Artificial Intelligence in IVF

Melbourne IVF, as part of the Virtus Health group, Australia’s leading fertility provider, is developing a new artificial intelligence system, called Ivy, to predict the likelihood of a viable pregnancy from transfer of an individual embryo in a woman undergoing IVF.

The Virtus Health group is the first provider of fertility treatment in Australia to use this innovative Ivy, artificial intelligence technology which allows embryologists to identify the embryo for each woman with the aim of achieving a successful pregnancy as quickly as possible.

 Watch the video below to find out more:

How will Ivy, Artificial Intelligence increase success rates in IVF pregnancies?

The accuracy of predicting which embryo will result in a viable pregnancy can be greatly increased using Ivy. Ivy can predict the potential of an embryo to develop as far as the stage of having a fetal heart, which is a good indicator of the likelihood of a successful pregnancy.

What is different from the current IVF process?

Currently, an embryologist will manually assess each embryo based on the physical appearance on the day of transfer according to standard grading systems. The development of new incubators with time-lapse imaging has opened up new possibilities for assessing embryos but, currently, none of the systems so far in use has significantly improved embryo selection.

Ivy artificial intelligence performs a comprehensive three dimensional assessment of the growth of the embryos through all stages of development and then relates this data to whether a fetal heart has developed or not. By objective analysis of the images, with no pre-determined parameters, Ivy has taught itself to identify the embryos with the greatest likelihood of developing as far as a fetal heart by allocating each embryo an embryoscore. The embryo with the highest embryoscore, and therefore the highest potential for leading to a viable fetus, can then be selected for transfer, with the aim of accelerating the chance of a healthy baby.

How did the Ivy Artificial Intelligence system teach itself?

Ivy is a self-improving artificial intelligence that continuously learns from the embryos that it analyses. The artificial intelligence iteratively retrains on powerful supercomputers and has been shown to consistently improve on previous versions of itself in identifying the embryo with the highest potential for a successful pregnancy.

As an artificial intelligence system, Ivy was able to review a massive amount of data, far more than any human could ever process. Hundreds of images from each embryo were reviewed across a large series of embryos. Patterns were identified in the pictures which were then related back to whether each embryo developed into an ongoing pregnancy. Ivy was then tested by repeating the comparison against the outcomes of a separate group of embryos. The results showed a predictive value for which embryo is going to implant that is a significant improvement on any system that has been used in human embryology before.

What are the potential benefits for patients?

The aim of this work is to select the right embryo more quickly so that women who have to go through IVF are able to conceive more quickly and thus lessen the strain of IVF on them and their families.

The next step of this project will be to carry out clinical trials to quantify the contribution of Ivy artificial intelligence in IVF, to individual women’s success rates and the time that it takes women to achieve a viable pregnancy.

When will Ivy artificial intelligence be available widely in IVF treatment?

Melbourne IVF is planning the randomised control trial of Ivy artificial intelligence. As the pioneers of this groundbreaking technology for Virtus Health, Melbourne IVF's scientists will continue to oversee the project, taking it across all our Virtus sites. We anticipate Ivy will be rolling out across our Melbourne IVF sites following the trial as part of our ongoing investment in the very highest standards of laboratory technology.

Full media release here.

Angus Tran