Classification of Sleep Quality and Aging as a Function of Brain Complexity: A Multiband Non-Linear EEG Analysis

Area of Science:

• Neuroscience
• Sleep Science
• Computational Biology

Background:

• Understanding brain states is crucial for sleep hygiene interventions.
• Non-linear electroencephalography (EEG) features can differentiate awake brain states.
• Age and sleep quality significantly impact brain activity patterns.

Purpose of the Study:

• To classify brain states based on age and sleep quality using EEG complexity.
• To evaluate an algorithm's accuracy in predicting age and sleep quality.
• To explore the potential for personalized sleep interventions.

Main Methods:

• Collected resting-state EEG data from 58 participants.
• Assessed sleep quality using the Pittsburgh Sleep Quality Inventory (PSQI).
• Extracted ten non-linear EEG features and applied cross-validation classifiers.

Main Results:

• Accurately predicted age in good sleepers (75% mean accuracy).
• Moderately predicted sleep quality in older adults (70-72% accuracy).
• Successfully differentiated younger good sleepers from older poor sleepers (85% mean accuracy).

Conclusions:

• Brain state complexity effectively distinguishes between age groups.
• The algorithm shows potential for predicting sleep quality in older adults.
• This approach may personalize sleep interventions based on individual brain complexity.