Robust PPG Peak Detection Using Dilated Convolutional Neural Networks

Kazemi, Kianoosh; Laitala, Juho; Azimi, Iman; Liljeberg, Pasi; Rahmani, Amir M.

doi:10.3390/s22166054

Cited by 25 publications

(11 citation statements)

References 58 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…An alternative explanation for measuring slightly too short stride lengths could also be attributed to the geometric considerations of Lidar placement at shin height: as the height of measurement increases, the recorded stride lengths appear shorter. Although optimizing the algorithm detecting peaks was not the focus of the present work, the problem could be overcome by using an adaptive peak detector based on dilated convolutional neural networks [ 52 ].…”

Section: Discussionmentioning

confidence: 99%

A Transferable Lidar-Based Method to Conduct Contactless Assessments of Gait Parameters in Diverse Home-like Environments

Single,

Bruhin,

Colombo

et al. 2024

Sensors

View full text Add to dashboard Cite

Gait abnormalities in older adults are linked to increased risks of falls, institutionalization, and mortality, necessitating accurate and frequent gait assessments beyond traditional clinical settings. Current methods, such as pressure-sensitive walkways, often lack the continuous natural environment monitoring needed to understand an individual’s gait fully during their daily activities. To address this gap, we present a Lidar-based method capable of unobtrusively and continuously tracking human leg movements in diverse home-like environments, aiming to match the accuracy of a clinical reference measurement system. We developed a calibration-free step extraction algorithm based on mathematical morphology to realize Lidar-based gait analysis. Clinical gait parameters of 45 healthy individuals were measured using Lidar and reference systems (a pressure-sensitive walkway and a video recording system). Each participant participated in three predefined ambulation experiments by walking over the walkway. We observed linear relationships with strong positive correlations (R2>0.9) between the values of the gait parameters (step and stride length, step and stride time, cadence, and velocity) measured with the Lidar sensors and the pressure-sensitive walkway reference system. Moreover, the lower and upper 95% confidence intervals of all gait parameters were tight. The proposed algorithm can accurately derive gait parameters from Lidar data captured in home-like environments, with a performance not significantly less accurate than clinical reference systems.

show abstract

Section: Discussionmentioning

confidence: 99%

A Transferable Lidar-Based Method to Conduct Contactless Assessments of Gait Parameters in Diverse Home-like Environments

Single,

Bruhin,

Colombo

et al. 2024

Sensors

View full text Add to dashboard Cite

show abstract

“…PPG peak detection —We used a deep-learning-based method introduced by Kazemi et al . [ 40 ] for PPG peak detection. The method outperforms other state-of-the-art methods [ 41 , 42 ], particularly when the signal is noisy.…”

Section: Methodsmentioning

confidence: 99%

A comprehensive accuracy assessment of Samsung smartwatch heart rate and heart rate variability

et al. 2022

Self Cite

View full text Add to dashboard Cite

Background Photoplethysmography (PPG) is a low-cost and easy-to-implement method to measure vital signs, including heart rate (HR) and pulse rate variability (PRV) which widely used as a substitute of heart rate variability (HRV). The method is used in various wearable devices. For example, Samsung smartwatches are PPG-based open-source wristbands used in remote well-being monitoring and fitness applications. However, PPG is highly susceptible to motion artifacts and environmental noise. A validation study is required to investigate the accuracy of PPG-based wearable devices in free-living conditions. Objective We evaluate the accuracy of PPG signals—collected by the Samsung Gear Sport smartwatch in free-living conditions—in terms of HR and time-domain and frequency-domain HRV parameters against a medical-grade chest electrocardiogram (ECG) monitor. Methods We conducted 24-hours monitoring using a Samsung Gear Sport smartwatch and a Shimmer3 ECG device. The monitoring included 28 participants (14 male and 14 female), where they engaged in their daily routines. We evaluated HR and HRV parameters during the sleep and awake time. The parameters extracted from the smartwatch were compared against the ECG reference. For the comparison, we employed the Pearson correlation coefficient, Bland-Altman plot, and linear regression methods. Results We found a significantly high positive correlation between the smartwatch’s and Shimmer ECG’s HR, time-domain HRV, LF, and HF and a significant moderate positive correlation between the smartwatch’s and shimmer ECG’s LF/HF during sleep time. The mean biases of HR, time-domain HRV, and LF/HF were low, while the biases of LF and HF were moderate during sleep. The regression analysis showed low error variances of HR, AVNN, and pNN50, moderate error variances of SDNN, RMSSD, LF, and HF, and high error variances of LF/HF during sleep. During the awake time, there was a significantly high positive correlation of AVNN and a moderate positive correlation of HR, while the other parameters indicated significantly low positive correlations. RMSSD and SDNN showed low mean biases, and the other parameters had moderate mean biases. In addition, AVNN had moderate error variance while the other parameters indicated high error variances. Conclusion The Samsung smartwatch provides acceptable HR, time-domain HRV, LF, and HF parameters during sleep time. In contrast, during the awake time, AVNN and HR show satisfactory accuracy, and the other HRV parameters have high errors.

show abstract

“…Systolic peaks are crucial elements for deriving vital signs from PPG signals, as they represent essential characteristics associated with the cardiovascular system. We employ a deep-learning-based method that we have developed and previously published 69 , specifically designed for the extraction of systolic peaks from PPG signals.…”

Section: Hr and Hrv Data Extractionmentioning

confidence: 99%

Preterm Birth Risk Stratification through Longitudinal Heart Rate and HRV Monitoring in Daily Life

Feli,

Azimi,

Sarhaddi

et al. 2024

Preprint

Self Cite

View full text Add to dashboard Cite

Preterm birth (PTB) remains a global health concern, impacting neonatal mortality and lifelong health consequences. Traditional methods for estimating PTB rely on electronic health records or biomedical signals, limited to short-term assessments in clinical settings. Recent studies have leveraged wearable technologies for in-home maternal health monitoring, offering continuous assessment of maternal autonomic nervous system (ANS) activity and facilitating the exploration of PTB risk. In this paper, we conduct a longitudinal study to assess the risk of PTB by examining maternal ANS activity through heart rate (HR) and heart rate variability (HRV). To achieve this, we collect long-term raw photoplethysmogram (PPG) signals from 58 pregnant women (including seven preterm cases) from gestational weeks 12-15 to three months post-delivery using smartwatches in daily life settings. We employ a PPG processing pipeline to accurately extract HR and HRV, and an autoencoder machine learning model with SHAP analysis to generate explainable abnormality scores indicative of PTB risk. Our results reveal distinctive patterns in PTB abnormality scores during the second pregnancy trimester, indicating the potential for early PTB risk estimation. Moreover, we find that HR, average of interbeat intervals (AVNN), SD1SD2 ratio, and standard deviation of interbeat intervals (SDNN) emerge as significant PTB indicators.

show abstract

Robust PPG Peak Detection Using Dilated Convolutional Neural Networks

Cited by 25 publications

References 58 publications

A Transferable Lidar-Based Method to Conduct Contactless Assessments of Gait Parameters in Diverse Home-like Environments

A Transferable Lidar-Based Method to Conduct Contactless Assessments of Gait Parameters in Diverse Home-like Environments

A comprehensive accuracy assessment of Samsung smartwatch heart rate and heart rate variability

Preterm Birth Risk Stratification through Longitudinal Heart Rate and HRV Monitoring in Daily Life

Contact Info

Product

Resources

About