Artificial Intelligence-Based Machine Learning to Diagnose and Classify Adenomyosis from Ultrasound Scans: a Multicentre Model Development Study
This study is exploring how artificial intelligence (AI) can help diagnose and classify a common womb condition called adenomyosis using ultrasound scans. Adenomyosis, where the womb lining grows into its muscle, can cause severe pain and heavy bleeding, yet it's often hard to diagnose clearly. Researchers are using a large collection of ultrasound images from women with and without adenomyosis to train AI systems. They want to see if AI can accurately identify adenomyosis and its severity, potentially saving healthcare professionals time and reducing differences in diagnosis between different doctors. Ultimately, this could lead to quicker, more consistent diagnoses, helping women get suitable support sooner.
At a glance
What is this study about?
This study is looking at an exciting new way to diagnose a common condition called adenomyosis, which affects many women. Adenomyosis happens when the cells that normally line the inside of your womb (uterus) grow into the muscular wall of the womb. This can lead to very heavy periods, severe period pain, ongoing pelvic pain, and can even affect pregnancy. Despite how common it is, adenomyosis is often missed or takes a long time to diagnose properly.
Currently, diagnosing adenomyosis and understanding how severe it is can be tricky and sometimes depends a lot on how experienced the doctor performing the ultrasound is. This study aims to make this process much easier and more consistent by using artificial intelligence (AI). Researchers are feeding a huge number of ultrasound images of wombs – some normal, some with adenomyosis of different severities – into special computer systems. These systems 'learn' to recognise the patterns associated with adenomyosis.
The main goal is to see if these AI systems can accurately identify adenomyosis and classify its severity automatically. If successful, this could mean that diagnosing adenomyosis with an ultrasound could become much quicker and less dependent on guesswork, helping doctors give women a clearer picture of their condition sooner. This could then lead to better tailored treatment plans and an improved quality of life for women living with adenomyosis.
Key takeaways
- AI is being trained to diagnose adenomyosis from ultrasound scans.
- The goal is to make diagnosis quicker and more accurate.
- It could help doctors classify the severity of adenomyosis better.
- This could lead to more consistent diagnoses across different healthcare professionals.
- No new patient involvement is required; existing scans are being used.
- Ultimately, this potential improvement in diagnosis aims to benefit women with adenomyosis.
Who may be eligible?
This study will look at ultrasound scans from women who visited the CARE Fertility centre between February 2022 and February 2024 for any reason. Researchers will be reviewing these existing ultrasound images.
The scans they use must be clear, good quality 2D and/or 3D images that clearly show either a normal womb or a womb with adenomyosis where the condition's features are visible. They are looking for images that allow a clear and certain diagnosis of adenomyosis or a normal womb.
They will not use scans from women who also have other conditions in their womb, such as fibroids (non-cancerous growths) or abnormalities inside the womb cavity. They will also exclude any scans that are not clear, are poor quality, or where it's impossible to tell if adenomyosis is present or absent.
Could this study suit you?
Answer these quick questions to see if you may be eligible. This is a guide only — the research team makes the final call.
- Did you have an ultrasound scan at CARE Fertility between February 2022 and February 2024?
- Was your scan clear and of good quality?
- Did your scan show either a normal womb or signs of adenomyosis?
- Were there no other conditions like fibroids or womb cavity problems seen in your scan?
What does participation involve?
As this study involves reviewing existing ultrasound scans, you would not need to do anything extra. This is not a study where new scans or treatments are given. Researchers are simply analysing images already taken as part of your routine care at the CARE Fertility centre between February 2022 and February 2024. Your involvement is therefore indirect and passive, meaning you wouldn't need to attend any appointments, take any medication, or have any follow-up related to this research.
Potential risks and benefits
Locations (1)
- CARE FertilityVerified postcodeBirmingham, United Kingdom
Common questions
What is adenomyosis?
Adenomyosis is a condition where the lining of the womb grows into its muscular wall, causing symptoms like heavy periods and pain.
How does AI help diagnose adenomyosis?
AI 'learns' from many ultrasound images to recognise patterns of adenomyosis, potentially making diagnosis quicker and more consistent.
Will I have to do anything if I participate?
No, this study uses existing, past ultrasound images, so you wouldn't need to do anything or attend any appointments.
Who is eligible for this study?
The study uses existing scans from women who had an ultrasound at CARE Fertility between February 2022 and February 2024, showing either a normal womb or adenomyosis.
What are the benefits of this research?
It could lead to faster, more accurate diagnoses of adenomyosis in the future, helping women get better care sooner.
How to find out more
Always speak to your GP or specialist before deciding to take part in a study.
Discussion
Community discussion
Powered by our forum at community.patient.info. Please be respectful — this is not medical advice.