Page 288 - PowerPoint Presentation

1 International Congress of Artificial Intelligence
st
in Medical Sciences Posters

Repro-AI: status and future prospects

Sousan Houshmandi* , Hossein vaseghi dodaran 2
1
1Department of Midwifery, Ardabil University of Medical science, Ardabil, Iran
2Founder of SPOO Healthtech Startup
Background and aims: Infertility rate in the world varies from 10 to 22%. However, couples
receive successful infertility treatment at low rates, leading to repeat treatment or treatment with-
drawal. Since the birth of the first IVF baby in 1978, more than eight million babies have been
born as a result of the assisted reproductive technique. Artificial intelligence is rapidly changing
the practice of medicine in various fields. Artificial intelligence entered the research world of as-
sisted reproductive technologies (ART) in the late 1990s with the creation of an algorithm aimed
at predicting the outcome of IVF. In reproductive medicine, artificial intelligence can significantly
reduce the highly manual and labor-intensive processes of ART. The aim of this paper is to pro-
vide a systematic review to establish the actual contribution of artificial intelligence for predicting
ART outcomes.
Method: The PubMed database was searched for citations indexed with “artificial intelligence”
and at least one of the medical subject heading terms between January 1, 2000 and April 30, 2020:
“artificial intelligence”. “Obstetrics and Gynecology”; “Assisted Reproductive Techniques, “or
“Fertility”.

Results: The PubMed search retrieved 750 citations and 55 publications met the selection criteria.
All ART subdomains were covered. Among these 55 articles, 15 were related to embryo selection,
25 were sperm evaluation, and 15 were related to egg selection and implantation technologies.
We observed a generally increasing trend in AI-related publications in assisted reproductive tech-
niques over the past two decades.

Conclusion: The development of new artificial intelligence frameworks to predict the ideal out-
come in reproductive medicine is a necessity. As a comprehensive result, this new system can
reduce the instability between observers, reduce risks during egg stimulation, reduce close and
personal clinical contacts, and from the financial aspect, increase clinical profitability and better
determination of sperm tests and evaluation of egg quality and embryo selection.

Keywords: Artificial intelligence, machine learning, assisted reproductive technology, Obstetrics
and Gynecology, fertility

266