Performance Evaluation of Processing Methods for Ballistocardiogram Peak Detection

Suliman, Ahmad; Carlson, Charles; Warren, Steve; Thompson, David E.

doi:10.1109/embc.2018.8512317

Cited by 5 publications

(8 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The novel sensor technologies may detect HR and HRV more accurately. In addition, several different algorithms are being used to detect BCG peaks, each with differing detecting abilities [11]. However, previous studies have been conducted under laboratory conditions.…”

Section: Introductionmentioning

confidence: 99%

A Contact-Free, Ballistocardiography-Based Monitoring System (Emfit QS) for Measuring Nocturnal Heart Rate and Heart Rate Variability: Validation Study

Ville¹

2020

JMIR Biomed Eng

View full text Add to dashboard Cite

Background Heart rate (HR) and heart rate variability (HRV) measurements are widely used to monitor stress and recovery status in sedentary people and athletes. However, effective HRV monitoring should occur on a daily basis because sparse measurements do not allow for a complete view of the stress-recovery balance. Morning electrocardiography (ECG) measurements with HR straps are time-consuming and arduous to perform every day, and thus compliance with regular measurements is poor. Contact-free, ballistocardiography (BCG)-based Emfit QS is effortless for daily monitoring. However, to the best of our knowledge, there is no study on the accuracy of nocturnal HR and HRV measured via BCG under real-life conditions. Objective The aim of this study was to evaluate the accuracy of Emfit QS in measuring nocturnal HR and HRV. Methods Healthy participants (n=20) completed nocturnal HR and HRV recordings at home using Emfit QS and an ECG-based reference device (Firstbeat BG2) during sleep. Emfit QS measures BCG by a ferroelectret sensor installed under a bed mattress. HR and the root mean square of successive differences between RR intervals (RMSSD) were determined for 3-minute epochs and the sleep period mean. Results A trivial mean bias was observed in the mean HR (mean –0.8 bpm [beats per minute], SD 2.3 bpm, P=.15) and Ln (natural logarithm) RMSSD (mean –0.05 ms, SD 0.25 ms, P=.33) between Emfit QS and ECG. In addition, very large correlations were found in the mean values of HR (r=0.90, P<.001) and Ln RMSSD (r=0.89, P<.001) between the devices. A greater amount of erroneous or missing data (P<.001) was observed in the Emfit QS measurements (28.3%, SD 14.4%) compared with the reference device (1.1%, SD 2.3%). The results showed that 5.0% of the mean HR and Ln RMSSD values were outside the limits of agreement. Conclusions Based on the present results, Emfit QS provides nocturnal HR and HRV data with an acceptable, small mean bias when calculating the mean of the sleep period. Thus, Emfit QS has the potential to be used for the long-term monitoring of nocturnal HR and HRV. However, further research is needed to assess reliability in HR and HRV detection.

show abstract

Section: Introductionmentioning

confidence: 99%

A Contact-Free, Ballistocardiography-Based Monitoring System (Emfit QS) for Measuring Nocturnal Heart Rate and Heart Rate Variability: Validation Study

Ville¹

2020

JMIR Biomed Eng

View full text Add to dashboard Cite

show abstract

“…The latter was chosen but led to two MA length parameters that required optimization. Contrary to the initial study [33], performance was substantially improved by setting this second parameter in a subject-specific manner.…”

Section: A Lee Methodsmentioning

confidence: 90%

“…The original codes from the mentioned studies were modified to address differences in sensing methods and sampling frequencies. The authors recently conducted a pilot study involving five participants in [33] to compare three peak detection methods [20], [24], [31]. The present study extends that work to include the performance of two additional methods [25], [27], where data from total of 30 participants supplement the original data in [33].…”

Section: Introductionmentioning

confidence: 83%

“…Any temporal shift introduced by a signal processing step was corrected so that the detected peaks in the BCG segment would align in time with their corresponding GT peaks. A detection, or true positive (TP), is defined as a detected peak that is within d = 0.06 sec of the GT peak [33]. WhileSprager and Zazula [14] suggest 0.075 sec instead, this stricter criteria improves specificity and exceeds the IEC standard for ECG QRS detection used in [14].…”

