Estimated reading time: 4 minutes
Diagnosing endometriosis can be a lengthy process, but researchers are developing an artificial intelligence tool that could eventually help clinicians identify the condition from a single pelvic scan.
The emerging AI framework, called EndoFusion, correctly distinguished between positive and negative cases of endometriosis 83% of the time in a recent study and produced results in just 18 milliseconds.
Researchers at Adelaide University developed the tool through IMAGENDO, an ongoing collaborative study focused on improving endometriosis diagnosis. The findings were recently published in Artificial Intelligence in Medicine.
The technology is still in the early stages of development, but researchers say it could address a limitation in endometriosis imaging: MRI and ultrasound can have different strengths when detecting signs of the disease.
“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,” said 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.
“Our AI tool can help address these shortcomings by combining data from both imaging tools, giving the framework the knowledge it needs to detect both signs of endometriosis through a single scan more effectively and efficiently.”
How EndoFusion Works
Researchers developed the AI framework using information from four datasets containing more than 9,000 female pelvic MRI scans and more than 800 transvaginal ultrasound sliding scans.
EndoFusion was trained to identify two major indicators of advanced endometriosis by combining MRI and ultrasound data.
“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.
“This is a positive step forward and moves us closer to a future where an AI tool can help clinicians to provide a faster, more accurate diagnosis without the need for surgery.”
Why This Matters for Nurses
More than 190 million women worldwide have endometriosis. Symptoms can include abdominal pain, heavy periods, bloating, anxiety, fatigue and infertility.
The condition can be difficult to diagnose and may rely on visually identifying lesions through surgery, a process researchers describe as slow, costly and risky.
“The development of accurate, non-invasive early diagnostic methods is critical to shorten the diagnostic timeline and reduce associated costs,” Avery said.
For nurses, the research is notable because patients with suspected endometriosis may experience persistent symptoms while seeking answers. If ultimately validated for clinical use, a non-invasive imaging tool could affect how nurses support patients through evaluation and the path toward diagnosis.
What Comes Next
Researchers plan to expand the dataset to incorporate additional markers of endometriosis to improve the tool’s classification accuracy.
“We envisage that clinicians will be able to use these tools to help make a determination about the presence of endometriosis from a single scan,” Zhang said.
“It could also potentially provide insights for research into other disease that require the use of multi modal imaging such as gynecological disorders, prostate and breast cancer and fetal abnormalities.”
The research involved collaborators from Flinders University, Benson Radiology, Omni Ultrasound and Gynaecological Care, the University of Surrey, McMaster University Medical Centre and Mohamed bin Zayed University of Artificial Intelligence.
For now, the 83% accuracy rate represents an early step rather than a replacement for existing diagnostic approaches. Further development and validation could move researchers closer to a faster, non-invasive approach to endometriosis diagnosis.
