Neurology, first mathematical models for reading electroencephalograms: The goal is to understand the early symptoms of Alzheimer's disease

Memory lapses, temporary amnesia that leads to forgetting names, places, or appointments. It can happen to anyone, especially older people. But how can we tell when these symptoms are the result of normal aging or a period of stress, or when they are a warning sign of a possible brain disorder, such as the onset of Alzheimer's disease? A study coordinated by the Sant'Anna School of Advanced Studies in Pisa, in collaboration with the Careggi University Hospital and the University of Florence, has begun developing a new method that, thanks to a combination of mathematical models and electroencephalograms, could help detect the symptoms of Alzheimer's at an earlier stage, paving the way for new diagnostic techniques that, with the necessary verification, could support doctors in their clinical practice in the future.
The study, entitled ‘Digital Twins and Non-Invasive Recordings Enable Early Diagnosis of Alzheimer's Disease’, has been published in the journal Alzheimer's Research & Therapy.

“Not only will we be able to provide an increasingly reliable prediction of the risk of developing Alzheimer's in people who do not yet have obvious clinical symptoms, but we have managed to do so using a completely new method that is potentially much easier to use for hospitals and patients than the methods currently in use,” said Alberto Mazzoni, associate professor of Bioengineering at the Institute of Biorobotics at the Sant'Anna School and coordinator of the study.

Understanding the symptoms of Alzheimer's in advance

Alzheimer's disease is one of the major challenges of modern medicine, with a growing impact on patients, families, and healthcare systems. In recent years, research has focused increasingly on the prodromal stages of the disease, i.e., the period before the onset of overt symptoms of dementia.

“The technology is promising,” says Valentina Bessi, head of the Center for Cognitive Disorders and Dementia at the Careggi University Hospital in Florence and associate professor of neurology at the University of Florence, "and can be an additional tool to help doctors, who are familiar with the physical, psychological, and social complexity of patients, in their diagnosis.
Identifying Alzheimer's when clinical signs are still mild but biological changes are already present is now considered essential. Early diagnosis is opening up new possibilities for intervention, allowing access to innovative treatments that could slow the progression of the disease and improve quality of life."

The aim of this research, which lasted over four years, is to provide a prognosis regarding the possible emergence of Alzheimer's dementia in people who do not yet have symptoms severe enough to be clinically significant. Until now, the only way to determine whether memory lapses are the first signs of Alzheimer's was to resort to complex tests such as brain PET scans or cerebrospinal fluid testing.

In the study, we analyzed data from 124 people, 86 of whom had only subjective mild cognitive impairment. Our approach allowed us to predict the outcome of cerebrospinal fluid testing in 88% of cases based solely on electroencephalograms. In addition, we were able to predict 7/7 conversions to objectively measurable cognitive decline. Of course, these numbers are not very large and, above all, the observation period is relatively limited for phenomena that take years to develop, so it will be necessary to expand the available data and continue to follow patients over the next few years.

“We used a mathematical model that describes the change in brain activity as Alzheimer's progresses to investigate the signals that herald the onset of the disease,” explains Lorenzo Gaetano Amato, PhD student in Biorobotics at the Sant'Anna School and lead author of the discovery. "The next step was to analyze the brain activity of over 100 elderly people with mild memory problems using a simple electroencephalogram recording.electroencephalogram. By combining these analyses, we developed a personalized version of the brain model for each of them, which allowed us to understand which of them were at risk of developing Alzheimer's."

“This combination of advanced brain simulation methods combined with a simple electroencephalogram works better than the methods used so far. For now, this has been an all-Italian study, but we are working on a broader validation that also includes collaborations with European centers,” concludes Alberto Mazzoni.