Treffer: A novel approach to exercise heart rate estimation combining PPG quality assessment with DNN modeling.
Title:
A novel approach to exercise heart rate estimation combining PPG quality assessment with DNN modeling.
Authors:
Source:
Medical & biological engineering & computing [Med Biol Eng Comput] 2025 Nov; Vol. 63 (11), pp. 3185-3202. Date of Electronic Publication: 2025 Jun 04.
Publication Type:
Journal Article
Language:
English
Journal Info:
Publisher: Springer Country of Publication: United States NLM ID: 7704869 Publication Model: Print-Electronic Cited Medium: Internet ISSN: 1741-0444 (Electronic) Linking ISSN: 01400118 NLM ISO Abbreviation: Med Biol Eng Comput Subsets: MEDLINE
Imprint Name(s):
Publication: New York, NY : Springer
Original Publication: Stevenage, Eng., Peregrinus.
Original Publication: Stevenage, Eng., Peregrinus.
MeSH Terms:
References:
Loh H, Xu S, Faust O (2022) Application of photoplethysmography signals for healthcare systems: an in-depth review[J]. Computer Methods and Programs in Biomedicine 216:106677. (PMID: 10.1016/j.cmpb.2022.106677)
Rahbar S, Ferrero R, Pegoraro PA, Toscani S (2024) Taylor-Fourier analysis of photoplethysmography signals for heart rate measurement, vol 2024. IEEE International Instrumentation and Measurement Technology Conference (I2MTC). Glasgow, United Kingdom pp 1–6.
Park P, Lee W, Cho S (2024) An adaptive filter based motion artifact cancellation technique using multi-wavelength PPG for accurate HR estimation [J]. IEEE Transactions on Biomedical Circuits and Systems 17(5):1074–1083. (PMID: 10.1109/TBCAS.2023.3315297)
Nishime EO, Cole CR, Blackstone EH (2000) Heart rate recovery and treadmill exercise score as predictors of mortality in patients referred for exercise ECG [J]. Jama 284(11):1392–1398. (PMID: 10.1001/jama.284.11.139210989401)
Sandvik L, Erikssen J, Ellestad M (1995) Heart rate increase and maximal heart rate during exercise as predictors of cardiovascular mortality: a 16-year follow-up study of 1960 healthy men[J]. Coronary artery disease 6(8):667–680. (PMID: 10.1097/00019501-199508000-000128574463)
Stein PK, Ehsani AA, Domitrovich PP (1999) Effect of exercise training on heart rate variability in healthy older adults[J]. American heart journal 138(3):567–576. (PMID: 10.1016/S0002-8703(99)70162-610467210)
Zhang Z (2015) Photoplethysmography-based heart rate monitoring in physical activities via joint sparse spectrum reconstruction[J]. IEEE transactions on biomedical engineering 62(8):1902–1910. (PMID: 10.1109/TBME.2015.240633226186747)
Chung H, Lee H, Lee J (2019) Finite state machine framework for instantaneous heart rate validation using wearable photoplethysmography during intensive exercise. IEEE Journal of Biomedical and Health Information 23(4):1595–1606. (PMID: 10.1109/JBHI.2018.287117730235152)
Temko A (2015) Estimation of heart rate from photoplethysmography during physical exercise using Wiener filtering and the phase vocoder, vol 2015. 37th annual international conference of the IEEE engineering in medicine and biology society (EMBC). IEEE, pp 1500–1503.
Xiong J, Cai L, Jiang D (2016) Spectral matrix decomposition-based motion artifacts removal in multi-channel PPG sensor signals[J]. IEEE Access 4:3076–3086. (PMID: 10.1109/ACCESS.2016.2580594)
Salehizadeh SMA, Dao D, Bolkhovsky J (2015) A novel time-varying spectral filtering algorithm for reconstruction of motion artifact corrupted heart rate signals during intense physical activities using a wearable photoplethysmogram sensor[J]. Sensors 16(1):10. (PMID: 10.3390/s16010010267036184732043)
Biswas D, Simões-Capela N, Van Hoof C (2019) Heart rate estimation from wrist-worn photoplethysmography: a review[J]. IEEE Sensors Journal 19(16):6560–6570. (PMID: 10.1109/JSEN.2019.2914166)
Zhou M, Selvaraj N (2020) Heart rate monitoring using sparse spectral curve tracing, vol 2020. 42nd Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC). IEEE, pp 5347–5352.
