Enhancing sleep stage classification with ballistocardiogram signals: feature selection using attention mechanism and XGBoost
Traditional sleep staging using contact sensors may compromise data validity. This study proposes a non-contact ballistocardiogram (BCG)-based method to improve both accuracy and comfort in sleep monitoring. BCG signals were processed using continuous wavelet transform and low-pass filtering to extr...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2025-07-01
|
| Series: | Frontiers in Public Health |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fpubh.2025.1608725/full |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1839612671585943552 |
|---|---|
| author | Chao Luo Banteng Liu Jiayu Chai Zhijian Teng |
| author_facet | Chao Luo Banteng Liu Jiayu Chai Zhijian Teng |
| author_sort | Chao Luo |
| collection | DOAJ |
| description | Traditional sleep staging using contact sensors may compromise data validity. This study proposes a non-contact ballistocardiogram (BCG)-based method to improve both accuracy and comfort in sleep monitoring. BCG signals were processed using continuous wavelet transform and low-pass filtering to extract heart rate variability (HRV) and respiratory rate variability (RRV). A novel feature selection model integrating attention mechanisms with XGBoost was developed, where attention weights are used to prioritize features before iterative refinement by XGBoost. Evaluated on 10,201 sleep segments, the Fast-ABC Boost model achieved an accuracy of 89.85%, along with superior precision, recall, F1-score, and Kappa values compared to conventional methods. The attention-XGBoost fusion effectively mitigates interference from noisy and redundant features while optimizing feature relevance, demonstrating robust adaptability to the complexity of sleep signals. This innovation advances the accuracy non-contact sleep staging, enabling practical applications in home healthcare and personalized sleep management, while improving user comfort. |
| format | Article |
| id | doaj-art-bada16a41a7f48a5988bd59e6dc828a8 |
| institution | Matheson Library |
| issn | 2296-2565 |
| language | English |
| publishDate | 2025-07-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Public Health |
| spelling | doaj-art-bada16a41a7f48a5988bd59e6dc828a82025-07-28T05:30:25ZengFrontiers Media S.A.Frontiers in Public Health2296-25652025-07-011310.3389/fpubh.2025.16087251608725Enhancing sleep stage classification with ballistocardiogram signals: feature selection using attention mechanism and XGBoostChao Luo0Banteng Liu1Jiayu Chai2Zhijian Teng3School of Information Engineering, Huzhou University, Huzhou, ChinaKey Laboratory of Artificial Organs, Computational Medicine in Zhejiang Province, College of Information Science and Technology, Zhejiang Shuren University, Hangzhou, ChinaKey Laboratory of Artificial Organs, Computational Medicine in Zhejiang Province, College of Information Science and Technology, Zhejiang Shuren University, Hangzhou, ChinaSchool of Information Engineering, Huzhou University, Huzhou, ChinaTraditional sleep staging using contact sensors may compromise data validity. This study proposes a non-contact ballistocardiogram (BCG)-based method to improve both accuracy and comfort in sleep monitoring. BCG signals were processed using continuous wavelet transform and low-pass filtering to extract heart rate variability (HRV) and respiratory rate variability (RRV). A novel feature selection model integrating attention mechanisms with XGBoost was developed, where attention weights are used to prioritize features before iterative refinement by XGBoost. Evaluated on 10,201 sleep segments, the Fast-ABC Boost model achieved an accuracy of 89.85%, along with superior precision, recall, F1-score, and Kappa values compared to conventional methods. The attention-XGBoost fusion effectively mitigates interference from noisy and redundant features while optimizing feature relevance, demonstrating robust adaptability to the complexity of sleep signals. This innovation advances the accuracy non-contact sleep staging, enabling practical applications in home healthcare and personalized sleep management, while improving user comfort.https://www.frontiersin.org/articles/10.3389/fpubh.2025.1608725/fullsleep stagingballistocardiogram (BCG)heart rate variability (HRV)respiratory rate variability (RRV)XGBoostfeature selection |
| spellingShingle | Chao Luo Banteng Liu Jiayu Chai Zhijian Teng Enhancing sleep stage classification with ballistocardiogram signals: feature selection using attention mechanism and XGBoost Frontiers in Public Health sleep staging ballistocardiogram (BCG) heart rate variability (HRV) respiratory rate variability (RRV) XGBoost feature selection |
| title | Enhancing sleep stage classification with ballistocardiogram signals: feature selection using attention mechanism and XGBoost |
| title_full | Enhancing sleep stage classification with ballistocardiogram signals: feature selection using attention mechanism and XGBoost |
| title_fullStr | Enhancing sleep stage classification with ballistocardiogram signals: feature selection using attention mechanism and XGBoost |
| title_full_unstemmed | Enhancing sleep stage classification with ballistocardiogram signals: feature selection using attention mechanism and XGBoost |
| title_short | Enhancing sleep stage classification with ballistocardiogram signals: feature selection using attention mechanism and XGBoost |
| title_sort | enhancing sleep stage classification with ballistocardiogram signals feature selection using attention mechanism and xgboost |
| topic | sleep staging ballistocardiogram (BCG) heart rate variability (HRV) respiratory rate variability (RRV) XGBoost feature selection |
| url | https://www.frontiersin.org/articles/10.3389/fpubh.2025.1608725/full |
| work_keys_str_mv | AT chaoluo enhancingsleepstageclassificationwithballistocardiogramsignalsfeatureselectionusingattentionmechanismandxgboost AT bantengliu enhancingsleepstageclassificationwithballistocardiogramsignalsfeatureselectionusingattentionmechanismandxgboost AT jiayuchai enhancingsleepstageclassificationwithballistocardiogramsignalsfeatureselectionusingattentionmechanismandxgboost AT zhijianteng enhancingsleepstageclassificationwithballistocardiogramsignalsfeatureselectionusingattentionmechanismandxgboost |