Project Details
Description
Foetal anatomy can be identified using video ultrasound techniques to determine a baby's health and growth trajectory in routine perinatal healthcare, but there is growing evidence that individual differences in foetal anatomy might also be predictive of later infant neurodevelopmental disorders such as autism and ADHD. Collaborators at the Cambridge Reproduction strategic research initiative collected 90 samples from 30 participants (25 weeks gestational age) of 4D ultrasound videos in families from a range of backgrounds.
The objective of this project is to develop a software pipeline using state-of-the-art deep learning methods for pose detection, and understand their feasibility when compared to conventional approaches.
The objective of this project is to develop a software pipeline using state-of-the-art deep learning methods for pose detection, and understand their feasibility when compared to conventional approaches.
Layman's description
Goal: AI-annotated Imaging Interface for Fertility Journey
1. What we hope to achieve: Through our "informed imaging" interface, developed with the Lister Hospital in London, we hope to bridge the gap between health care providers, parents to be, cutting edge researchers predicting fertility and child outcomes from earlier data.
2. Who we are: We are a team of researchers funded by University of Cambridge to deploy computer vision and artificial intelligence to identify predictors of successful pregnancies and healthy infant development.
3. How we hope to accomplish it: By annotating own parent's medical images with features known to predict perinatal outcomes, we hope to create a visual roadmap for parents-to-be guiding them along their fertility journey.
| Short title | Womb 2 World |
|---|---|
| Acronym | W2W |
| Status | Active |
| Effective start/end date | 1/12/24 → 1/12/26 |
| Links | https://www.wombs2world.com/ |
Funding
- University of Cambridge: £125,000.00