Artificial intelligence would improve IVF’s success rate, research finds

Using artificial intelligence to determine embryos with the best chance of producing successful pregnancies through in-vitro fertilisation (IVF) would increase the procedure’s rate of success, according to a study.

The research, presented today at the 33rd Annual Meeting of the European Society of Reproduction and Embryology (ESHRE) in Geneva, found that using AI to standardise the selection of ‘good quality’ embryos would make viable embryo selection more accurate and so increase the chances of a successful pregnancy.

A popular choice for couples with fertility problems, IVF involves removing eggs from the woman’s ovaries, fertilising them with sperm to produce embryos and then implanting the most viable back into the womb, where they – all being well – will develop as normal into a healthy baby.

As part of this, embryologists determine which embryos are most viable for implantation, however despite this selection process, between 30 and 60% fail to implant successfully. Part of the reason for this is patient age, but there is also the fact that different embryologists will make different calls about whether an embryo is viable.

“The issue is that morphological grading by humans leads to wide inter and intra-operator variation,” said investigator Professor José Celso Rocha, from São Paulo State University, Brazil. In other words, embryologists’ assessment of an embryo’s shape and development results in significantly different conclusions depending on who is doing the assessing. And it is this variation that Rocha believes AI can help with.

“To classify images automatically will increase the predictive value of our embryo assessment,” he said. “By increasing objectivity and repeatability in embryo assessment, we can improve the accuracy of diagnosing embryo viability. Clinics can use this information as ‘artificial intelligence’ to customise treatment strategies and better predict a patient’s chance of pregnancy.”

The study which is the focus of this argument involved bovine embryos, with 482 seven-day-old embryos used to ‘train’ the AI system to recognise viable embryos from non-viable ones. The system assessed the embryos against 36 variables to determine viability, resulting in an accuracy of 76% – an improvement on conventional methods – and increased general consistency.

Now the research has moved onto human embryos, and although it is in the early stages of development, it is hoped that it will produce a highly repeatable system that will make future embryo classification far more consistent.

However, while AI might be poised to take over a part of the IVF process, it is human expertise that it will draw from.

“The artificial intelligence system must be based on learning from a human being,” said Rocha. “That is, the experienced embryologists who set the standards of assessment to train the system.”

AI will ultimately need globally agreed laws: DeepMind CEO

Once artificial intelligence has sufficiently advanced, it will need to be bound by rules and regulations that are agreed upon by governments across the world, Demis Hassabis, co-founder and CEO of Google-owned AI giant DeepMind, has said.

“What we hope is once we understand these systems better and understand what we would be legislating for, there would be some stronger form of governance that would be agreed by world governments,” he said.

“In the course of the next few decades as these systems develop, we will understand better what kind of control systems are required, how to check and interpret what these systems are doing and I think we will solve those problems on the way to building AI.”

Speaking on the latest episode of BBC Radio 4’s Desert Island Discs, Hassabis argued that AI was ultimately a positive force, in contrast to doom-ridden comments made by figures such as Stephen Hawking.

“The way I think about AI is as this amazing tool that we can use to enhance our own goals as humans,” he explained.

“I think there’s a lot of research left to go, but we have to think about what goals we give these systems, what values we give these systems and how we make sure that we stick to the goals that we give them.”

DeepMind CEO and co-founder Demis Hassabis. Image courtesy of DeepMind

Hassabis also rejected the notion that we should be concerned about current AI technology falling into the wrong hands, arguing that it was too complex and too small a field for this to realistically happen at present.

“One good thing about it at the moment is that these are incredibly tough systems to build, they’re incredibly complicated – there are only a few hundred people in the world who can build these things,” he said.

“And, you know, most of them know each other was well. We’re quite a small community – I couldn’t do it now on my own; we have 400 people, 250 PhDs from the top places in the world, coming together and even then we’re only making small progress towards this.”

After being bought by Google in 2014, DeepMind made headlines when its Go-playing AI software AlphaGo beat world-class professional Go player Lee Sedol.

The match, which was held in March 2016, was hailed as a watershed moment for AI as it required the system to develop its own methods to play the game, rather than being equipped with a set of routines it could run through.

However, Hassabis was keen to stress that although the resulting methods drew surprise, this was not evidence of a machine breaking its programming to think for itself.

“We were astonished by the capability the machine had, and our program that we created actually came up with its own ideas and its own motifs that even stunned the Go world,” he said.

“But we weren’t surprised that it could play Go well; we were surprised that it could create its own moves and we were delighted by what it created within the domain of what we made it to do, which was to play Go.“