Researchers in Britain, led by a Greek woman from the Diaspora, have developed an innovative artificial intelligence system that can recognize the first signs of dementia in the brain, before the first symptoms appear in patients and before neurologists are able to make a diagnosis.
The system, who does not need more than one brain image to draw conclusions, thanks to a special algorithm, has already begun to be clinically tested, although it is too early to say when it will be possible to put it into practice.
The algorithm has been "trained" with thousands of brain images of people with dementia, so he can then make comparisons when asked to "read" a new brain image. He is able to recognize early signs of dementia that can not "catch" the eye of even a specialist doctor, thus paving the way for the detection of neurodegeneration of the brain in a timely manner, several years earlier than,what is possible today.
Investigators, led by Professor Zoe Kourtzis of the University of Cambridge and the Alan Turing Institute, which created the system, According to the BBC and the British "Guardian" and "Independent", optimistic that this will be a valuable assistant to neurologists in the future. As mentioned by Dr. Kourtzi, "If we intervene in time, treatments will be able to start earlier and slow the progression of the disease, while at the same time a greater brain damage will be avoided ".
Preclinical tests have shown that the system can diagnose dementia even when there is not the slightest obvious sign of damage to the brain. also, can predict whether dementia will remain stable for many years and whether it will develop sooner or later.
Clinical trials at Adenbrook Hospital in Cambridge and other British clinics, with the participation of approx 500 led by Cam Ridge neurologist Dr. Tim Rittman, they will show how much the system "works" in practice. As he said, the new "smart" system is "a fantastic development".
Mrs.. Kourtzis studied Experimental Psychology at the University of Crete and received her PhD in Cognitive Neuroscience from Rutgers University in the USA. He did postdoctoral research at MIT and Harvard, specializing in Cognitive Neuroimaging, and then conducted research at the German Max Planck Institute for Biological Government and at the British University of Birmingham. From 2013 teaches Experimental Psychology at the University of Cambridge and is considered a specialist in Cognitive Computational Neuroscience, being a pioneer in the use of Artificial Intelligence in mind and brain health.
The development of algorithms for clinical use is a rapidly growing field internationally, "Marrying" computer science with Neurobiology and Medicine. Various "smart" systems for early diagnosis have already been developed - from imaging tests- various diseases.