It’s estimated that more than 190 million women worldwide have endometriosis, a chronic illness that causes uterine tissue to grow outside of the uterus to painful effect. Symptoms can range from abdominal pain and heavy periods to fatigue and infertility.
Beyond a lack of research into women’s reproductive health, let alone endometriosis, the biggest hurdle many face is accurate diagnosis, meaning that figure of 190 million is likely far higher than we know.
An emerging AI tool, from Adelaide University, however, shows early promise of improving the process, with researchers finding it can identify two major indicators of advanced endometriosis from pelvic scans in just 18 milliseconds.
Currently, ultrasound and MRI are used for detection, with each method scanning for different signs of the disease. This can mean patients only receiving one type of scan may be disadvantaged depending on the signs of the condition they have. Formal diagnosis requires going one step further with surgery to visually identify lesions, which proves costly and invasive.
“Current scanning methods each have their own strengths when it comes to detecting two common markers that indicate the likelihood of endometriosis and patients will often only have access to one of them,” says study author Associate Professor Jodie Avery, Research Co Lead of Chronic Reproductive Conditions in the Endometriosis Research Group at Adelaide University’s Robinson Research Institute.
“This means that some patients could be disadvantaged if they are scanned by the less optimal option for their particular signs. Some of the imaging tools also rely on operator experience and can be costly.”
This is where the AI tool, EndoFusion, shows its greatest potential. Still in early stages of development, the tool analysed thousands of scans, both ultrasound and MRI, to build the framework to better identify cases of the disease.
“We looked at the how well EndoFusion was able to distinguish between positive and negative cases of endometriosis and found it was able to provide a correct diagnosis 83% of the time, which is more accurate than all competing models,” said lead author Dr Yuan Zhang from Adelaide University’s Robinson Research Institute and the Australian Institute for Machine Learning.
The study will continue to advance its AI capability and the accuracy of EndoFusion by expanding the data set to include additional endometriosis markers. It’s promising, but yet to reach clinical diagnostic stages.
“The development of accurate, non-invasive early diagnostic methods is critical to shorten the diagnostic timeline and reduce associated costs,” said Associate Professor Avery.
Related stories
Native ad body.
Native ad body.
Native ad body.
Share this post