A Two Stage Approach for the Automatic Detection of Insomnia

Mostafa Shahin, Lamana Mulaffer, Thomas Penzel, Beena Ahmed

Research output: Contribution to journalArticle

Abstract

Chronic insomnia can significantly impair an individual's quality of life leading to a high societal cost. Unfortunately, limited automated tools exist that can assist clinicians in the timely detection of insomnia. In this paper, we propose a two stage approach to automatically detect insomnia from an overnight EEG recording. In the first stage we trained a sleep stage scoring model and an epoch-level insomnia detection model. Both models are deep neural network (DNN)- based which are fed by a set of temporal and spectral features derived from 2 EEG channels. In the second stage we computed two subject-level feature sets. One is computed using the output of the sleep stage scoring model and consists of the sleep stage ratios, the stage pair ratios and the stage transition ratios. The second feature set is derived from the output of the epoch-level insomnia detection model and represents the ratio of detected insomniac epochs in each stage and their average posterior probability. These features are then used to train a final binary classifier to classify each subject as control, i.e., with no sleep complaints, or insomniac. We compared 5 different binary classifiers, namely the linear discriminant analysis (LDA), the classification and regression trees (CART) and the support vector machine (SVM) with linear, Gaussian and sigmoid kernels. The system was evaluated against data collected from 115 participants, 61 control and 54 with insomnia, and achieved $F1$ score, sensitivity and specificity of 0.88, 84% and 91% respectively.

Fingerprint

Sleep Initiation and Maintenance Disorders
Sleep Stages
Electroencephalography
Classifiers
Discriminant analysis
Neural Networks (Computer)
Discriminant Analysis
Sigmoid Colon
Support vector machines
Sleep
Quality of Life
Costs and Cost Analysis
Sensitivity and Specificity
Costs

ASJC Scopus subject areas

  • Signal Processing
  • Biomedical Engineering
  • Computer Vision and Pattern Recognition
  • Health Informatics

Cite this

@article{697211873d6848719b0e31c146fbf32c,
title = "A Two Stage Approach for the Automatic Detection of Insomnia",
abstract = "Chronic insomnia can significantly impair an individual's quality of life leading to a high societal cost. Unfortunately, limited automated tools exist that can assist clinicians in the timely detection of insomnia. In this paper, we propose a two stage approach to automatically detect insomnia from an overnight EEG recording. In the first stage we trained a sleep stage scoring model and an epoch-level insomnia detection model. Both models are deep neural network (DNN)- based which are fed by a set of temporal and spectral features derived from 2 EEG channels. In the second stage we computed two subject-level feature sets. One is computed using the output of the sleep stage scoring model and consists of the sleep stage ratios, the stage pair ratios and the stage transition ratios. The second feature set is derived from the output of the epoch-level insomnia detection model and represents the ratio of detected insomniac epochs in each stage and their average posterior probability. These features are then used to train a final binary classifier to classify each subject as control, i.e., with no sleep complaints, or insomniac. We compared 5 different binary classifiers, namely the linear discriminant analysis (LDA), the classification and regression trees (CART) and the support vector machine (SVM) with linear, Gaussian and sigmoid kernels. The system was evaluated against data collected from 115 participants, 61 control and 54 with insomnia, and achieved $F1$ score, sensitivity and specificity of 0.88, 84{\%} and 91{\%} respectively.",
author = "Mostafa Shahin and Lamana Mulaffer and Thomas Penzel and Beena Ahmed",
year = "2018",
month = "7",
day = "1",
doi = "10.1109/EMBC.2018.8512360",
language = "English",
volume = "2018",
pages = "466--469",
journal = "Annual International Conference of the IEEE Engineering in Medicine and Biology - Proceedings",
issn = "0589-1019",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - JOUR

T1 - A Two Stage Approach for the Automatic Detection of Insomnia

AU - Shahin, Mostafa

AU - Mulaffer, Lamana

AU - Penzel, Thomas

AU - Ahmed, Beena

PY - 2018/7/1

Y1 - 2018/7/1

N2 - Chronic insomnia can significantly impair an individual's quality of life leading to a high societal cost. Unfortunately, limited automated tools exist that can assist clinicians in the timely detection of insomnia. In this paper, we propose a two stage approach to automatically detect insomnia from an overnight EEG recording. In the first stage we trained a sleep stage scoring model and an epoch-level insomnia detection model. Both models are deep neural network (DNN)- based which are fed by a set of temporal and spectral features derived from 2 EEG channels. In the second stage we computed two subject-level feature sets. One is computed using the output of the sleep stage scoring model and consists of the sleep stage ratios, the stage pair ratios and the stage transition ratios. The second feature set is derived from the output of the epoch-level insomnia detection model and represents the ratio of detected insomniac epochs in each stage and their average posterior probability. These features are then used to train a final binary classifier to classify each subject as control, i.e., with no sleep complaints, or insomniac. We compared 5 different binary classifiers, namely the linear discriminant analysis (LDA), the classification and regression trees (CART) and the support vector machine (SVM) with linear, Gaussian and sigmoid kernels. The system was evaluated against data collected from 115 participants, 61 control and 54 with insomnia, and achieved $F1$ score, sensitivity and specificity of 0.88, 84% and 91% respectively.

AB - Chronic insomnia can significantly impair an individual's quality of life leading to a high societal cost. Unfortunately, limited automated tools exist that can assist clinicians in the timely detection of insomnia. In this paper, we propose a two stage approach to automatically detect insomnia from an overnight EEG recording. In the first stage we trained a sleep stage scoring model and an epoch-level insomnia detection model. Both models are deep neural network (DNN)- based which are fed by a set of temporal and spectral features derived from 2 EEG channels. In the second stage we computed two subject-level feature sets. One is computed using the output of the sleep stage scoring model and consists of the sleep stage ratios, the stage pair ratios and the stage transition ratios. The second feature set is derived from the output of the epoch-level insomnia detection model and represents the ratio of detected insomniac epochs in each stage and their average posterior probability. These features are then used to train a final binary classifier to classify each subject as control, i.e., with no sleep complaints, or insomniac. We compared 5 different binary classifiers, namely the linear discriminant analysis (LDA), the classification and regression trees (CART) and the support vector machine (SVM) with linear, Gaussian and sigmoid kernels. The system was evaluated against data collected from 115 participants, 61 control and 54 with insomnia, and achieved $F1$ score, sensitivity and specificity of 0.88, 84% and 91% respectively.

UR - http://www.scopus.com/inward/record.url?scp=85056653950&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85056653950&partnerID=8YFLogxK

U2 - 10.1109/EMBC.2018.8512360

DO - 10.1109/EMBC.2018.8512360

M3 - Article

VL - 2018

SP - 466

EP - 469

JO - Annual International Conference of the IEEE Engineering in Medicine and Biology - Proceedings

JF - Annual International Conference of the IEEE Engineering in Medicine and Biology - Proceedings

SN - 0589-1019

ER -