Although Alzheimer's disease affects tens of millions of people around the world, it remains difficult to detect at an early stage. But scientists who deal with the possibilities of artificial intelligence in medicine have discovered that technology can help to diagnose early traumatic disease. The California team recently published a report on its study in the Radiology journal and demonstrated that once you trained the neural network, it was able to accurately diagnose Alzheimer's disease in a limited number of patients based on visual imaging of the brain years before the patients concerned are diagnosed by a doctor.
The team uses brain imaging (FDG-PET imaging) to train and test the neural network. In the FDG, images of the patient's bloodstream are injected with a type of radioactive glucose, and then his body tissue, including the brain, pushes him to the surface. Scientists and physicians can use PET scan to detect the metabolic activity of this tissue, depending on the amount of FDG administered.
The FDG-PET method is used to diagnose Alzheimer's disease, and patients with this disease usually display lower levels of metabolic activity in certain parts of the brain. However, experts need to analyze these images to find evidence of the disease and this becomes very difficult because moderate cognitive impairment and Alzheimer's disease can result in similar scan results.
Therefore, the team uses 2,109 FDG-PET images from 1002 patients, forming their neural network at 90% and testing it at the remaining 10%. It also performs a single set of 40 scanned patients between 2006 and 2016, then compares the findings of artificial intelligence with those of a group of specialists who analyze the same data.
With a separate set of test data, Artificial Intelligence is able to diagnose Alzheimer's patients with an accuracy of 100% and with an accuracy of 82% those who do not suffer from treacherous disease. He can also make predictions on average more than six years before. By comparison, the group of physicians who examined the same scanned images identified patients with Alzheimer's disease in 57% of cases and those without disease – in 91%. However, differences in machine and human performance are not so noticeable when it comes to diagnosing mild cognitive impairment that is not typical of Alzheimer's disease.
Researchers have noted that their research has several limitations, including a small amount of test data and limited types of training data.