The success rate of IVF could be increased to 85% with a new revolutionary test using artificial intelligence

Thousands of women undergoing IVF could be spared from the agony caused by miscarriages and stillbirths by artificial intelligence.

A machine has been developed to select embryos that can lead to a live birth with a staggering accuracy of 85%.

The researchers say that robot learning will "revolutionize" IVF and will be available to NHS hospitals by five years ago.

The desperate couples of children could see a marked improvement in the birth rate, because the British doctors emailed a picture of their embryo and got an answer in a few minutes.

The technology was developed jointly by Imperial College London and Cornell University in New York.

He uses accelerated pictures of embryos in an incubator, which currently helps doctors to choose the most likely to give birth to a baby.

About half of the losses in early pregnancy are due to an abnormal number of chromosomes in the embryo.

This breakthrough was rewarded this week with an award at Denver's largest fertility conference in the world.

Dr. Nikica Zaninovic, head of study at Cornell University, said: "If the AI ​​recognizes normal embryos and those with a chromosomal abnormality, the rate of miscarriage and stillbirth will be reduced, which is really our main reason search.

"This research is something really new and an 85% success rate is huge."

For women under 35 with no other underlying health problem, this could increase the live birth rate by IVF up to 70%, said Dr. Zaninovic.

Currently on the NHS, it is about 30% for under 35s.

The use of time lapse technique to select embryos, without the help of AI, has been around since 2010.

Embryos stay in an incubator and a photo is taken every 10 minutes to measure growth.

The computers then produce data to help doctors choose an embryo to implant in the mother.

This prevents doctors from manipulating and choosing according to their shape and appearance

Scientists have powered a New York-based supercomputer – dubbed The Beast – thousands of historical images to "teach" him what to look for in successful embryos.

It has been said to those who have resulted in a live birth, where pregnancies have failed or have resulted in a stillbirth.

Dr. Zaninovic, whose team is currently patenting technology, said, "I only need information about the patients of a London hospital and the image of the patient. 39, embryo, that I can put in the computer.

"Everything is web-based. That does not mean that all hospitals will have that. This can simply be done on the Internet very easily. "

He added: "In five years, it will be commonly used in the clinic."

The Beast has received more than 50,000 images of more than 10,000 embryos.

In a retrospective single-image trial of 328 embryos implanted in potential mothers.

When asked to indicate which of these would have resulted in a biological birth, he selected 280 correctly.

The 85% rate is well above what one would expect from a human embryologist, according to the researchers, with current standards varying from one clinic to the next.

Researchers say it is "close to 100%" because about 15% of the reasons why live births do not occur are due to problems in the uterus.

The findings were unveiled at the annual meeting of the American Society for Reproductive Medicine (ASRM).

Some clinics across Britain use time-lapse photographic technology to help specialists make a crucial decision about the choice of embryo.

Evidence as to whether, in its current form, it improves successful birth rates are mixed.

Some believe that these measures help improve the success rate, in part because the embryos do not need to be handled by scientists for five days.

Professor Allan Pacey, British expert in fertility, from the University of Sheffield, said: "The ability to observe the development of embryos with the help of time lapse systems. has developed considerably in recent years and these devices have become commonplace in clinical practice.

"However, to date, the data do not show that they actually help to select the best embryo and improve the chances of pregnancy.

"Therefore, the application of artificial intelligence to data collected from time lapse systems is a very good idea, as it can detect patterns and algorithms that are not visible at all. human eye. "

Professor Charles Kingsland founded the UK's largest NHS IVF unit, Liverpool's Hewitt IVF, which was the first to successfully use time-lapse technology.

Prof. Kingsland, clinical director of CARE Fertility, said, "Although the numbers in the study are small, these results are encouraging.

"By adding artificial intelligence to the computer, it pushes it even further into the future.

"The final arbiter of embryo quality will always be a highly qualified embryologist or clinician.

"There are only many years left in the in vitro fertilization laboratory (IVF)."