Reiss A, Indlekofer I, Schmidt P (2019) Deep PPG: Large-scale heart rate estimation with convolutional neural networks[J]. Sensors 19(14):3079. (PMID: 10.3390/s19143079313368946679242)
Wilkosz M, Szczęsna A (2021) Multi-headed conv-lstm network for heart rate estimation during daily living activities [J]. Sensors 21(15):5212. (PMID: 10.3390/s21155212343724478348622)
Song SB, Nam JW, Kim JH (2021) NAS-PPG: PPG-based heart rate estimation using neural architecture search[J]. IEEE Sensors Journal 21(13):14941–14949. (PMID: 10.1109/JSEN.2021.3073047)
Risso M Burrello A, Pagliari DJ (2021) Robust and energy-efficient PPG-based heart-rate monitoring, vol 2021. 2021 IEEE International Symposium on Circuits and Systems (ISCAS) IEEE, pp 1–5.
Burrello A, Pagliari DJ, Risso M (2021) Q-PPG: energy-efficient PPG-based heart rate monitoring on wearable devices[J]. IEEE Transactions on Biomedical Circuits and Systems 15(6):1196–1209. (PMID: 10.1109/TBCAS.2021.3122017)
Burrello A, Pagliari DJ, Bianco M (2022) Improving PPG-based heart-rate monitoring with synthetically generated data, vol 2022. IEEE Biomedical Circuits and Systems Conference (BioCAS) IEEE, pp 153–157.
Kasnesis P, Toumanidis L, Burrello A (2023) Feature-level cross-attentional PPG and motion signal fusion for heart rate estimation, vol 2023. IEEE 47th 47th Annual Computers, Software, and Applications Conference (COMPSAC) IEEE, pp 1731–1736.
Bieri V, Streli P, Demirel BU, Holz C (2023) BeliefPPG uncertainty-aware heart rate estimation from PPG signals via belief propagation, vol 216. Proceedings of the 39th Conference on Uncertainty in Artificial Intelligence (UAI 2023) PMLR pp 173–183.
Younesi A, Ansari M, Fazli M, Ejlali A, Shafique M, Henkel J (2024) A comprehensive survey of convolutions in deep learning: applications, challenges, and future trends[J]. IEEE Access 12:41180–41218. (PMID: 10.1109/ACCESS.2024.3376441)
Shiri FM, Perumal T, Mustapha N, Mohamed R (2024) A comprehensive overview and comparative analysis on deep learning models: CNN, RNN, LSTM, GRU https://doi.org/10.48550/arXiv.2305.17473.
Lee H, Chung H, Lee J (2019) Motion artifact cancellation in wearable photoplethysmography using gyroscope [J]. IEEE Sensors 19:1166–1175. (PMID: 10.1109/JSEN.2018.2879970)
Jarchi D, Casson AJ (2017) Description of a database containing wrist PPG signals recorded during physical exercise with both accelerometer and gyroscope measures of motion[J]. Data 2(1):1. (PMID: 10.3390/data2010001)
Schmidt P, Reiss A, Duerichen R (2018) Introducing WESAD, a multimodal dataset for wearable stress and affect detection. Proceedings of the 20th ACM international conference on multimodal interaction, pp 400-408 https://doi.org/10.1145/3242969.3242985.
Naeini EK, Sahaddi F, Azimi I (2023) A deep learning–based PPG quality assessment approach for heart rate and heart rate variability. ACM Transactions on Computing for Healthcare, 4(4):1-2 Article 24.
McLean MK, Weaver RG, Lane A (2023) A sliding scale signal quality metric of photoplethysmography applicable to measuring heart rate across clinical contexts with chest mounting as a case study. Sensors 23:3429. (PMID: 10.3390/s230734293705048810098585)
Shyam A, Ravichandran V, Preejith SP, Joseph J, Sivaprakasam M (2019) PPGnet: deep network for device independent heart rate estimation from photoplethysmogram. 41st Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), pp 1899–1902 https://doi.org/10.48550/arXiv.1903.08912.
