[ad_1]
A "super computer" IVF could spare women the sorrow of miscarriage by spotting pregnancies that could fail due to "bad" embryos
- Women could be spared the grief of a miscarriage through the AI
- Breakthrough works with IVF treatment with computer harvesting of "good" embryos
- The computer was able to predict which embryos would lead to a live birth with an accuracy of 85%
Victoria Allen for the Daily Mail
Women could be spared the grief of miscarriage through artificial intelligence that could help detect pregnancies that fail.
The breakthrough operates with IVF treatment, using a supercomputer to select "good" embryos – or "bad" ones that could prevent a woman from giving birth or having her miscarryed.
The researchers found that the computer, nicknamed The Beast, was able to predict which embryos would lead to a live birth with an accuracy of 85%, correctly selecting 280 out of 328.
The breakthrough works with an IVF treatment, using a super-computer to select the "right" embryos – or the "bad ones" that could prevent a woman from giving birth or having her miscarryed
Doctors who email a picture of an embryo to the computer can expect a result in minutes.
And technology is only five years from the provision of couples, say the researchers. About half of the miscarriages are caused by an embryo problem, which has an abnormal number of chromosomes.
The results of the study, involving Cornell University in New York and Imperial College London, will be presented today at the annual meeting of the American Society for Reproductive Medicine in New York. Denver. Dr. Nikica Zaninovic, senior researcher in the United States, supported by Dr. Zev Rosenwaks, said: "If AI can recognize normal embryos and those that have a chromosomal abnormality, it will result in a reduction the rate of miscarriage and stillbirth, which is really our reason for doing this research.
At present, fertility clinics tend to judge whether embryos develop normally by examining their shape, size, or growth rate.
The researchers formed the supercomputer using nearly 700 images of five-day old embryos. They used accelerated photography, where a photograph of embryos in an incubator is taken every ten minutes, thus avoiding the risks associated with handling an embryo.
UK Professor Allan Pacey, a professor at Sheffield University, said: "Time-lapse system data is a very good idea because it can find models and algorithms invisible to the human eye."
The new method could also prevent doctors from implanting two embryos to increase the chances of pregnancy. Dr. Zaninovic said, "We could help women avoid twins and multiple births, which cause all kinds of complications."
The scientist, who seeks to patent the technology, added: "Everything is web-based. That does not mean that all hospitals will have that. This can be done very easily on the Internet. "
Source link
