Researchers from Severo Ochoa University Hospital and Carlos III University of Madrid have developed an artificial intelligence system that analyses electrical brain activity during sleep to detect Alzheimer's in its early stages. The study, published in GeroScience, paves the way for a non-invasive and cost-effective diagnosis.
Artificial intelligence has made its way into patients' beds to monitor their brains while they sleep. A team of researchers from Severo Ochoa University Hospital and Carlos III University of Madrid (UC3M) has demonstrated that analysing nocturnal brain waves using machine learning allows for the identification of early neuronal alterations associated with Alzheimer's. The work, published in the journal GeroScience, classifies patients into three biological subgroups based on their biomarker signatures.
A window into the brain while we sleep
Sleep is not just rest: during the night, the brain performs a cellular cleaning of metabolic waste, such as beta-amyloid protein. In fact, the accumulation of this protein is one of the first signs of Alzheimer's. “Records of electrical activity during sleep allow us to have a window into the biological processes occurring that, over the years, can lead to forgetfulness,” explains Arrate Muñoz-Barrutia, professor in the Department of Neuroscience and Biomedical Sciences at UC3M and principal investigator of the study.
The research has analysed a unique database that combines polysomnographies (records of nocturnal electrical activity) with protein expression data from the cerebrospinal fluid of 42 patients with mild to moderate Alzheimer's and 58 healthy controls. The data was collected by a team from Santa María University Hospital in Lleida and IRBLleida, led by doctors Gerard Piñol Ripoll and Ferran Barbé.
AI to detect Alzheimer's before symptoms
Alzheimer's is a neurodegenerative disease whose clinical symptoms appear between 10 and 20 years after the pathological processes in the brain begin. Currently, drugs are only effective if administered in the early stages, but diagnostic tests such as positron emission tomography (PET) or lumbar puncture are invasive and costly. “With our research, we seek a tool that allows for early, non-invasive, cost-effective diagnosis that can be extended to the majority of the population,” states Anna Michela Gaeta, a specialist in Pulmonology at Severo Ochoa University Hospital.
The AI algorithm analyses the electrical signals captured by electrodes placed on the skull throughout the night. “We use AI to identify changes that detect the accumulation of proteins in the brain, which will then lead to neurodegenerative diseases,” details Lorena Gallego Viñarás, from the Department of Neuroscience and Biomedical Sciences at UC3M. The AI has accurately differentiated healthy individuals from Alzheimer's patients and, by cross-referencing the data with biomarkers such as beta-amyloid protein (A42), phosphorylated tau (p-tau181), and neurofilament light chain (NfL), has identified three subgroups of patients with distinct biological signatures.
“AI does not aim to replace any type of medical test, but rather to support early diagnosis. This allows for a much more detailed analysis of information to facilitate earlier diagnoses and, consequently, to start treatments sooner,” adds Gallego Viñarás.
A future with home screening
In the future, these tools could work complementarily to blood tests, such as those for the p-tau217 protein, which are already beginning to be used in some hospitals. “There could be an economical screening pathway that can be carried out from home to detect preclinical Alzheimer's and simultaneously address sleep disorders,” the researchers point out. The next step will be to validate the results in a larger sample and develop a portable device that allows for recording at home, without the need to go to a hospital.
Severo Ochoa University Hospital, located in Leganés, thus establishes itself as a benchmark in sleep research and neurodegenerative diseases. For residents of the Community of Madrid, this advancement represents hope for more accessible and earlier diagnosis, crucial for combating a disease that affects over 800,000 people in Spain.
You may also like
