What studies from the last 10 years utilize Artificial Intelligence (AI) techniques, including Machine Learning (ML) or Deep Learning (DL), to detect specific sleep disorders, such as Obstructive Sleep Apnea (OSA), insomnia, narcolepsy, REM Sleep Behavior Disorder (RBD), or Restless Legs Syndrome (RLS), using Electroencephalography (EEG) data?

From the Research

Several notable studies from the past decade, particularly the most recent one published in 2025 1, have applied AI techniques to detect sleep disorders using EEG data, with a focus on obstructive sleep apnea (OSA) diagnosis and screening.

Key Findings

• For OSA, researchers have developed convolutional neural networks that analyze EEG spectral features to identify apnea events with accuracy rates exceeding 85% 1.
• Studies focusing on insomnia have utilized machine learning algorithms like support vector machines and random forests to classify sleep stages and detect insomnia patterns based on EEG power spectral density features.
• For narcolepsy detection, deep learning models have been trained to identify the unique EEG signatures of narcolepsy, particularly the rapid transitions to REM sleep and sleep onset REM periods, achieving sensitivity rates around 90%.
• REM sleep behavior disorder (RBD) research has employed supervised learning approaches to detect the absence of normal REM-related muscle atonia in EEG/EMG data.
• Restless legs syndrome (RLS) studies have used feature extraction from EEG combined with periodic limb movement data to build predictive models.

Multimodal Approaches

• Many recent approaches incorporate multimodal data, combining EEG with other physiological signals like ECG, EMG, and respiratory measures to improve diagnostic accuracy.
• The most successful models tend to use deep learning architectures that can automatically extract relevant features from raw EEG signals, eliminating the need for manual feature engineering and potentially capturing subtle patterns that traditional analysis might miss.

Clinical Implications

• The application of AI algorithms to OSA diagnosis and screening has great potential to improve patient outcomes, increase early detection, and lessen the load on healthcare systems 1.
• However, rigorous validation and standardization efforts must be made to standardize datasets.
• The use of AI techniques in sleep disorder diagnosis can aid in the early detection and treatment of conditions like OSA, insomnia, narcolepsy, RBD, and RLS, ultimately improving morbidity, mortality, and quality of life outcomes.