New AI Tool By Mass General Brigham Predicts Brain Decline Years in Advance

The findings, published in the Journal of Alzheimer's Disease, highlight the tool's potential to transform dementia prevention and management.

Researchers at Mass General Brigham have developed an artificial intelligence (AI) tool capable of predicting brain decline years before symptoms appear, potentially allowing for earlier intervention and treatment.

The AI tool analyzes subtle changes in brain activity during sleep using electroencephalography (EEG) data to detect early signs of cognitive impairment.

The findings, published in the Journal of Alzheimer's Disease, highlight the tool's potential to transform dementia prevention and management. The study involved sleep data from a group of women over 65 who were tracked for five years.

Out of 281 participants with normal cognitive function at the start of the study, 96 developed cognitive impairment by the time they were reassessed five years later.

High Accuracy in Prediction

The AI tool was trained to extract brainwave patterns from EEG data, particularly focusing on subtle changes in the gamma band frequencies during deep sleep.

The tool successfully flagged 85% of the participants who eventually developed cognitive decline, achieving an overall accuracy rate of 77%.

"This could completely change how we approach dementia prevention," said Dr Shahab Haghayegh, lead author of the study.

"Using advanced AI and information theory tools, we can detect subtle changes in brain wave patterns during sleep that signal future cognitive impairment, offering a window of opportunity for intervention years before symptoms appear."

Potential for Early Intervention

According to the World Health Organization (WHO), approximately 55 million people worldwide are living with dementia. Research indicates that subtle changes in behavior and physiological functions often precede cognitive decline.

Early detection could provide an opportunity for lifestyle changes such as increased physical activity, mental stimulation, and dietary adjustments, which may help preserve cognitive health.

Dr Haghayegh emphasized that the tool's early detection capabilities could empower individuals to adopt preventive measures before the onset of symptoms, improving long-term health outcomes.

While the results are promising, the researchers stressed the need for more extensive studies involving more diverse populations to validate and refine the tool's effectiveness.

Expanding the study across different demographic groups will help ensure that the tool's predictive accuracy remains consistent and reliable across various population segments.

Stay tuned for more such updates on Digital Health News.