Artificial intelligence can be used to recognize Alzheimer's disease six years before the patient is diagnosed normally, studies show.
Physicians used a computer to detect changes in brain scan that are too subtle for humans.
The system was able to identify dementia in 40 patients on average six years before they were formally diagnosed.
British AI expert prof. Noel Sharkey of the University of Sheffield said: "This is precisely a task for which deep relaxation is excluded – searching for quality samples in data.
"Although the sample size and test sets were relatively small, the results are so promising that there would be much more study."
Boffins from the University of California trained a computer using over 2100 scans of 1,002 patients.
The images measure brain activity by monitoring the uptake of radioactive liquid injected into the blood.
Research has linked the development of Alzheimer's disease with specific changes in some areas of the brain, but these are hard to find out.
The Alzheimer's algorithm was able to learn to recognize patterns in the brain scans that indicated the disease.
As a final test, he received a set of 40 scans from 40 patients he had never examined before.
It turned out to be 100 percent accurate in detecting Alzheimer's disease many years before the patient was later diagnosed.
Dr. Jae Ho Sohn, who worked on the project, said: "We were very happy with the performance of the algorithm.
"He was able to predict every case that went into Alzheimer's disease."
Early detection of Alzheimer's disease could open the door to new ways of slowing down or even halting disease progression.
Dr. Carol Routledge, of Alzheimer's UK research, said: "Diseases that cause dementia begin in the brain for up to 20 years before signs begin to emerge, which is a vital opportunity for intervention before extensive damage.
"This study highlights the potential of machine learning to help with early detection of diseases such as Alzheimer's disease, but findings will have to be confirmed in much larger groups of people than we will be able to properly assess the strength of this approach."
The research is published in the latest issue of the journal Radiology.
Click on more information from The Sun.