Evaluate different machine learning techniques for classifying sleep stages on single-channel EEG

Qureshi, Shahnawaz; Vanichayobon, Sirirut

doi:10.1109/jcsse.2017.8025949

Cited by 10 publications

(6 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The multilayer perceptron (MLP) and decision tree (DT) have their best performances and confusion matrices shown in Table 6-13. Furthermore, our previous study [25] is based on different machine learning techniques to classify the sleep stages.…”

Section: Results and Discussion For Chebyshev Distance With K =mentioning

confidence: 99%

Human Sleep Scoring Based on K-Nearest Neighbors

Qureshi¹,

Karrila²,

Vanichayobon³

2018

Turk J Elec Eng & Comp Sci

Self Cite

View full text Add to dashboard Cite

Human sleep is one of the essential indicators that gauge the overall health and well-being. Presently, it is common for people to face issues related to sleep. Various biomedical signals including electroencephalogram (EEG), electrooculography (EMG), and electrooculography (EOG) are utilized in the diagnosis and during the treatment of sleep disorder cases. An automatic classification to diagnose sleep problems can help in the analysis of sleep EEG data. In this current study, an effort is made to classify the sleep stages from a single EEG channel (C4-A1) based on K-nearest neighbors (K-NN) with three alternative distance metrics. The Euclidean distance is the most commonly used distance measure in K-NN, and no prior study of sleep EEG data has inspected the classification performance of K-NN with various distance measures. Therefore, this study aimed to investigate whether the distance function affects the performance of K-NN in the classification of sleep data. Euclidean, Manhattan and Chebyshev distance measures were individually tested with K-NN classification, and their performances were compared based on accuracy, sensitivity, specificity, F-measure, Kappa statistic and computation time for both Rechtschaffen & Kales and American Academy of Sleep Medicine standard labelings of the sleep stages. The experimental results show that the Manhattan distance function with K = 5 was the best choice for classification of the sleep stages, achieving 98.46% and 98.77% correct rates for the two labelings with comparatively rapid computations.

show abstract

Section: Results and Discussion For Chebyshev Distance With K =mentioning

confidence: 99%

Human Sleep Scoring Based on K-Nearest Neighbors

Qureshi¹,

Karrila²,

Vanichayobon³

2018

Turk J Elec Eng & Comp Sci

Self Cite

View full text Add to dashboard Cite

show abstract

“…In recent years, thanks to the blossoming of artificial intelligence and big data in this era and the dramatic evolution of microelectronics as well, EEG applications have expanded from research-oriented tools to more practical use. EEG has become one of the main evaluation tools of brain disease including sleep disorders [1][2][3][4][5][6] and epileptic seizures [7][8][9][10][11] clinically, potential applications in stroke recovery [12][13][14][15] and head trauma. [16][17][18][19] To date, developers have further extended EEG applications' reach outside of medical use to other fields such as sports training and condition monitoring for athletes, [20][21][22] robot controls, [23][24][25][26] evaluation of driver vigilance.…”

Section: Overview Of Eegmentioning

confidence: 99%

“…We will then report the developed wearable EEGs that target the most widely applied brain-related diseases and disorders. For example, BCIs, 29–31 epilepsy diagnosis, 7–11,32,33 sleep disorder diagnosis, 1–6 and mental health evaluation. 34,35…”

Section: Introductionmentioning

confidence: 99%

Design of hydrogel-based wearable EEG electrodes for medical applications

Hsieh

Wang

et al. 2022

J. Mater. Chem. B

View full text Add to dashboard Cite

show abstract

“…Deep learning was selected for this classification problem because physiological responses to sleep onset and REM sleep involve complicated processes that are best elucidated through advanced feature detection methods (30,31). Specifically, an FFNN with one hidden layer consisting of 120 neurons was implemented because this model allows for sufficient variable complexity but limits very-high-order interactions that may be subject dependent or a result of overfitting (31). Neuron weights and biases were trained with conjugate gradient backpropagation on a set of 262 input features after min-max normalization.…”

Section: Sleep Analysis and Classification Of Sleep Stages With A Neu...mentioning

confidence: 99%