Sunday , October 17 2021

AI recognizes Alzheimer's for many years before the diagnosis

Shooting human brains with PET

Thousands of PET images of early-stage Alzheimer's patients used scientists to train their artificial intelligence. (Photo: Radiological Society of North America)

BerlinIn the fight against Alzheimer's disease, early detection is particularly important. If still incurable dementia is detected early, it can at least slow down their treatment with medication.

"If we diagnose Alzheimer's disease only if there are clear symptoms, the loss of brain volume is so high that it is usually too late for effective intervention," explains Jae Ho Sohn.

Together with his team at the University of California, San Francisco, the doctor developed a new tool for early detection of Alzheimer's disease: an adaptive algorithm that predicts dementia reliably for many years before a doctor's diagnosis.

Researchers focused on the development of subtle metabolic changes in the brain that are caused by the onset of the disease. Such changes can be visualized using the imaging technique known as positron emission tomography (PET).

Seven factors favoring Alzheimer's disease

Traces in the early stages of the disease are so weak that it is difficult to recognize them even for experienced doctors. "It's easier for people to find specific biomarkers of disease," explains Sohn. "But metabolic changes are much more subtle processes."

Researchers have trained their artificial intelligence using data from the Allergy Neuroimaging Injection (ADNI) initiative. This dataset contains, among others, thousands of photos of Alzheimer's patients with PET in the very early stages of the disease. 90 percent of these recordings, scientists used the algorithm to train, the remaining 10 percent to control success.

For the final test, AI had to finally analyze 40 images that were not presented to her until then. The result describes the son as follows: "The algorithm was able to reliably detect every case that later came to the beginning of Alzheimer's disease."

In addition to the 100 percent hit rate, doctors were especially impressed by very early case identification. On average, the system recognized symptoms more than six years before the actual diagnosis of the disease. "We were delighted with this result," says Son. However, the doctor also knows that the test series was still relatively small, and further tests must confirm the result.

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

Source link