Automated and precise heartbeat detection in ballistocardiography signals using bidirectional LSTM
Ballistocardiography (BCG) provides an unobtrusive, mechanical measurement of heartbeat for early detection and longitudinal monitoring of cardiovascular conditions, although it is plagued by poor signal to noise ratio. Despite tremendous efforts over the last decade in related advanced signal proce...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2022-08-01
|
| Series: | Franklin Open |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2773186322000019 |
| _version_ | 1827876668550676480 |
|---|---|
| author | Shiyu Zhang Haihong Zhang Zhiping Lin Soon Huat Ng |
| author_facet | Shiyu Zhang Haihong Zhang Zhiping Lin Soon Huat Ng |
| author_sort | Shiyu Zhang |
| collection | DOAJ |
| description | Ballistocardiography (BCG) provides an unobtrusive, mechanical measurement of heartbeat for early detection and longitudinal monitoring of cardiovascular conditions, although it is plagued by poor signal to noise ratio. Despite tremendous efforts over the last decade in related advanced signal processing and deep learning research, automated and precise heartbeat detection in BCG remains a major technical challenge. In this work, we design and study a sequential deep neural network based approach that addresses three key issues in automated BCG learning and detection: auto-labeling of BCG samples in the raw training data; design of a recursive neural network that can learn the association between the continuous BCG waves and the label sequence in the form of sample-by-sample predictions; and finally, heartbeat detection from the output sequence. We evaluate the proposed method using our data set comprising 8 human subjects under varying cardiac output conditions: pre-exercise and post-exercise. Using multiple performance metrics, we report that the proposed method can achieve a heart rate accuracy of 98.68% and root-mean-square-error of 1.37 bpm with 98.85% coverage, which compare favorably to two state-of-the-art methods in the same test. |
| first_indexed | 2024-03-12T17:24:38Z |
| format | Article |
| id | doaj.art-96dca0896e5e47f5b4d698d5e657cecc |
| institution | Directory Open Access Journal |
| issn | 2773-1863 |
| language | English |
| last_indexed | 2024-03-12T17:24:38Z |
| publishDate | 2022-08-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Franklin Open |
| spelling | doaj.art-96dca0896e5e47f5b4d698d5e657cecc2023-08-05T05:18:28ZengElsevierFranklin Open2773-18632022-08-0113038Automated and precise heartbeat detection in ballistocardiography signals using bidirectional LSTMShiyu Zhang0Haihong Zhang1Zhiping Lin2Soon Huat Ng3School of Electrical and Electronic Engineering, Nanyang Technological University, SingaporeInstitute for Infocomm Research, Agency for Science, Technology and Research, SingaporeSchool of Electrical and Electronic Engineering, Nanyang Technological University, Singapore; Corresponding author.Institute for Infocomm Research, Agency for Science, Technology and Research, SingaporeBallistocardiography (BCG) provides an unobtrusive, mechanical measurement of heartbeat for early detection and longitudinal monitoring of cardiovascular conditions, although it is plagued by poor signal to noise ratio. Despite tremendous efforts over the last decade in related advanced signal processing and deep learning research, automated and precise heartbeat detection in BCG remains a major technical challenge. In this work, we design and study a sequential deep neural network based approach that addresses three key issues in automated BCG learning and detection: auto-labeling of BCG samples in the raw training data; design of a recursive neural network that can learn the association between the continuous BCG waves and the label sequence in the form of sample-by-sample predictions; and finally, heartbeat detection from the output sequence. We evaluate the proposed method using our data set comprising 8 human subjects under varying cardiac output conditions: pre-exercise and post-exercise. Using multiple performance metrics, we report that the proposed method can achieve a heart rate accuracy of 98.68% and root-mean-square-error of 1.37 bpm with 98.85% coverage, which compare favorably to two state-of-the-art methods in the same test.http://www.sciencedirect.com/science/article/pii/S2773186322000019BallistocardiographyRecursive deep learningHeartbeat detection |
| spellingShingle | Shiyu Zhang Haihong Zhang Zhiping Lin Soon Huat Ng Automated and precise heartbeat detection in ballistocardiography signals using bidirectional LSTM Franklin Open Ballistocardiography Recursive deep learning Heartbeat detection |
| title | Automated and precise heartbeat detection in ballistocardiography signals using bidirectional LSTM |
| title_full | Automated and precise heartbeat detection in ballistocardiography signals using bidirectional LSTM |
| title_fullStr | Automated and precise heartbeat detection in ballistocardiography signals using bidirectional LSTM |
| title_full_unstemmed | Automated and precise heartbeat detection in ballistocardiography signals using bidirectional LSTM |
| title_short | Automated and precise heartbeat detection in ballistocardiography signals using bidirectional LSTM |
| title_sort | automated and precise heartbeat detection in ballistocardiography signals using bidirectional lstm |
| topic | Ballistocardiography Recursive deep learning Heartbeat detection |
| url | http://www.sciencedirect.com/science/article/pii/S2773186322000019 |
| work_keys_str_mv | AT shiyuzhang automatedandpreciseheartbeatdetectioninballistocardiographysignalsusingbidirectionallstm AT haihongzhang automatedandpreciseheartbeatdetectioninballistocardiographysignalsusingbidirectionallstm AT zhipinglin automatedandpreciseheartbeatdetectioninballistocardiographysignalsusingbidirectionallstm AT soonhuatng automatedandpreciseheartbeatdetectioninballistocardiographysignalsusingbidirectionallstm |