AI recognizes Alzheimer’s years before diagnosis

The earlier Alzheimer is diagnosed, the greater the chances of at least slowing down the disease. An adaptive algorithm is now setting new standards.

n the fight against Alzheimer’s disease, early detection is particularly important. If the still incurable dementia is diagnosed at an early stage, drugs can at least slow down its progression.

“If we do not diagnose Alzheimer’s until clear symptoms appear, the loss of brain volume is already so great that it is usually too late for effective intervention,” explains Jae Ho Sohn.

Together with his team from the University of California in San Francisco, the physician has developed a new tool for the early detection of Alzheimer’s: an adaptive algorithm that can reliably predict dementia years before it is diagnosed by the physician.

The researchers focused their development on subtle metabolic changes in the brain caused by the onset of the disease. Such changes can be visualized using an imaging method known as positron emission tomography.

However, the traces in the early stages of the disease are so weak that even experienced physicians can hardly recognize them. “It is easier for people to find specific biomarkers of a disease,” explains Sohn. “However, metabolic changes are much more subtle processes.

The researchers trained their artificial intelligence using data from the Alzheimer’s Disease Neuroimaging Initiative (ADNI). Among other things, this data collection contains thousands of PET images of Alzheimer’s patients in very early stages of the disease. The researchers used 90 percent of these images to train the algorithm and the remaining 10 percent to monitor success.

For the final test, the AI had to analyse 40 images that had not yet been presented to it. Sohn describes the result as follows: “The algorithm was able to reliably identify every case that later led to the outbreak of Alzheimer’s disease”.

In addition to the 100 percent hit rate, the physicians were particularly impressed by the very early identification of disease cases. On average, the system identified the symptoms more than six years before the actual diagnosis of the disease. “We were delighted with this result,” said Sohn. However, the physician also knows that the test series was still comparatively small and that further tests have to confirm the result.

Nevertheless, he sees in his algorithm the potential for an important tool in the treatment of Alzheimer’s disease: “If we succeed in recognizing the disease earlier, the researchers will be able to find better ways to slow down the course of the disease or even stop it altogether.

Leave a Reply

Your email address will not be published. Required fields are marked *