Artificial intelligence can be used to detect the six years of Alzheimer's before a patient is usually diagnosed, according to a study.
Doctors used the self-learning computer to detect changes in brain scans too subtle for human beings to see.
The system was able to identify dementia in 40 patients an average of six years before being diagnosed formally.
The British expert from AI, Prof. Noel Sharkey, from the University of Sheffield, said of the findings: "This is precisely the type of task that cuts deep learning: finding high-level standards in the data.
"Although the sample sizes and test groups were relatively small, the results are so promising that a much more useful study would be worthwhile."
The Boffins of the University of California trained the computer with more than 2,100 explorations of 1,002 patients.
Scans measure brain activity following the absorption of a radioactive liquid injected into the blood.
Research has linked Alzheimer's development with particular changes in certain brain regions, but it can be difficult to detect.
The Alzheimer's algorithm was able to teach to recognize patterns in brain scans that indicated disease.
As a final test, 40 sets were received from 40 patients who had never studied before.
It turned out to be 100 percent accurate in detecting Alzheimer's disease many years before the patient was diagnosed later.
Dr. Jae Ho Sohn, who worked on the project, said: "We were very pleased with the performance of the algorithm.
"He was able to predict each case that progressed to Alzheimer's disease."
Early detection of Alzheimer's may open new ways of slowing down or even slowing down the progression of the disease.
"The diseases that cause dementia begin in the brain until 20 years before the symptoms begin to show themselves, presenting a showcase of opportunity so that we can intervene before widespread damage occurs," said Carol Routledge, of Alzheimer's & # 39; s Research UK.
"This study highlights the potential of automatic learning to help early detection of Alzheimer's disease, but the results must be confirmed in groups of much older people before being able to adequately assess the power of this approach."
The research is published in the latest issue of Radiology magazine.
Click for more from The Sun.