The NHS is to use neural networking technology in a bid to stop fraud by patients and staff.
The NHS Counter Fraud Service (CFS) will use the technology to target its investigations more effectively. It has investigated 503 cases of potential fraud in the last year - leading to just 45 prosecutions.
Jim Gee, director of the CFS, told vnunet.com that he is looking to reach an agreement with software company SAS to use its tools for the analysis of fraud data.
"We would like to use some of the neural network software so that every time an incidence of fraud is discovered we can build it in," he said. "We think we have the data to be able to develop the process."
Neural newtorks are artificial systems capable of sophisticated - almost "intelligent" - calculations, similar to the human brain.
Gee said the CFS is also looking to build a "fraud map" from the data to show any patterns that might be useful to its investigations. "You have to use the information you have [in order] to make your systems as fraud-proof as possible," he said.
"Our data tells us how much fraud is taking place but we want to know the nature of the problem. We want to use the software to analyse the data to see the problem," he added.
Gee said NHS frauds tend to be individual cases of a medium to high value, but not very frequent. The CFS has already cut pharmaceutical patient fraud by £48m and dental patient fraud by £10m. The NHS has more than a million employees and a budget of £50bn.
In one case alone, £1.2m was paid back to the NHS; in another, assets worth £2m were frozen.
But Gee warned that the fight against fraud is an uphill struggle. "Each time you make an improvement the fraudsters make two," he said.
Found by calculating the strength of the material deep inside the crust of neutron stars
Can highlight in real-time the relevant regions of an image being described
Double legal trouble for Musk as he also faces civil lawsuit over renewed British pot-holer 'paedo' claims
Battery development could help boost performance of smartphones