Want to freeze your eggs for the future? AI can evaluate the likelihood that they will produce good quality embryos
The peer-reviewed journal Reproductive BioMedicine Online (RBMO) has published results of a global clinical study showing that Artificial Intelligence (AI) can predict the likelihood of an individual human egg (oocyte) developing into an embryo that’s suitable for use in IVF. Importantly, the AI can also predict how many oocytes are likely to develop into embryos in any given IVF cycle, which can help with those critical decisions around how many IVF cycles should be performed for a successful outcome. This is great news for women freezing their eggs, helping to provide peace of mind that their frozen eggs will form suitable embryos when needed in the future. In some cases, the AI may help reduce unnecessary IVF cycles, alleviating some of the financial and physical burden associated with the IVF process.
The study was conducted in collaboration with an extensive group of IVF clinics who collected upwards of ten thousand oocyte images from almost 1500 IVF cycles. The AI was trained using a unique Federated AI method, where patient data were not transferred out of the region of origin, and no-one except the treating healthcare professional saw the data. In this study, the AI “saw” the data by travelling to servers in each region, keeping patient data safe and secure. Contributors included California Fertility Partners in the USA, Asada Institute for Reproductive Medicine in Japan, Alpha IVF & Women’s Specialists and Kensington Green Specialist Centre in Malaysia, Indore Infertility Clinic in India, Trinidad and Tobago IVF in Trinidad and Tobago, and Dokuz Eylül University in Turkey – making this a truly international effort to create an AI that can accurately assess egg quality prior to fertilization.
The AI evaluates the quality of eggs at a specific level of maturity (metaphase II) before they undergo fertilization using a process called Intracytoplasmic Sperm Injection (ICSI). The current understanding in the field of what features indicate a good quality egg has been largely inconclusive until now. Any development in this area that brings clarity to evaluating egg quality will go a long way to assisting clinical decision-making with regards to planning IVF cycles, with the ultimate aim of achieving a successful live birth via IVF.
The AI algorithm, called Life Whisperer Oocytes (a product by Presagen), evaluates how likely oocytes are to form a good quality embryo, or blastocyst. The AI is able to look at individual oocytes, as well as a whole group of oocytes in a single image, providing individual oocyte scores as well as an overall estimation of how many blastocysts are likely to develop in a cycle.
Presagen’s Chief Medical Science Officer Dr Sonya Diakiw explained “For patients undergoing IVF, this knowledge could help their doctors decide whether to perform a fresh embryo transfer with corresponding progesterone treatment, or whether a freeze-all cycle approach might be more appropriate. The AI could be a valuable decision-making tool for evaluating whether to conduct additional rounds of hormone stimulation and oocyte retrieval, and it could also be used to evaluate the quality of donor oocytes for Egg Banks.”
Presagen’s Chief Scientist Dr Jonathan Hall said “We are excited to be able to assist patients in their infertility journey, by providing a technology that gives such useful information for decision-making, helping to mitigate stress and improve transparency around the IVF and egg assessment process.”
Life Whisperer Oocytes is due to be released in the next software update for Life Whisperer. It can be used in combination with Life Whisperer Genetics and Viability, which assess if a resulting embryo is likely to be genetically normal, and likely to result in a pregnancy. Together, the Life Whisperer AI capabilities span across multiple clinical touchpoints, from pre-ICSI, through Day 1 fertilization outcomes, all the way to Day 5 and 6 prior to transfer back to the patient, giving unique, point-in-time assessment through the IVF journey.
Paper Title
Use of Federated Learning on distributed data to develop an artificial intelligence for predicting usable blastocyst formation from pre-ICSI oocyte images
https://www.sciencedirect.com/science/article/pii/S1472648324005923
Authors
J. M. M. Hall a,b,c , T. V. Nguyen a, A. W. Dinsmore d, D. Perugini a, M. Perugini a,e, N. Fukunaga f,
Y. Asada g, M. Schiewe d, A. Y. X. Lim h, C. Lee h, N. Patel i, H. Bhadarka i, J. Chiang j, D. P. Bose k,
S. Mankee-Sookram l, C. Minto-Bain l, E. Bilen m, S. M. Diakiw a
a Life Whisperer Diagnostics (a subsidiary of Presagen), San Francisco CA 94404, USA, and Adelaide
SA 5000, Australia
b Australian Research Council Centre of Excellence for Nanoscale BioPhotonics, Adelaide SA 5000,
Australia
c Adelaide Business School, The University of Adelaide, Adelaide SA 5005, Australia
d California Fertility Partners, Los Angeles CA 90025, USA
e Adelaide Medical School, The University of Adelaide, Adelaide SA 5000, Australia
f Asada Institute for Reproductive Medicine, Nagoya, Japan
g Asada Ladies Clinic, Nagoya, Japan
h Alpha IVF & Women's Specialists, Petaling Jaya, Selangor 47810, Malaysia
i Akanksha Hospital and Research Institute, Anand, Gujarat 387310, India
j Kensington Green Specialist Centre, Iskandar Puteri, Johor 79100, Malaysia
k Indore Infertility Clinic, Indore, Madhya Pradesh 452008, India
l Trinidad and Tobago IVF and Fertility Centre, Maraval, Trinidad, Trinidad and Tobago
m Dokuz Eylül University, Inciraltı 35330 Balçova/İZMİR, Turkey
Paper Abstract
Research question
Can Federated Learning be used to develop an artificial intelligence (AI) for evaluating oocyte competence using two-dimensional images of denuded oocytes in metaphase II prior to intracytoplasmic sperm injection (ICSI)?
Design
10,677 oocyte images with associated metadata were prospectively collected by 8 in vitro fertilization (IVF) clinics across 6 countries. AI training used Federated Learning, where data were retained on regional servers to comply with data privacy laws. The final AI required a single image as input to evaluate oocyte competence, which was defined by the formation of a usable blastocyst (≥expansion grade 3 by Day 5/6 post-ICSI).
Results
The oocytes AI demonstrated area under the curve (AUC) up to 0.65 on two blind test datasets. A high sensitivity for predicting competent oocytes (83-88%) was offset by a lower specificity (26-36%). Exclusion of confounding biological variables (male factor infertility and maternal age ≥35 years) improved AUC up to 14%, primarily due to increased specificity. AI scores correlated with size of the zona pellucida and perivitelline space, and ooplasm appearance. AI scores also correlated with blastocyst expansion grade and morphological quality. The sum of AI scores from oocytes in group culture images predicted formation of ≥2 usable blastocysts (AUC=0.77).
Conclusion
An AI for evaluating oocyte competence was developed using Federated Learning, representing an essential step in protecting patient data. The AI was significantly predictive of oocyte competence as defined by usable blastocyst formation, which is a critical factor for IVF success. Potential clinical utility ranges from selective oocyte fertilization, to guiding treatment decisions regarding additional rounds of oocyte retrieval.
About Presagen and Life Whisperer
Presagen is an AI healthcare company that is changing the way clinics, patients, and medical data from around the world are connected through AI. Its platform, The Social Network for Healthcare, connects clinics and patients globally, and enables collaboration and data sharing to create scalable AI healthcare products that are affordable and accessible for all. The decentralized network democratizes the creation of AI products, promotes collaboration through incentives, and protects data privacy and ownership. With a focus on improving Women’s Health outcomes globally, Presagen’s first product, Life Whisperer, is being used by IVF clinics globally to improve pregnancy outcomes for couples struggling with fertility. With a vision of creating the largest network of clinics, patients, and medical data from around the world, Presagen is driving the future of AI Enhanced Healthcare.