1. School of Information Engineering, Huzhou University, Huzhou 313000, China 2. College of Information Science and Technology, Zhejiang Shuren University, Hangzhou 310015, China 3. State Key Laboratory of Industrial Control Technology, Zhejiang University, Hangzhou 310027, China
A simple and efficient sleep analysis algorithm was designed based on single-channel electrocardiogram (ECG) signals, in order to reduce the dependence on polysomnography (PSG) system. First, maximum overlap discrete wavelet transform（MODWT）was used to perform multi-resolution analysis on the original signal, then to furture extract peak information. Then, the multi-dimensional heart rate variability（HRV）features were extracted based on the first-order deviation of the peak position. To further screen the HRV features with a strong correlation with different sleep stages, a feature extraction method was proposed based on the ReliefF algorithm and Gini index. On this basis, the INFO-ABCLogitBoost method was used to mine the correlation between HRV and different sleep stages, thereby achieving a fine classification of sleep stages. Experimental results on actual public data sets showed that the proposed model had an overall accuracy of 83.67%, an accuracy rate of 82.59%, a Kappa coefficient of 77.94%, and an F1-Score value of 82.97% in the sleep staging task. Compared with conventional models in sleep staging tasks, the proposed method shows more efficient and convenient sleep quality assessment performance, which helps realize sleep monitoring in home or mobile medical scenarios.
Bingyang ZHU,Jianfeng WU,Ke WANG,Zhangquan WANG,Banteng LIU. Sleep staging based on single-channel ECG signal and INFO-ABCLogitBoost model. Journal of ZheJiang University (Engineering Science), 2024, 58(12): 2547-2555.
Tab.3HRV feature statistics extracted from sleep stage task
Fig.4Mean and standard deviation of importance of HRV features calculated based on Gini index
Fig.5Mean and standard deviation of HRV feature weights calculated based on ReliefF algorithm
算法
n/个
t/min
ACC/%
Kappa/%
F1/%
—
77
632.95
82.43
76.22
81.39
Gini
52
379.06
83.35
77.51
82.66
ReliefF
38
261.83
83.67
77.94
82.97
Tab.4Experimental results of feature dimensionality reduction
方法
ACC/%
PRE/%
Kappa/%
F1/%
ELM
65.76
65.42
53.81
65.37
SVM
69.89
68.40
58.43
67.98
GRU
70.71
68.83
59.84
69.05
LSTM
70.64
68.85
59.66
68.91
BiLSTM
71.49
69.77
61.03
70.06
CNN
72.26
71.97
62.74
72.08
CNN-LSTM
72.05
71.45
62.39
71.65
ResNet
73.54
73.28
64.48
73.34
XGBoost
70.71
68.83
59.84
69.05
Bag
79.13
77.89
71.65
78.19
RF
80.37
79.10
73.38
79.45
GBDT
80.73
79.43
73.90
79.82
本研究
83.67
82.59
77.94
82.97
Tab.5Test results of performance of sleep stage classification model
方法
Recall/%
NREM1
NREM2
NREM3
REM
W
ELM
18.77
74.59
61.70
63.31
71.67
SVM
11.14
80.66
62.91
55.40
83.55
GRU
12.02
78.73
61.01
64.39
85.07
LSTM
12.32
78.52
61.70
61.15
86.12
BiLSTM
14.96
77.87
64.12
65.47
85.87
CNN
25.81
76.16
72.96
67.81
82.26
CNN-LSTM
24.68
77.05
68.01
65.33
84.71
ResNet
26.41
77.48
71.28
70.67
84.71
XGBoost
12.02
78.73
61.01
64.39
85.07
Bag
24.63
85.08
76.78
75.90
89.89
RF
24.93
85.87
79.38
78.78
90.53
GBDT
25.22
85.94
81.80
79.50
90.13
本研究
29.03
89.08
86.31
83.27
91.49
Tab.6Recall rates of models for different sleep stages in sleep staging task
数据类别
ACC/%
PRE/%
Kappa/%
F1/%
No pathology
86.78
86.28
81.10
86.19
NFLE
77.94
77.76
68.21
77.31
RBD
88.07
88.16
83.52
87.98
PLM
85.78
85.89
79.71
85.40
SDB
74.63
73.97
58.32
73.09
Insomnia
87.98
88.09
82.46
87.72
Narcolepsy
84.22
83.90
79.11
83.76
Tab.7Validation results of samples from different categories
文献
方法
睡眠阶段
ACC/%
Kappa/%
F1/%
[8]
GRU
3
80.43
—
80.07
[9]
LSTM
4
77.00
61.00
—
[10]
BiLSTM
4
75.90
60.00
—
[11]
CNN
4
91.72
88.44
88.50
[12]
CNN- LSTM
5
72.54
58.50
70.10
[13]
GBDT
5
82.02
—
72.88
[14]
ResNet
5 4
77.34 85.32
67.58 77.11
68.03 81.87
[15]
RF
5
75.90
66.30
76.00
[16]
PSO- ELM
6 4
62.66 71.52
— —
— —
本研究
INFO- ABC- LogitBoost
5
83.67
77.94
82.97
Tab.8Comparison of advanced research results with results of proposed model
[1]
PANDA S, HOGENESCH J B, KAY S A Circadian rhythms from flies to human[J]. Nature, 2002, 417 (6886): 329- 335
doi: 10.1038/417329a
[2]
SHOCHAT T Impact of lifestyle and technology developments on sleep[J]. Nature and Science of Sleep, 2012, 4 (6): 19- 31
