AI Tool Integrates MRI and Ultrasound to Detect Endometriosis

Endometriosis is a chronic condition in which tissue similar to the uterine lining grows outside the uterus, causing pain, heavy bleeding, and infertility. Diagnosis often depends on surgical visualization, while MRI and ultrasound can miss disease features and vary with operator experience. Researchers have now developed an AI tool designed to identify signs of endometriosis from a single scan.

EndoFusion is a recent development from IMAGENDO, an ongoing collaborative study led by Adelaide University researchers. The tool is designed to consolidate pelvic MRI and transvaginal ultrasound into one decision framework. Its goal is to detect two major indicators of advanced endometriosis quickly and without surgery.

The framework integrates data from both modalities to overcome the tendency of each scan type to preferentially detect one marker. It produces a classification in 18 milliseconds following image input. The researchers report that this approach is intended to mitigate operator dependence and streamline access to appropriate imaging.

Development drew on four datasets containing more than 9,000 female pelvic MRI scans and more than 800 transvaginal ultrasound sliding scans. Investigators assessed the model’s ability to distinguish positive from negative cases and found it delivered a correct diagnosis 83% of the time. The framework outperformed all competing models evaluated.

The results were recently published in Artificial Intelligence in Medicine. Collaborators included Flinders University, Benson Radiology, Omni Ultrasound and Gynaecological Care, the University of Surrey, the McMaster University Medical Centre, and Mohamed bin Zayed University of Artificial Intelligence. Next steps will expand the dataset to include additional endometriosis markers to improve classification accuracy and broaden applicability.

“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 associate professor Jodie Avery, research co-lead of Chronic Reproductive Conditions in the Endometriosis Research Group at Adelaide University’s Robinson Research Institute.

“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.”

Related Links
Adelaide University