By Clare Murphy
BBC News health reporter
For many women, the scan which shows their unborn squirming, kicking and sucking its thumb is one of the important milestones of pregnancy.
So is it just a waste of money?
For the vast majority it provides reassurance that all is well, and it enables parents to prepare if all is not.
But an eminent ultrasound specialist is determined to kickstart a debate on the value of the scan within an NHS increasingly strapped for cash.
In a paper published in Ultrasounds this week, retired Dr Hylton Meire not only argues there is no scientific evidence to prove the 20-week scan is worthwhile, he also casts doubt on the reliability of the principal method of testing for Down's Syndrome - the nuchal fold measurement.
These tests do not give a yes or no answer to whether a baby has Down's, but an indication of risk. Those deemed to have a higher possibility are offered an amniocentesis, where a needle is inserted into the womb to give a much more accurate analysis.
Every amniocentesis carries a small risk of miscarriage, so women who are not carrying a disabled foetus in the first place can end up losing a perfectly healthy baby.
Using various figures, Dr Meire, formerly of King's College Hospital, calculates that as many as 3,200 healthy babies are lost in this way each year.
For every 50 live births of children with Down's Syndrome prevented, he says 160 women miscarry non-affected babies.
Dr Meire has reservations about aborting babies with Down's Syndrome in the first place, and admits this forms part of his hostility to testing for the condition.
The nuchal fold test involves measuring the fluid at the back of the neck of an 11-13 week old foetus.
Babies with abnormalities tend to accumulate more fluid at the back of their neck during the first three months after conception, causing this clear space to be larger than average.
Age of the mother is one of the factors taken into account when risk is being calculated, and many hospitals also carry out blood tests to give an even more accurate picture.
Low risk is a result where the chances of having a baby with Down's syndrome is less than one in 800, while high risk is anything greater than one in 300.
Dr Meire argues that blood tests notwithstanding, the test is hugely problematic because of the difficulties of measuring anything accurately on a blurry ultrasound screen.
"Women are being referred on to amniocentesis on the basis of a very flimsy test," he says.
"And I think they need to understand just how inaccurate it can be."
Jane Fisher from Antenatal Results and Choices (ARC) says Dr Meire does have a point, but that most parents are all too aware the test is fallible.
Weighing up the various risks - a one in 200 chance of carrying a child with Down's syndrome against a one in 200 chance of miscarrying after an amniocentesis - is an immensely difficult decision.
"The problem is Dr Meire wants absolutes," she says.
"The technology isn't there to provide that yet, and in the absence of it, we have to make do with what there is. But of course counselling is important, and a proper discussion about what the risks are."
In any event, ARC stresses that screening now means fewer older mothers have an amniocentesis as a matter of course, as was the case all the way up until 2001 when the scans became available nationally.
The test is offered in around three quarters of hospitals, but not necessarily to all women. Older women are at greater risk of having a child with Down's Syndrome, and some hospitals only screen those over the age of 35 or even 38.
But the vast majority of women in the UK - 95% - are offered the 20 week anomaly scans which Dr Meire also targets in his paper.
Costing between £25 and £50 a go, the specialist says there is no evidence to prove they are worth the money.
"There's no data that shows - in regard to children living or dying or succumbing to serious illness - that these scans have any effect in population terms. No-one has been able to prove otherwise," he says.
"We need to really think about why we do these, and the money we spend on them, and ask ourselves if it's really worthwhile."
Conditions detected at 20 weeks include serious problems such as heart defects, as well as more minor conditions like cleft palate.
Sometimes you've also got to look at things from an individual perspective, and what it means for the parents - maybe the pure joy of parents seeing their unborn child is something that's worth paying for
Obstetrics consultant Patrick O'Brien says it is very important to know in advance if a child has heart defects, as it means preparations can be made for delivery in a specialist unit if need be.
But he concedes that viewed at the level of pure statistics, it is possible to argue that scans only benefit a small minority of people. Often parents learn information which they do not necessarily need to know before the birth.
But while finding out a child is to be born with a cleft palate may not be necessity, learning through a scan means parents can be prepared for what can be a huge shock at delivery.
"They are shown pictures of what can be done and understand how easy it is to treat," says Mr O'Brien.
"I certainly wouldn't want to go back to the pre-scan days when horrified parents simply shunned the child when it was born."
"But sometimes you've also got to look at things from an individual perspective, and what it means for the parents.
"Maybe the pure joy of parents seeing their unborn child is something that's worth paying for."