Submitted by Pooja Pandey on Thu, 14/11/2024 - 18:03
Meet Cavendish physicist Mo Vali who is using innovative AI methods to revolutionise fertility treatments along with a fertility doctor and an infant neuroscience researcher. Mo Vali is a PhD student in Diana Fusco’s (Biophysics) research group and his project is one of the five ai@cam initiatives that the University has chosen to fund to help tackle important problems using AI.
The project, ‘Revolutionising women’s health, female fertility and early infant neurodevelopment using AI’, involves using machine learning techniques to improve fertility and IVF outcomes. It aims to develop affordable, non-invasive, AI-assisted tests to support the conception to childhood journey, improving diagnosis accuracy for women’s health and personalising fertility and IVF outcomes.
The researchers are responding to a birth rate crisis.In many advanced economies, the global fertility rate has more than halved over the past 50 years. This is exacerbated by poor birth outcomes for couples who are finding it difficult to have children. Those people often turn to in vitro fertilisation (IVF). However, the current technology is inadequate as the IVF outcomes are poor and expensive. Scientifically, this area is not well understood.
“Most IVF innovations have been commercially led, tinkering around the edges and trying to make technical improvements to the IVF process,” said Mo Vali. “However, there’s a lot of unexplored territory, with open questions about the underlying molecular mechanisms of life. We still don’t really know the underlying mechanisms for how life takes shape, and we hope to put the science on a more rigorous footing – improving outcomes, reducing misinformation and exploring some of these fundamental questions.”
The scientists are envisioning a comprehensive personalised monitoring and treatment pathway from preconception through infancy, using AI to help shed light on the still largely unknown underlying cellular, molecular and genetic mechanisms for how a life takes shape, including the genesis of neurodevelopmental disorders like autism. Currently they are gathering extensive data from fertility clinics, including ultrasound images, blood samples, and follicular fluid. They are also developing tools to provide parents with real-time insights into their baby’s development, with the ultimate goal of creating an online dashboard that offers a precise timeline of pregnancy and foetal growth. By personalising fertility treatments to individual profiles, the researchers hope to enhance IVF outcomes for around 55,000 UK women. Advanced imaging and computer vision will allow for early disease diagnosis, reducing costs and enhancing care within the NHS.
Recently the scientists also presented their initial research at a global conference for clinicians, the RCOG World Congress in Oman. At this conference, mainly for research in obstetrics and gynaecology, the group won first place in the e-poster research prize, out of a possible 1,109 posters. “This initial research involved using multi-modal deep learning architectures to predict embryo transfer outcomes, with an emphasis on interpretability to help clinicians understand how the model is making its predictions,” said Mo Vali, who presented the poster at the conference with his collaborator (and sister) Saaliha Vali.
“We are utilising ultrasound images, patient medical histories, and blood test results to make a diagnosis. We're hoping to translate this work to develop a diagnostics platform for IVF clinics to help better diagnose fertility conditions as early as possible during the course of treatment.”
“We are also addressing any ethical risks and data privacy concern of managing highly confidential data by having a dedicated Graphics Processing Unit on a separate server. We have also ensured all data is anonymised at source upon collection, with highly restricted and monitored access. There is no way to identify patients from our data and this helps to head off any data privacy concerns,” concludes Vali.
The project is a collaboration between the University of Cambridge’s Computer Science and Physics department, among others – in partnership with two leading specialty hospitals – Addenbrooke’s Hospital and The Lister Hospital.
This initiative is actively seeking collaborations with IVF clinics and hospitals looking to address foundational questions in women’s, reproductive and infant health using state-of-the-art AI methods.
If you would like to learn more about this research project or are interested in a potential collaboration—particularly in spectroscopy on biofluids or multi-modal machine learning—please email Mo at mv487@cam.ac.uk, Staci at smw95@cam.ac.uk and Sal at s.vali@nhs.net.
This article has been adapted from the original article From womb to world: Using innovative AI methods to revolutionise fertility treatments.