Login / Signup

Non-invasive tests for endometriosis are here; how reliable are they, and what should we do with the results?

Simon M ScheckClaire E HenryNick BedfordJason A AbbottMichael Wynn-WilliamsAnusch YazdaniSimon McDowell
Published in: The Australian & New Zealand journal of obstetrics & gynaecology (2023)
A reliable non-invasive biomarker for endometriosis is highly likely in the coming years. In the lead-up to this, clinicians need to be aware of commercially available tests as they become accessible, be aware of the level of evidence to support them and be prepared to counsel and manage patients who present with the results of such tests. One such test gaining popularity in Europe was developed using a machine-based learning algorithm to analyse thousands of microRNAs based on a 200-patient cohort with suspected endometriosis in France. We explore the background science for this commercially available test; outline the questions that remain to be answered; and caution against its use outside of a research setting.
Keyphrases
  • deep learning
  • machine learning
  • case report
  • pulmonary embolism