AUTHOR=Li Yabing , Dong Xinglong 

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

JOURNAL=Frontiers in Neuroscience

VOLUME=Volume 17 - 2023

YEAR=2023

URL=https://www.frontiersin.org/journals/neuroscience/articles/10.3389/fnins.2023.1108059

DOI=10.3389/fnins.2023.1108059

ISSN=1662-453X

ABSTRACT=Background: K-complexes detection traditionally relied on expert clinicians, which is time-consuming and onerous. Lots of automatic k-complexes detection-based machine learning are presented. However, these methods always suffered from imbalanced datasets, which impede the subsequent processing steps.
New method: In this study, an efficient method for k-complexes detection using electroencephalogram (EEG) based multi-domain features extraction and selection method coupled with a RUSBoosted trees model is presented. EEG signals are first decomposed using tunable-Q factor wavelet transform (TQWT). And then multi-domain features based on TQWT are pulled out from each sub-band and then forwarded to a feature selection based on consistency-based filter to select features for detection of k-complexes. Finally, an adaptive-feature set is used to detect the k-complexes characteristics using the RUSBoosted trees model.
Results: The proposed method for the detection of the k-complexes achieves an average performance of recall measure, AUC, and F_10-score of 82.42%±22.14%, 95.4%±4.32%, and 81.19%±21.32%.
Comparison to state-of-the-art methods: The RUSBoosted trees 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 Kappa coefficient, recall measure, and F_10-score provided the evidence that the proposed model surpassed other algorithms in the detection of the k-complexes. 
Conclusions: In summary, the RUSBoosted Trees 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.