How AI can now tell sex of baby three DAYS into pregnancy by spotting tiny differences in embryos 

ARTIFICIAL intelligence could reveal the gender of a baby as early as three days after fertilisation, scientists reckon.

The AI tech was more accurate than humans in determining the sex of IVF embryos when viewed under a microscope.

Illustration of a newborn baby's hand reaching out to a robot's hand in a hospital.
AI could reveal the gender of a baby as early as three days after fertilisation, scientists reckonCredit: Shutterstock

The Cambridge University team trained an algorithm to compare 515 videos of embryos prepared for in-vitro fertilisation under a microscope in very early stages of development.

They were no more than a blob of cells days after the egg was fertilised.
Scientists then pitted three human experts against the AI model.

People were not able to distinguish male from female and were right only 46 per cent of the time — no better than if they were guessing.

But the algorithm could sometimes detect tiny differences invisible to the human eye, achieving 61 per cent accuracy on average — better than random.

MUM DEATH

Brit died of rabies after dog scratch ‘when medics struggled to diagnose illness’


‘SCANDALOUS’

Record number of Brits PAYING privately for chemotherapy due to long NHS waits

Mums-to-be are currently not told the sex of their baby until about halfway through pregnancy, at 20 weeks, when identification is easier on scans.

Blood tests and biopsies can determine an embryo’s sex earlier but are invasive and higher risk.

The study suggests ultrasounds could one day be analysed to detect changes sooner.

Medics already use AI to read breast and lung scans to look for signs of cancer as it is much faster and more sensitive than people.

Study author Prof Magdalena Zernicka-Goetz said: “Subtle sex- related differences may begin to emerge around day three.

“Studying sex differences at this early stage may enhance our understanding of why some embryos fail to develop.

“Our findings open the possibility for an early, non-invasive detection tool that could assist in identifying and addressing developmental abnormalities.”

Source link

Related Posts

Load More Posts Loading...No More Posts.