A Sleep Disorder Detection Model based on EEG Cross-Frequency Coupling and Random Forest

Abstract

Study objectives: Sleep disorders are medical disorders of the sleep architecture of a subject, and based on their severity they can interfere with mental, emotional and physical functioning. The most common ones are insomnia, narcolepsy, sleep apnea, bruxism, etc. There is an increment risk of developing sleep disorders in elderly like insomnia, periodic leg movements, rapid eye movement (REM) behaviour disorders, sleep disorder breathing, etc. Consequently, their accurate diagnosis and classification are important steps towards an early stage treatment that could save the life of a patient. The Electroencephalographic (EEG) signal is the most sensitive and important biosignal, which is able to capture the brain sleep activity that is sensitive to sleep. In this study, we attempt to analyse EEG sleep activity via complementary cross-frequency coupling (CFC) estimates that will further feed a classifier, aiming to discriminate sleep disorders. Methods: We adapted an open EEG Physionet Database with recordings that were grouped into seven sleep disorders and a healthy control. The EEG brain activity from common sensors has been analysed with two basic types of cross-frequency coupling (CFC). Finally, a Random Forest (RF) classification model was built on CFC patterns , that were extracted from non-cyclic alternating pattern (CAP) epochs. Results: Our RFCFC model succeeded a 74% multiclass accuracy (accuracy via random guessing 1/8 = 12.5%). Both types of CFC, PAC and AAC patterns contribute to the accuracy of the RF model , thus supporting their complementary information. Conclusion: CFC patterns, in conjunction with the RF classifier proved a valuable biomarker for the classification of sleep disorders.

Ongoing neural oscillations predict the post-stimulus outcome of closed loop auditory stimulation during slow-wave sleep

“…Random forest was therefore chosen because it presented the best trade-off between accuracy and computation time. Likewise, RF has also proved to be efficient in other problems based on sleep EEG (da Silveira et al, 2017;Dimitriadis et al, 2020). Consequently, we independently trained classification models for SHAM and STIM conditions using the RF algorithm.…”

A hybrid unsupervised and supervised learning approach for postictal generalized EEG suppression detection

“…Misclassification and out-of-bag metrics are used to adjust the weight of each tree (Li et al, 2017 ). Random forest has been leveraged for detecting various clinic events from EEG recordings (Wei et al, 2020 ; Abou-Abbas et al, 2021 ; Dimitriadis et al, 2021 ; Messaoud and Chavez, 2021 ).…”

EEG based Major Depressive disorder and Bipolar disorder detection using Neural Networks: A review

“…Electroencephalogram (EEG) is a non-invasive, effective, and powerful tool for recording the electrical activity of the brain and for the diagnosis of various mental disorders such as MDD [8], BD [9], anxiety [10], schizophrenia [11], and sleep disorders [12]. Due to these mental disorders or anomalies specifically depression and bipolar disorder, the body releases cortisol to the brain which affects the neurons production and communication and consequently slowing down the functionality of some parts of brain and changing the electrical activities patterns.…”