Evaluating and Improving Automatic Sleep Spindle Detection by Using Multi-Objective Evolutionary Algorithms
Sleep spindles are brief bursts of brain activity in the sigma frequency range (11–16 Hz) measured by electroencephalography (EEG) mostly during non-rapid eye movement (NREM) stage 2 sleep. These oscillations are of great biological and clinical interests because they potentially play an important r...
Main Authors: | , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Frontiers Media S.A.
2017-05-01
|
Series: | Frontiers in Human Neuroscience |
Subjects: | |
Online Access: | http://journal.frontiersin.org/article/10.3389/fnhum.2017.00261/full |
id |
doaj-1168d82aace6490d8c02e7b44f77ac14 |
---|---|
record_format |
Article |
spelling |
doaj-1168d82aace6490d8c02e7b44f77ac142020-11-25T02:36:30ZengFrontiers Media S.A.Frontiers in Human Neuroscience1662-51612017-05-011110.3389/fnhum.2017.00261243651Evaluating and Improving Automatic Sleep Spindle Detection by Using Multi-Objective Evolutionary AlgorithmsMin-Yin Liu0Adam Huang1Norden E. Huang2Norden E. Huang3Department of Biomedical Sciences and Engineering, Institute of Systems Biology and Bioinformatics, National Central UniversityTaoyuan, TaiwanResearch Center for Adaptive Data Analysis, National Central UniversityTaoyuan, TaiwanDepartment of Biomedical Sciences and Engineering, Institute of Systems Biology and Bioinformatics, National Central UniversityTaoyuan, TaiwanResearch Center for Adaptive Data Analysis, National Central UniversityTaoyuan, TaiwanSleep spindles are brief bursts of brain activity in the sigma frequency range (11–16 Hz) measured by electroencephalography (EEG) mostly during non-rapid eye movement (NREM) stage 2 sleep. These oscillations are of great biological and clinical interests because they potentially play an important role in identifying and characterizing the processes of various neurological disorders. Conventionally, sleep spindles are identified by expert sleep clinicians via visual inspection of EEG signals. The process is laborious and the results are inconsistent among different experts. To resolve the problem, numerous computerized methods have been developed to automate the process of sleep spindle identification. Still, the performance of these automated sleep spindle detection methods varies inconsistently from study to study. There are two reasons: (1) the lack of common benchmark databases, and (2) the lack of commonly accepted evaluation metrics. In this study, we focus on tackling the second problem by proposing to evaluate the performance of a spindle detector in a multi-objective optimization context and hypothesize that using the resultant Pareto fronts for deriving evaluation metrics will improve automatic sleep spindle detection. We use a popular multi-objective evolutionary algorithm (MOEA), the Strength Pareto Evolutionary Algorithm (SPEA2), to optimize six existing frequency-based sleep spindle detection algorithms. They include three Fourier, one continuous wavelet transform (CWT), and two Hilbert-Huang transform (HHT) based algorithms. We also explore three hybrid approaches. Trained and tested on open-access DREAMS and MASS databases, two new hybrid methods of combining Fourier with HHT algorithms show significant performance improvement with F1-scores of 0.726–0.737.http://journal.frontiersin.org/article/10.3389/fnhum.2017.00261/fullsleep spindlesautomatic detectionHilbert-Huang transformperformance assessmentmulti-objective evolutionary algorithmPareto front |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Min-Yin Liu Adam Huang Norden E. Huang Norden E. Huang |
spellingShingle |
Min-Yin Liu Adam Huang Norden E. Huang Norden E. Huang Evaluating and Improving Automatic Sleep Spindle Detection by Using Multi-Objective Evolutionary Algorithms Frontiers in Human Neuroscience sleep spindles automatic detection Hilbert-Huang transform performance assessment multi-objective evolutionary algorithm Pareto front |
author_facet |
Min-Yin Liu Adam Huang Norden E. Huang Norden E. Huang |
author_sort |
Min-Yin Liu |
title |
Evaluating and Improving Automatic Sleep Spindle Detection by Using Multi-Objective Evolutionary Algorithms |
title_short |
Evaluating and Improving Automatic Sleep Spindle Detection by Using Multi-Objective Evolutionary Algorithms |
title_full |
Evaluating and Improving Automatic Sleep Spindle Detection by Using Multi-Objective Evolutionary Algorithms |
title_fullStr |
Evaluating and Improving Automatic Sleep Spindle Detection by Using Multi-Objective Evolutionary Algorithms |
title_full_unstemmed |
Evaluating and Improving Automatic Sleep Spindle Detection by Using Multi-Objective Evolutionary Algorithms |
title_sort |
evaluating and improving automatic sleep spindle detection by using multi-objective evolutionary algorithms |
publisher |
Frontiers Media S.A. |
series |
Frontiers in Human Neuroscience |
issn |
1662-5161 |
publishDate |
2017-05-01 |
description |
Sleep spindles are brief bursts of brain activity in the sigma frequency range (11–16 Hz) measured by electroencephalography (EEG) mostly during non-rapid eye movement (NREM) stage 2 sleep. These oscillations are of great biological and clinical interests because they potentially play an important role in identifying and characterizing the processes of various neurological disorders. Conventionally, sleep spindles are identified by expert sleep clinicians via visual inspection of EEG signals. The process is laborious and the results are inconsistent among different experts. To resolve the problem, numerous computerized methods have been developed to automate the process of sleep spindle identification. Still, the performance of these automated sleep spindle detection methods varies inconsistently from study to study. There are two reasons: (1) the lack of common benchmark databases, and (2) the lack of commonly accepted evaluation metrics. In this study, we focus on tackling the second problem by proposing to evaluate the performance of a spindle detector in a multi-objective optimization context and hypothesize that using the resultant Pareto fronts for deriving evaluation metrics will improve automatic sleep spindle detection. We use a popular multi-objective evolutionary algorithm (MOEA), the Strength Pareto Evolutionary Algorithm (SPEA2), to optimize six existing frequency-based sleep spindle detection algorithms. They include three Fourier, one continuous wavelet transform (CWT), and two Hilbert-Huang transform (HHT) based algorithms. We also explore three hybrid approaches. Trained and tested on open-access DREAMS and MASS databases, two new hybrid methods of combining Fourier with HHT algorithms show significant performance improvement with F1-scores of 0.726–0.737. |
topic |
sleep spindles automatic detection Hilbert-Huang transform performance assessment multi-objective evolutionary algorithm Pareto front |
url |
http://journal.frontiersin.org/article/10.3389/fnhum.2017.00261/full |
work_keys_str_mv |
AT minyinliu evaluatingandimprovingautomaticsleepspindledetectionbyusingmultiobjectiveevolutionaryalgorithms AT adamhuang evaluatingandimprovingautomaticsleepspindledetectionbyusingmultiobjectiveevolutionaryalgorithms AT nordenehuang evaluatingandimprovingautomaticsleepspindledetectionbyusingmultiobjectiveevolutionaryalgorithms AT nordenehuang evaluatingandimprovingautomaticsleepspindledetectionbyusingmultiobjectiveevolutionaryalgorithms |
_version_ |
1724799816253308928 |