Rocha LG (2020) Binary CorNET: accelerator for HR estimation from wrist-PPG[J]. IEEE Transactions on Biomedical Circuits and Systems 14(4):715–726. (PMID: 10.1109/TBCAS.2020.3001675)
Zhang Y (2019) Motion artifact reduction for wrist-worn photoplethysmograph sensors based on different wavelengths [J]. Sensors 19(3):673. (PMID: 10.3390/s19030673307363956387309)
Feli M, Kazemi K, Azimi I, Wang Y, Rahmani AM, Liljeberg P (2023) End-to-end PPG processing pipeline for wearables: from quality assessment and motion artifacts removal to HR/HRV feature extraction. 2023 IEEE International Conference on Bioinformatics and Biomedicine (BIBM) Istanbul, Turkiye 1895-1900 https://doi.org/10.1109/BIBM58861.2023.10385998.
Wang Y, Pan Z, Pan Y (2019) A training data set cleaning method by classification ability ranking for the k-nearest neighbor classifier[J]. IEEE transactions on neural networks and learning systems 31(5):1544–1556. (PMID: 10.1109/TNNLS.2019.292086431265416)
Hao D, Zhang L, Sumkin J (2020) Inaccurate labels in weakly-supervised deep learning: automatic identification and correction and their impact on classification performance[J]. IEEE journal of biomedical and health informatics 24(9):2701–2710. (PMID: 10.1109/JBHI.2020.2974425320785707429345)
Krishnan R, Natarajan B, Warren S (2008) Analysis and detection of motion artifact in photoplethysmographic data using higher order statistics, vol 2008 . 2008 IEEE International Conference on Acoustics, Speech and Signal Processing. IEEE, pp 613-616 https://doi.org/10.1109/ICASSP.2008.4517684.
Mashhadi MB, Asadi E, Eskandari M (2015) Heart rate tracking using wrist-type photoplethysmographic (PPG) signals during physical exercise with simultaneous accelerometry[J]. IEEE Signal Processing Letters 23(2):227–231. (PMID: 10.1109/LSP.2015.2509868)
Huang N, Selvaraj N (2020) Robust PPG-based ambulatory heart rate tracking algorithm, vol 2020. 2020 42nd Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC), pp 5929–5934 https://doi.org/10.1109/EMBC44109.2020.9175346.
Huang J, Dai J, Li Y (2023) Research on PCA-Kmeans++ clustering algorithm considering spatiotemporal dimension, vol 2023. 2023 2nd International Conference on 3D Immersion, Interaction and Multi-sensory Experiences (ICDIIME), Madrid, Spain, pp 195-201 https://doi.org/10.1109/ICDIIME59043.2023.00042.
Wu J, Luo S, Chen S, Lyu Y, Ji X Meng Y, Wang D (2024) Continuous authentication via wrist photoplethysmogram: an extensive study. IEEE Transactions on Mobile Computing (early access) https://doi.org/10.1109/TMC.2024.3443208.
Ribeiro L, Hu X, Oliveira HP, Pereira T (2023) AI-based models to predict the heart rate using PPG and accelerometer signals during physical exercise, 2023. 2023 IEEE International Conference on Bioinformatics and Biomedicine (BIBM), Istanbul, Turkiye, pp 4398-4403 https://doi.org/10.1109/BIBM58861.2023.10385871.
Bai S, Kolter JZ, Koltun V (2018) An empirical evaluation of generic convolutional and recurrent networks for sequence modeling. https://doi.org/10.48550/arXiv.1803.01271.
Kanoga S, Hoshino T, Kamei S (2023) Comparison of seven shallow and deep regressors in continuous blood pressure and heart rate estimation using single-channel photoplethysmograms under three evaluation cases[J]. Biomedical Signal Processing and Control 85:105029. (PMID: 10.1016/j.bspc.2023.105029)
Rahbar S, Ferrero R, Pegoraro PA, Toscani S (2024) Taylor-Fourier analysis of photoplethysmography signals for heart rate measurement, vol 2024. IEEE International Instrumentation and Measurement Technology Conference (I2MTC). Glasgow, United Kingdom pp 1–6.
