Scientists are developing an artificial glass liver to improve understanding of how the organ works.
Dr Walker is developing the model
The team, from the University of Leeds, hope their work will eventually lead to better treatments for people with liver disease.
They also hope that the lessons learned will be used to engineer liver tissue to replace that damaged by disease.
The first aim will be to replicate exactly the way the real organ is geometrically constructed.
Once this is achieved, it should then be possible to study how the cells function within that structure.
Lead researcher Dr Peter Walker said the liver was a very complex organ, which was still not fully understood.
He hoped his artificial creation would begin to help solve some of the mysteries.
"It may provide an alternative to animal testing for hepatic drugs and bring us one step closer to being able to engineer liver tissue."
The artificial organ - to be used like a dialysis machine - will be constructed of tiny hexagonal glass plates with channels running from their edges to the centre.
The minute channels - less than 1/200 of a millimetre wide - are lined with liver cells which reproduce the cleaning work of the organ.
The blood flows to the edge of each hexagon and down the channels, cleaned by cells as it goes, before exiting through a central "vein".
The liver is the only organ in the body where blood from veins and arteries is mixed together.
The arterial blood, fresh from the lungs, provides the cells with the oxygen they need to function, while venal blood contains the impurities for the liver to clean out.
Dr Walker said: "A major problem with artificial livers is that as the blood runs through, it loses too much oxygen, so cells at the end of the line are no longer effective.
"Reproducing the exact layout of the liver should overcome this problem, as cells will behave as they do in the natural liver, performing different cleaning functions dependent on their position and the level of oxygen they receive."
He is creating computer simulations to model the liver, so optimum channel size, flow rate and density of cells can be calculated as accurately as possible, before the artificial liver is constructed in the laboratory.
Professor Humphrey Hodgson, of the Royal Free Hospital, London, told BBC News Online that other researchers were working on similar projects around the world.
He said the theory was sound, but it would probably prove very difficult to develop a machine which could be of practical benefit to patients, not least because liver cells did not survive for very long outside the body.
"If it works it would be fantastic, so jolly good luck to them," he said.