Completed Pregnancy, Children & Inherited Conditions Bones, Joints & Muscles

A computer-guided imaging system for prenatal screening and comprehensive diagnosis of fetal abnormalities.

In plain English

AI plain-English summary

Half of all congenital abnormalities are missed by standard ultrasound scans. This project builds a computer-guided system using multiple ultrasound probes that automatically capture 3D images of the fetus, then compares them against population-based atlases to flag abnormalities. Current screening depends heavily on skilled sonographers, and detection rates vary widely by region. The system aims to remove that human bottleneck, completing a scan in minutes and delivering consistent, higher detection rates regardless of where a patient lives. It also generates high-quality 3D images of the placenta, amniotic fluid, and fetus, which could be linked to genetic and environmental data in large population studies. If successful, the technology would transform prenatal screening from a subjective, operator-dependent process into an automated, standardised one. The immediate impact is on medical diagnostics—specifically, the national screening programme for fetal abnormalities. The richer imaging data could also feed into research on how genetic and environmental factors shape fetal development, though that is a downstream benefit, not the project’s primary aim.

View original technical description
Ultrasound is a powerful tool and the mainstay of fetal imaging. However, the detection rate for many congenital abnormalities from the national screening programme is around 50%. In this application we propose the development of new computer guided ultrasound technologies, which will allow screening of fetal abnormalities in an automated and uniform fashion. The novelty in our approach is to use multiple ultrasound probes that can simultaneously acquire large 3-D datasets and be guided automatically to acquire the optimal data. This then allows the application of image processing techniques to improve the ultrasound image quality and the use of population-based atlases to identify different organs and systems of the fetus and help differentiate between normality and abnormality. This will be a radical change to fetal screening, largely removing the need for expert ultrasonographers to acquire and interpret the images. It will allow the initial screening scans to be done in a few minutes, and provide a consistently higher detection rate for major abnormalities, which is not subject to regional variation. In addition, the high quality 3D imaging phenotype of the placenta, amniotic fluid and fetus, could be used across large population studies linked to genetic and environmental factors.

View the original record at the funder ↗

Researchers

Reza Razavi (EPMC Awardee)

Related Research

Grants with similar aims, by meaning.

Perception Ultrasound for Reassuring Fetal Echo Clinical Trainees
Computer Aided Diagnosis for the Analysis of the Fetal Brain and Heart on Ultrasound
Improving reliability of three-dimensional fetal echocardiography through motion corrected slice-to-volume registration and machine learning
The Pregnancy Ultrasound ResourcE study (PURe study)
Ultrasound-Based Assessment of Brain Folding Patterns in Early Pregnancy

Original classification

EPSRC/WT Innovative Engineering

Plain English summaries and category classifications on this site are generated by AI and may not perfectly reflect the original research.