He, Chloe; (2025) Embryos Across Time, Space and Society: On the Development and Implications of Biologically-Informed Artificial Intelligence in Clinical Embryology and Beyond. Doctoral thesis (Ph.D), UCL (University College London).

Text He_Thesis.pdf - Accepted Version Access restricted to UCL open access staff until 1 August 2026. Download (71MB)

Abstract

Human reproduction in the modern world has been marked by two trends: a steep decline in fertility rates and an increase in the use of assisted reproductive technolo gies (ART) such as in-vitro fertilisation (IVF). With infertility estimated to affect 17.5% of the worldwide population owing to social, economic, ecological and ge ographical factors, millions of individuals and couples have turned to ART to have children. Despite its increasing popularity, IVF faces several challenges. In its cur rent form, IVF carries with it a great economic and emotional cost to patients, typ ically involving multiple failures before a successful pregnancy is obtained. More over, at an international level, nearly every nation boasting a sizeable IVF sector grapples with a shortage of skilled embryology staff. In recent years, a growing body of work has explored the use of artificial intel ligence in IVF. One task that has received particular attention is that of automating embryo selection, a specialist laboratory process in which embryos deemed most likely to lead to a successful pregnancy are prioritised for transfer back into a pa tient. Automatic embryo selection algorithms have the potential to reduce the num ber of attempts needed to achieve a pregnancy, as well as to increase embryologist bandwidth and standardisation. In this thesis, we explore several topics within automated embryo selection. The thesis opens with an investigation into challenges associated with standardising embryo imaging across clinics, proposing a new technique for standardising embryo data collected across different clinics. The thesis then introduces a novel pipeline for the 3-dimensional reconstruction of human embryos in a clinical setting. The clinical utility of the pipeline is validated through two studies which illustrate its potential for enabling single-cell and spatial analysis to be applied in clinical em bryology. Finally, we close the thesis with two case studies on the intersection of embryo selection, IVF and society at large.

Download activity - last month

Export as XMLCSVJSON

Download activity - last 12 months

Export as JSONXMLCSV

Downloads by country - last 12 months

Export as XMLCSVJSON

Archive Staff Only

View Item