Early diagnosis and treatment planning for Endometriosis
Endometriosis is not well understood and women often suffer from debilitating symptoms
Area of work
Endometriosis
Imaging modality
MRI, Ultrasound
Size of data set
100
Project lead
Dr Ippokratis Sarris
Number of NHS partners
1The average length of time for a woman to receive a diagnosis from the onset of symptoms in the UK is currently 8 years.
Unmet need
Endometriosis is a condition where endometrial tissue is found outside the uterus. It affects around 10% of people assigned female at birth worldwide, which equates to approximately 1.5 million in the UK.
Symptoms include significant pelvic and abdominal pain during menstruation, painful intercourse and spontaneous pain outside menstruation. In some cases, it can lead to infertility issues with as high as 30-50% of women with infertility also being diagnosed with endometriosis.
The current gold standard for diagnosis is usually made by laparoscopic inspection of the pelvis, which is a very invasive procedure. Less invasive methods may be used, including ultrasound and MRI scanning, however, more research is required into less invasive methods.
Project aim
This study aims to work with 100 patients identified through Kings Fertility clinic who are awaiting laparoscopic surgery.
The study named DEFEND – Developing a US-MRI biomarker fusion model for Endometriosis – aims to develop a database of ultrasound and MRI images for women with symptoms of endometriosis.
Project outcomes
- To develop a database of ultrasound and MRI images for women with symptoms of endometriosis and associated medical history.
- To investigate variability in diagnosing endometriosis using 2D and 3D ultrasound, and whether or not 3D-US is better at diagnosing the condition.
- To assess the use of quantitative MRI sequences in identifying and diagnosing endometriosis
- To assess the use of image fusion, based on quantitative MRI sequences and/or ultrasound images, and analyse their added value, to the diagnostic pathway.
- To develop prototype algorithms for image fusion and quantitative mapping of MRI/US scans for endometriosis

Video case study: DEFEND
March 8, 2023
The DEFEND (Developing an US-MRI-biomarker fusion model for Endometriosis) study: AI could end the need for invasive surgery to diagnose…

International Women’s Day | Meet Dr Laura Ismail
March 8, 2022

International Women’s Day 2022 | Meet Professor Sally Collins
March 8, 2022

International Women’s Day 2022 | Meet Gemma Greenall
March 8, 2022