Park P, Lee W, Cho S (2024) An adaptive filter based motion artifact cancellation technique using multi-wavelength PPG for accurate HR estimation [J]. IEEE Transactions on Biomedical Circuits and Systems 17(5):1074–1083. (PMID: 10.1109/TBCAS.2023.3315297)
Nishime EO, Cole CR, Blackstone EH (2000) Heart rate recovery and treadmill exercise score as predictors of mortality in patients referred for exercise ECG [J]. Jama 284(11):1392–1398. (PMID: 10.1001/jama.284.11.139210989401)
Sandvik L, Erikssen J, Ellestad M (1995) Heart rate increase and maximal heart rate during exercise as predictors of cardiovascular mortality: a 16-year follow-up study of 1960 healthy men[J]. Coronary artery disease 6(8):667–680. (PMID: 10.1097/00019501-199508000-000128574463)
Stein PK, Ehsani AA, Domitrovich PP (1999) Effect of exercise training on heart rate variability in healthy older adults[J]. American heart journal 138(3):567–576. (PMID: 10.1016/S0002-8703(99)70162-610467210)
Zhang Z (2015) Photoplethysmography-based heart rate monitoring in physical activities via joint sparse spectrum reconstruction[J]. IEEE transactions on biomedical engineering 62(8):1902–1910. (PMID: 10.1109/TBME.2015.240633226186747)
Chung H, Lee H, Lee J (2019) Finite state machine framework for instantaneous heart rate validation using wearable photoplethysmography during intensive exercise. IEEE Journal of Biomedical and Health Information 23(4):1595–1606. (PMID: 10.1109/JBHI.2018.287117730235152)
Temko A (2015) Estimation of heart rate from photoplethysmography during physical exercise using Wiener filtering and the phase vocoder, vol 2015. 37th annual international conference of the IEEE engineering in medicine and biology society (EMBC). IEEE, pp 1500–1503.
Xiong J, Cai L, Jiang D (2016) Spectral matrix decomposition-based motion artifacts removal in multi-channel PPG sensor signals[J]. IEEE Access 4:3076–3086. (PMID: 10.1109/ACCESS.2016.2580594)
Salehizadeh SMA, Dao D, Bolkhovsky J (2015) A novel time-varying spectral filtering algorithm for reconstruction of motion artifact corrupted heart rate signals during intense physical activities using a wearable photoplethysmogram sensor[J]. Sensors 16(1):10. (PMID: 10.3390/s16010010267036184732043)
Biswas D, Simões-Capela N, Van Hoof C (2019) Heart rate estimation from wrist-worn photoplethysmography: a review[J]. IEEE Sensors Journal 19(16):6560–6570. (PMID: 10.1109/JSEN.2019.2914166)
Zhou M, Selvaraj N (2020) Heart rate monitoring using sparse spectral curve tracing, vol 2020. 42nd Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC). IEEE, pp 5347–5352.
Reiss A, Indlekofer I, Schmidt P (2019) Deep PPG: Large-scale heart rate estimation with convolutional neural networks[J]. Sensors 19(14):3079. (PMID: 10.3390/s19143079313368946679242)
Wilkosz M, Szczęsna A (2021) Multi-headed conv-lstm network for heart rate estimation during daily living activities [J]. Sensors 21(15):5212. (PMID: 10.3390/s21155212343724478348622)
Song SB, Nam JW, Kim JH (2021) NAS-PPG: PPG-based heart rate estimation using neural architecture search[J]. IEEE Sensors Journal 21(13):14941–14949. (PMID: 10.1109/JSEN.2021.3073047)
Risso M Burrello A, Pagliari DJ (2021) Robust and energy-efficient PPG-based heart-rate monitoring, vol 2021. 2021 IEEE International Symposium on Circuits and Systems (ISCAS) IEEE, pp 1–5.