[3]
KILLICK R, STRANKS L, HOYOS C M Sleep deficiency and cardiometabolic disease[J]. Clinics in Chest Medicine, 2022, 43 (2): 319- 336
doi: 10.1016/j.ccm.2022.02.011
[4]
BERRY R B, BROOKS R, GAMALDO C, et al AASM scoring manual updates for 2017 (version 2.4)[J]. Journal of Clinical Sleep Medicine, 2017, 13 (5): 665- 666
doi: 10.5664/jcsm.6576
[5]
MADOR M J, KUFEL T J, MAGALANG U J, et al Prevalence of positional sleep apnea in patients undergoing polysomnography[J]. Chest, 2005, 128 (4): 2130- 2137
doi: 10.1378/chest.128.4.2130
[6]
GULLETT N, ZAJKOWSKA Z, WALSH A, et al Heart rate variability (HRV) as a way to understand associations between the autonomic nervous system (ANS) and affective states: a critical review of the literature[J]. International Journal of Psychophysiology, 2023, 192 (5): 35- 42
[7]
GONZALES J U, ELAVSKY S, CIPRYAN L, et al Influence of sleep duration and sex on age-related differences in heart rate variability: findings from program 4 of the HAIE study[J]. Sleep Medicine, 2023, 106 (14): 69- 77
[8]
ERDENEBAYAR U, KIM Y, PARK J U, et al Automatic classification of sleep stage from an ecg signal using a gated-recurrent unit[J]. International Journal of Fuzzy Logic and Intelligent Systems, 2020, 20 (3): 181- 187
doi: 10.5391/IJFIS.2020.20.3.181
[9]
RADHA M, FONSECA P, MOREAU A, et al Sleep stage classification from heart-rate variability using long short-term memory neural networks[J]. Scientific Reports, 2019, 9 (1): 14149
doi: 10.1038/s41598-019-49703-y
[10]
FONSECA P, VAN GILST M M, RADHA M, et al Automatic sleep staging using heart rate variability, body movements, and recurrent neural networks in a sleep disordered population[J]. Sleep, 2020, 43 (9): 48
[11]
GENG D Y, WANG J X, WANG Y, et al Convolutional neural network is a good technique for sleep staging based on HRV: a comparative analysis[J]. Neuroscience Letters, 2022, 779: 136550
doi: 10.1016/j.neulet.2022.136550
[12]
SUN H, GANGLBERGER W, PANNEERSELVAM E, et al Sleep staging from electrocardiography and respiration with deep learning[J]. Sleep, 2020, 43 (7): 306
doi: 10.1093/sleep/zsz306
[13]
WANG W B, QIN D, FANG Y, et al Automatic multi-class sleep staging method based on novel hybrid features[J]. Journal of Electrical Engineering and Technology, 2024, 19 (1): 709- 722
doi: 10.1007/s42835-023-01570-4
[14]
MATHUNJWA B M, LIN Y T, LIN C H, et al Automatic IHR-based sleep stage detection using features of residual neural network[J]. Biomedical Signal Processing and Control, 2023, 85: 105070
doi: 10.1016/j.bspc.2023.105070
[15]
CHAKRABORTY S, GOYAL M, GOYAL P, et al. A machine learning approach for automatic sleep staging using heart rate variability [EB/OL]. (2022-05-13)[2023-08-08]. https://www.biorxiv.org/content/10.1101/2022.05.13.491872v2.full.pdf.
[16]
SURANTHA N, LESMANA T F, ISA S M Sleep stage classification using extreme learning machine and particle swarm optimization for healthcare big data[J]. Journal of Big Data, 2021, 8 (1): 1- 17
doi: 10.1186/s40537-020-00387-6
[17]
彭向东, 潘从成, 柯泽浚, 等 基于并行架构和时空注意力机制的心电分类方法[J]. 浙江大学学报: 工学版, 2022, 56 (10): 1912- 1923 PENG Xiangdong, PAN Congcheng, KE Zejun, et al Classification method for electrocardiograph signals based on parallel architecture model and spatial-temporal attention mechanism[J]. Journal of Zhejiang University: Engineering Science, 2022, 56 (10): 1912- 1923
[18]
CHANDRA S, SHARMA A, SINGH G K Feature extraction of ECG signal[J]. Journal of Medical Engineering and Technology, 2018, 42 (4): 306- 316
doi: 10.1080/03091902.2018.1492039
MENG F, ZHAO D, ZHANG X A fair consensus adjustment mechanism for large-scale group decision making in term of Gini coefficient[J]. Engineering Applications of Artificial Intelligence, 2023, 126 (23): 106962
[21]
ZHANG B, LI Y, CHAI Z A novel random multi-subspace based ReliefF for feature selection[J]. Knowledge-based Systems, 2022, 252 (12): 109400
[22]
LI P. Abc-boost: adaptive base class boost for multi-class classification [C] // Proceedings of the 26th Annual international conference on Machine Learning . New York: Association for Computing Machinery, 2009: 625-632.
[23]
PONOMAREVA N, COLTHURST T, HENDRY G, et al. Compact multi-class boosted trees [C] // 2017 IEEE International Conference on Big Data . Boston: IEEE, 2017: 47-56.
