Cookies on this website

We use cookies to ensure that we give you the best experience on our website. If you click 'Accept all cookies' we'll assume that you are happy to receive all cookies and you won't see this message again. If you click 'Reject all non-essential cookies' only necessary cookies providing core functionality such as security, network management, and accessibility will be enabled. Click 'Find out more' for information on how to change your cookie settings.

The placenta is the most indispensable, disposable organ you ever had. It decrees both how well we grow and develop in the womb and our future health well into adulthood. Our research explores two key themes: 1) Imaging the placenta early in pregnancy in order to predict adverse pregnancy outcomes including fetal growth restriction and 2) Diagnosis and management of Placenta Accreta Spectrum Disorder (PAS) - also known as abnormally invasive placenta (AIP).

To facilitate real-time identification of the placenta we have used machine learning to teach the computer to 'see' the placenta in an ultrasound image

First trimester Placental imaging (OxPLUS study)

Every baby has an inherent growth potential which, under ideal conditions, will produce a healthy, appropriately-sized infant. Attaining this potential relies on good maternal health and a well-functioning placenta. If conditions are sub-optimal, babies will not achieve their full growth potential. This is called Fetal Growth Restriction (FGR). Failure of the placenta to develop properly is the most common cause of FGR.

The consequences of FGR can be devastating. It is the single biggest risk for stillbirth, is associated with pre-term birth and cerebral palsy, and has long-term health implications in adulthood including increased risk of obesity, diabetes and high blood pressure.

FGR cannot be prevented at present but early identification with timely delivery can prevent the baby from dying before it is born. Stillbirth currently devastates 8 families a day in the UK and costs the NHS an average of £4191 for each lost baby. Identifying and monitoring those pregnancies at high risk of FGR and delivering small babies before they succumb should significantly reduce this dreadful statistic.

FGR can be diagnosed with serial ultrasound scans. This is expensive and can medicalise a normal pregnancy causing unnecessary anxiety. Hence, only women deemed at high risk of FGR have these scans. In the UK low-risk women only have more than two scans if their doctor or midwife thinks the baby is small when they measure their belly with a tape-measure. Unfortunately, this approach is unreliable and as many as three-quarters of FGR babies go undetected before birth.

FGR babies have small placentas. Previous research has shown that small placentas can be detected as early as the dating scan at 11-13+6 weeks’ gestation. Until now though, it has been impractical to calculate placental volume and vascularity from a 3-D scan, as the tools were either unavailable or too time-consuming and operator dependent. We have developed a method that has fully computerised this process, giving reliable measurements of both the size of the placenta and its overall perfusion. This should make it possible to develop a population-based screening test for FGR. Our group are working on this.

A reliable test to screen for FGR would mean that the women truly at risk of FGR will have serial ultrasound scans enabling timely delivery of small babies who are getting into trouble and potentially reduce the number of stillbirths without unnecessarily worrying women with normal pregnancies. Also, if we know which pregnancies are at high-risk we can offer them the chance to try novel treatments to improve the function of the placenta and potentially prevent FGR from occurring. As the consequences of FGR include a long-term increased risk of obesity, diabetes & cardiovascular disease, the health benefits of any such successful treatment to the baby itself, and society as a whole, would be considerable.


placenta accreta spectrum (OxPAT Web page)

During a typical pregnancy, the placenta attaches to the top or side of the uterus and after a baby is delivered, the placenta follows. Delivery of the entire placenta is vital to a woman’s health after giving birth. When placenta accreta spectrum (PAS) occurs, it means there is a lack of spontaneous separation after a baby is born and the placenta is retained. Forced removal of the placenta can lead to a catastrophic maternal haemorrhage, putting mother and baby's lives at risk. PAS is the most common indication for caesarean hysterectomy. Average estimated blood loss has been quoted to be 3-5 litres with 90% of women requiring blood transfusion. Some of the major issues with PAS is that it is a spectrum disorder and there was no ‘gold-standard’ clinical diagnosis, and there are many different types of the condition; for example: focal accreta, accreta, increta or percreta. 

Following delivery for PAS, some women may experience Post Traumatic Stress Disorder. For example they feel they have experienced “actual or threatened injury to the baby” or had to endure an unplanned caesarean section and instrumental delivery.This can lead to negative consequences such as a relationship dissatisfaction and/or marital break-up; failure to bond with the child (‘stressor’); unable to cope with activities of daily living; deliberate self-harm or poor ‘health behaviours’ and suicide.  Maternal mortality has been reported to be as high as 7% for placenta percreta (900 times the background risk). 
Our research is focused on raising awareness of PAS amongst clinicians and patients, so that maternal mortality and morbidity is significantly decreased by antenatal diagnosis and appropriate management. We work with Healthcare professionals need to be educated about PAS and its potential negative psychological impact to facilitate help and support for these women. 

International Society for PAS (IS-PAS)

Prof Sally Collins represents the UK and is the Chair of the International Society for PAS (IS-PAS)

It consists of 63 clinicians and researchers across 14 countries and aims to:

  • Advance techniques for diagnosis and treatment
  • Promote research
  • Raise awareness of PAS and evidence-based management strategies.


Our team

Introducing OxNNet - how it uses AI to identify poor placenta health

Addressing the risks of stillbirth

VIDEO: Fully Automated 3D segmentation of the placenta.

Fully Automated 3D segmentation of the placenta: The ‘Authors Take’ video accompanying our recent JCI Insight article.

Some recent publications

Related research themes