To identify Barrett's and look for possible dysplasia, samples of cells (biopsies) need to be taken and examined by microscope.

The "Seattle Protocol", introduced about 20 years ago, calls for biopsies to be taken quadrantically over each centimetre length of the suspected area.

And it's sometimes not that easy to tell exactly where the suspected Barrett's begins and ends - and it's unlikely to be a tidy ring as in this picture - far more likely to have tongues or islets and the surface of the oesophagus isn't nice and smooth.

The Seattle protocol has come in for discussion with research papers appearing to contradict each other:

A good endoscopist may be able to spot the suspect areas but it can be like looking for a needle in a haystack. Those who were at the 2011 annual meeting of Barrett's Wessex may remember the slide Dr Patel showed with a small blue cornflower hidden amongst rows of red and white blooms which showed the challenge.

Attempting to make the identification of suspect cells easier, various techniques have been used.