Section: Performance Metricsmentioning

confidence: 99%

“…The authors recently conducted a pilot study involving five participants in [33] to compare three peak detection methods [20], [24], [31]. The present study extends that work to include the performance of two additional methods [25], [27], where data from total of 30 participants supplement the original data in [33]. For clarity in this paper, each of these five methods is referred to by the last name of the corresponding first author: Lee et al [24], Lydon et al [20], Brüser et al [31], Alvarado-Serrano et al [25] and Sadek et al [27].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Performance Comparison for Ballistocardiogram Peak Detection Methods

et al. 2019

Self Cite

View full text Add to dashboard Cite

A number of research groups have proposed methods for ballistocardiogram (BCG) peak detection toward the identification of individual cardiac cycles. However, objective comparisons of these proposed methods are lacking. This paper, therefore, conducts a systematic and objective performance evaluation and comparison of several of these approaches. Five peak-detection methods (three replicated from the literature and two adapted from code provided by the methods' authors) are compared using data from 30 volunteers. A basic cross-correlation approach was also included as a sixth method. Two high-performing methods were identified: the method proposed by Sadek et al. and the method proposed by Brüser et al. The first achieved the highest average peak-detection rate of 94%, the lowest average false alarm rate of 0.0552 false alarms per second, and a relatively small mean absolute error between the real and detected peaks: 0.0175 seconds. The second method achieved the lowest mean absolute error of 0.0088 seconds between the real and detected peaks, an average peak-detection success rate of 89%, and 0.0766 false alarms per second. All metrics are averaged across participants. INDEX TERMS Ballistocardiogram, heartbeat, heartbeat interval, heart rate, heart rate variability, load cells, statistical signal analysis, wavelets.

show abstract

A Contact-Free, Ballistocardiography-Based Monitoring System (Emfit QS) for Measuring Nocturnal Heart Rate and Heart Rate Variability: Validation Study (Preprint)

Vesterinen¹,

Rinkinen²,

Nummela³

2019

Preprint

View full text Add to dashboard Cite

BACKGROUND Heart rate (HR) and heart rate variability (HRV) measurements are widely used to monitor stress and recovery status in sedentary people and athletes. However, effective HRV monitoring should occur on a daily basis because sparse measurements do not allow for a complete view of the stress-recovery balance. Morning electrocardiography (ECG) measurements with HR straps are time-consuming and arduous to perform every day, and thus compliance with regular measurements is poor. Contact-free, ballistocardiography (BCG)-based Emfit QS is effortless for daily monitoring. However, to the best of our knowledge, there is no study on the accuracy of nocturnal HR and HRV measured via BCG under real-life conditions. OBJECTIVE The aim of this study was to evaluate the accuracy of Emfit QS in measuring nocturnal HR and HRV. METHODS Healthy participants (n=20) completed nocturnal HR and HRV recordings at home using Emfit QS and an ECG-based reference device (Firstbeat BG2) during sleep. Emfit QS measures BCG by a ferroelectret sensor installed under a bed mattress. HR and the root mean square of successive differences between RR intervals (RMSSD) were determined for 3-minute epochs and the sleep period mean. RESULTS A trivial mean bias was observed in the mean HR (mean –0.8 bpm [beats per minute], SD 2.3 bpm, P=.15) and Ln (natural logarithm) RMSSD (mean –0.05 ms, SD 0.25 ms, P=.33) between Emfit QS and ECG. In addition, very large correlations were found in the mean values of HR (r=0.90, P<.001) and Ln RMSSD (r=0.89, P<.001) between the devices. A greater amount of erroneous or missing data (P<.001) was observed in the Emfit QS measurements (28.3%, SD 14.4%) compared with the reference device (1.1%, SD 2.3%). The results showed that 5.0% of the mean HR and Ln RMSSD values were outside the limits of agreement. CONCLUSIONS Based on the present results, Emfit QS provides nocturnal HR and HRV data with an acceptable, small mean bias when calculating the mean of the sleep period. Thus, Emfit QS has the potential to be used for the long-term monitoring of nocturnal HR and HRV. However, further research is needed to assess reliability in HR and HRV detection.

show abstract

Performance Evaluation of Processing Methods for Ballistocardiogram Peak Detection

Cited by 5 publications

References 25 publications

A Contact-Free, Ballistocardiography-Based Monitoring System (Emfit QS) for Measuring Nocturnal Heart Rate and Heart Rate Variability: Validation Study

A Contact-Free, Ballistocardiography-Based Monitoring System (Emfit QS) for Measuring Nocturnal Heart Rate and Heart Rate Variability: Validation Study

Performance Comparison for Ballistocardiogram Peak Detection Methods

A Contact-Free, Ballistocardiography-Based Monitoring System (Emfit QS) for Measuring Nocturnal Heart Rate and Heart Rate Variability: Validation Study (Preprint)

Contact Info

Product

Resources

About