Scientists believe that a revolutionary new implant that can help restore memory in people who suffered a stroke or brain injury could start human trials in the next two years.
Researchers from the University of Southern California and North Carolina's Wake Forest University among others believe they will see a memory device being implanted in human volunteers within two years and it may be made available to patients in five to 10 years.
"I never thought I'd see this in my lifetime," said Ted Berger, professor of biomedical engineering at the University of Southern California in Los Angeles.
The scientists believe they have cracked how long-term memories are made, stored and retrieved and how to replicate this process in brains that are damaged, particularly by stroke or localised injury, CNN reported.
Berger said they record a memory being made, in an undamaged area of the brain, then use that data to predict what a damaged area "downstream" should be doing.
Electrodes are then used to stimulate the damaged area to replicate the action of the undamaged cells.
They concentrate on the hippocampus - part of the cerebral cortex which sits deep in the brain - where short-term memories become long-term ones.
Berger has looked at how electrical signals travel through neurons there to form those long-term memories and has used his expertise in mathematical modelling to mimic these movements using electronics.
The team's experiments on rats and monkeys have shown that certain brain functions can be replaced with signals via electrodes.
"Right now it's not a device, it's a fair amount of equipment. We're probably looking at devices in the five to 10 year range for human patients," said Rob Hampson, associate professor of physiology and pharmacology at Wake Forest University.
Berger believes in the future drugs and implants could be used together to treat early dementia.
Drugs could be used to enhance the action of cells that surround the most damaged areas, and the team's memory implant could be used to replace a lot of the lost cells in the centre of the damaged area.
"I think the best strategy is going to involve both drugs and devices," he said.
The team has unfortunately found that its method can't help patients with advanced dementia.
"When looking at a patient with mild memory loss, there's probably enough residual signal to work with, but not when there's significant memory loss," Hampson said.