Thursday , March 4 2021

AI recognizes Alzheimer's already years before the diagnosis



Shooting human brain with PET

Thousands of PET images from Alzheimer's patients at an early stage used researchers to train AI. (Photo: North American Radiological Society)

BerlinEarly detection of Alzheimer's disease is particularly important. If an incurable dementia is detected at an early stage, it may at least slow the course of the medication.

"If we diagnose Alzheimer's disease only when clear symptoms occur, the volume of the brain is so large that it is usually too late to be effective," Jae Ho Sohn explains.

At California University in San Francisco, his doctor developed a new tool for early detection of Alzheimer's disease: an adaptive algorithm that predicts dementia a year before medical diagnosis.

Researchers focused on developing subtle metabolic changes in the brain due to the outbreak of the disease. Such changes can be visualized using the known imaging technology of positron emission tomography (PET).

Seven factors that favor Alzheimer's disease

The traces in the early stages of the disease are, however, so weak that they are barely recognizable even to experienced doctors. "It is easier for people to find certain biomarkers of illness," explains Sohn. "But metabolic changes are much more subtle processes."

Scientists trained artificial intelligence using the Alzheimer's Disease Neuroscience Initiative (ADNI). Among other things, this data collection includes thousands of PET images of Alzheimer's patients at a very early stage. 90 percent of these recordings, researchers used algorithm training and the remaining 10 percent to manage success.

For the final test, AI had to finally analyze 40 images that had not been delivered to her until now. The result describes the boy as follows: "The algorithm was able to reliably detect all cases that later arose in the onset of Alzheimer's disease."

In addition to the 100% gain in productivity, doctors primarily affected the early identification of cases. On average, the system identified the symptoms more than six years before the actual diagnosis of the disease. "We have been happy about this result," says Son. However, the doctor also knows that the test kit was still relatively small and further studies need to confirm the result.

Yet, in his algorithm, he sees the potential of an important tool in Alzheimer's care: "If we can detect the disease earlier, it will allow scientists to find better ways to slow down or even stop the disease process."


Source link