A study that was published recently has shown that a deep neural network model is actually able to predict the age of the brain in healthy patients. Researchers managed to do it based on electroencephalogram data recorded during sleep at night. The EEG has predicted the age of the brain, and it can uncover plenty of information about people with different diseases.
The model predicted the age with an error of 4.6 years. Researchers also found that there is a relationship between the Absolute Brain Age Index and seizure disorders, epilepsy, stroke, and poor sleep at night. We also learned that patients with depression, diabetes, hypertension, excessive daytime sleepiness, and memory problems have an elevated Brain Age Index in comparison to healthy people.
The study has shown that these results show that the health conditions are actually closely related to the deviations of one’s predicted age from one’s chronological age.
“While clinicians can only grossly estimate or quantify the age of a patient based on their EEG, this study shows an artificial intelligence model can predict a patient’s age with high precision. The model’s precision enables shifts in the predicted age from the chronological age to express correlations with major disease families and comorbidities. This presents the potential for identifying novel clinical phenotypes that exist within physiological signals utilizing AI model deviations,” stated lead author Yoav Nygate.
In order to find out the age of patients, scientists used EEG signals recorded during clinical sleep studies. They were performed by using overnight polysomnography. The results showed the initial evidence of using AI to learn about the age of the patient. The aim of the study is that, one day, the brain age index will play a considerable role in diagnosing brain health, as well as high blood pressure.