RETRACTED ARTICLE: A novel system for automatic detection of K-complexes in sleep EEG

Yücelbaş, Cüneyt; Yücelbaş, Şule; Özşen, Seral; Tezel, Gülay; Küççüktürk, Serkan; Yosunkaya, Şebnem

doi:10.1007/s00521-017-2865-3

Cited by 22 publications

(7 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The k-complex [7], [8] is one of the important parameters in determining the epoch as N-REM stage-2, which would help the automatic sleep staging system in a great deal. Yucelbas [9] have presented a fusing-method with logic AND operations and Ranjan et al [10] propounded a fuzzy neural network approach to find out whether there were k-complex in related epochs using the time and frequency analysis, automatically. Camilleri et al [11] investigated the use of switching linear Gaussian state space models for the segmentation and automatic labelling of Stage 2 sleep EEG data characterised by spindles and K-complexes.…”

Section: Introductionmentioning

confidence: 99%

An Accurate Sleep Stages Classification Method Based on State Space Model

Shen

Guez

et al. 2019

IEEE Access

View full text Add to dashboard Cite

The classification of sleep stages is the process which helps to evaluate the quality of sleep and detect the sleep related disorders. Through analyzing the electroencephalography, the sleep stages can be discriminated manually by specialists. However, this can be a laboriousness work because of the huge datasets. Until now, several studies have been conducted based on the automatic analysis of electroencephalography. Still, as the development of wearable technology, there is a need for an accurate and single-channel electroencephalography based sleep stages identification system. In this paper, a statespace based sleep stages classification method is proposed using the proposed model based essence features extraction method. This method employed the state-space model to establish the intrinsic models based on the single-channel electroencephalography, from which the features used for further classification can be extracted. For 2-stage to 6-stage classification of sleep states, the verification system can achieve 98.6%, 94.9%, 93.0%, 92.3%, 91.8% accuracy on the Sleep-EDF database, and also reach 94.9%, 87.7%, 82.7%, 80.9%, 78.2% on Dreams Subjects database.

show abstract

Section: Introductionmentioning

confidence: 99%

An Accurate Sleep Stages Classification Method Based on State Space Model

Shen

Guez

et al. 2019

IEEE Access

View full text Add to dashboard Cite

show abstract

“…The TP rate, FP rate, Prc, F and MCC value were obtained using equations [41, equations (1)-( 5)]. More information can be found in some of recent studies [42][43][44].…”

Section: Classification Performancementioning

confidence: 99%

Optimization of an artificial neural network to estimate laser ablation efficiency

Yildiz¹

2019

Laser Phys.

View full text Add to dashboard Cite

The use of fiber laser in surgery has many advantages due to its small size, efficiency, operation in continuous and pulse mode and easy coupling to fiber optics for various applications. Laser surgery requires very precise control on laser parameters, and imprecision can cause unacceptable damages during operation, particularly in neurosurgical applications. There have been many experimental studies looking to optimize laser parameters to control laser thermal therapy. In this study, we propose an artificial neural network method (ANN) to predict laser ablation damage as a function of laser parameters and temperature. The purpose of this study was to investigate the performance of artificial neural networks to predict the ablation efficiency of a 1940 nm thulium fiber laser on ovine brain tissue. Twelve experimental data were used to train an ANN model. Tissue type (cortical/subcortical), laser mode, laser power, laser energy, time, temperature and temperature change were used as inputs for the ANN and the ablation efficiency (ablated area/total thermally altered area) was the output of the model. Four untrained data were used to validate the ANN prediction ability after finding correlation between the ANN inputs and output. Experimental and predicted data were compared to find the accuracy of the model. Moreover, optimum laser mode (continuous/pulse) selection was also studied. Five different machine learning methods were used for laser mode selection, and the results were compared. Our results showed that prediction for the ablation efficiency of an ANN is lower than 15% and 87.5% classification accuracy was obtained for optimum laser mode selection.

show abstract

“…The TQWT is also applied to decompose an EEG signal into various sub-bands at different levels; the findings showed that the proposed scheme with estimating the Hjorth parameters preserves efficiency and is appropriate for the automated identification of EEG signals (Geetika et al, 2022 ). Some time and frequency analysis methods based on variational mode decomposition were utilized to determine the k-complex, and the highest average accuracy was obtained at 92.29% (Yücelbaş et al, 2017 ). Wessam proposed an efficient method based on fractal dimension to detect k-complexes from EEG signals, and the findings revealed that the proposed method yields better classification results than other existing methods (Al-Salman et al, 2019b ).…”

Section: Introductionmentioning

confidence: 99%

A RUSBoosted tree method for k-complex detection using tunable Q-factor wavelet transform and multi-domain feature extraction

Dong²

2023

Front. Neurosci.

View full text Add to dashboard Cite

BackgroundK-complex detection traditionally relied on expert clinicians, which is time-consuming and onerous. Various automatic k-complex detection-based machine learning methods are presented. However, these methods always suffered from imbalanced datasets, which impede the subsequent processing steps.New methodIn this study, an efficient method for k-complex detection using electroencephalogram (EEG)-based multi-domain features extraction and selection method coupled with a RUSBoosted tree model is presented. EEG signals are first decomposed using a tunable Q-factor wavelet transform (TQWT). Then, multi-domain features based on TQWT are pulled out from TQWT sub-bands, and a self-adaptive feature set is obtained from a feature selection based on the consistency-based filter for the detection of k-complexes. Finally, the RUSBoosted tree model is used to perform k-complex detection.ResultsExperimental outcomes manifest the efficacy of our proposed scheme in terms of the average performance of recall measure, AUC, and F10-score. The proposed method yields 92.41 ± 7.47%, 95.4 ± 4.32%, and 83.13 ± 8.59% for k-complex detection in Scenario 1 and also achieves similar results in Scenario 2.Comparison to state-of-the-art methodsThe RUSBoosted tree model was compared with three other machine learning classifiers [i.e., linear discriminant analysis (LDA), logistic regression, and linear support vector machine (SVM)]. The performance based on the kappa coefficient, recall measure, and F10-score provided evidence that the proposed model surpassed other algorithms in the detection of the k-complexes, especially for the recall measure.ConclusionIn summary, the RUSBoosted tree model presents a promising performance in dealing with highly imbalanced data. It can be an effective tool for doctors and neurologists to diagnose and treat sleep disorders.

show abstract

RETRACTED ARTICLE: A novel system for automatic detection of K-complexes in sleep EEG

Cited by 22 publications

References 29 publications

An Accurate Sleep Stages Classification Method Based on State Space Model

An Accurate Sleep Stages Classification Method Based on State Space Model

Optimization of an artificial neural network to estimate laser ablation efficiency

A RUSBoosted tree method for k-complex detection using tunable Q-factor wavelet transform and multi-domain feature extraction

Contact Info

Product

Resources

About