endometriosis or not.”
The $1.9 million grant, awarded by the Medical Research Future Fund, will provide funding for three years, at which point the researchers hope to have the algorithm tested and ready to develop into a diagnostic product.
“It will reduce diagnostic delay for women, which is really important,” says Avery.
“It will also increase their accessibility to diagnosis, and it will improve mental health by validating a patient’s pain experience. We can also look at preventative interventions for chronic pain and infertility.”