Burrello A, Pagliari DJ, Risso M (2021) Q-PPG: energy-efficient PPG-based heart rate monitoring on wearable devices[J]. IEEE Transactions on Biomedical Circuits and Systems 15(6):1196–1209. (PMID: 10.1109/TBCAS.2021.3122017)
Burrello A, Pagliari DJ, Bianco M (2022) Improving PPG-based heart-rate monitoring with synthetically generated data, vol 2022. IEEE Biomedical Circuits and Systems Conference (BioCAS) IEEE, pp 153–157.
Kasnesis P, Toumanidis L, Burrello A (2023) Feature-level cross-attentional PPG and motion signal fusion for heart rate estimation, vol 2023. IEEE 47th 47th Annual Computers, Software, and Applications Conference (COMPSAC) IEEE, pp 1731–1736.
Bieri V, Streli P, Demirel BU, Holz C (2023) BeliefPPG uncertainty-aware heart rate estimation from PPG signals via belief propagation, vol 216. Proceedings of the 39th Conference on Uncertainty in Artificial Intelligence (UAI 2023) PMLR pp 173–183.
Younesi A, Ansari M, Fazli M, Ejlali A, Shafique M, Henkel J (2024) A comprehensive survey of convolutions in deep learning: applications, challenges, and future trends[J]. IEEE Access 12:41180–41218. (PMID: 10.1109/ACCESS.2024.3376441)
Shiri FM, Perumal T, Mustapha N, Mohamed R (2024) A comprehensive overview and comparative analysis on deep learning models: CNN, RNN, LSTM, GRU https://doi.org/10.48550/arXiv.2305.17473.
Lee H, Chung H, Lee J (2019) Motion artifact cancellation in wearable photoplethysmography using gyroscope [J]. IEEE Sensors 19:1166–1175. (PMID: 10.1109/JSEN.2018.2879970)
Jarchi D, Casson AJ (2017) Description of a database containing wrist PPG signals recorded during physical exercise with both accelerometer and gyroscope measures of motion[J]. Data 2(1):1. (PMID: 10.3390/data2010001)
Schmidt P, Reiss A, Duerichen R (2018) Introducing WESAD, a multimodal dataset for wearable stress and affect detection. Proceedings of the 20th ACM international conference on multimodal interaction, pp 400-408 https://doi.org/10.1145/3242969.3242985.
Naeini EK, Sahaddi F, Azimi I (2023) A deep learning–based PPG quality assessment approach for heart rate and heart rate variability. ACM Transactions on Computing for Healthcare, 4(4):1-2 Article 24.
McLean MK, Weaver RG, Lane A (2023) A sliding scale signal quality metric of photoplethysmography applicable to measuring heart rate across clinical contexts with chest mounting as a case study. Sensors 23:3429. (PMID: 10.3390/s230734293705048810098585)
Shyam A, Ravichandran V, Preejith SP, Joseph J, Sivaprakasam M (2019) PPGnet: deep network for device independent heart rate estimation from photoplethysmogram. 41st Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), pp 1899–1902 https://doi.org/10.48550/arXiv.1903.08912.
Rocha LG (2020) Binary CorNET: accelerator for HR estimation from wrist-PPG[J]. IEEE Transactions on Biomedical Circuits and Systems 14(4):715–726. (PMID: 10.1109/TBCAS.2020.3001675)
Zhang Y (2019) Motion artifact reduction for wrist-worn photoplethysmograph sensors based on different wavelengths [J]. Sensors 19(3):673. (PMID: 10.3390/s19030673307363956387309)
Feli M, Kazemi K, Azimi I, Wang Y, Rahmani AM, Liljeberg P (2023) End-to-end PPG processing pipeline for wearables: from quality assessment and motion artifacts removal to HR/HRV feature extraction. 2023 IEEE International Conference on Bioinformatics and Biomedicine (BIBM) Istanbul, Turkiye 1895-1900 https://doi.org/10.1109/BIBM58861.2023.10385998.
Wang Y, Pan Z, Pan Y (2019) A training data set cleaning method by classification ability ranking for the k-nearest neighbor classifier[J]. IEEE transactions on neural networks and learning systems 31(5):1544–1556. (PMID: 10.1109/TNNLS.2019.292086431265416)
Hao D, Zhang L, Sumkin J (2020) Inaccurate labels in weakly-supervised deep learning: automatic identification and correction and their impact on classification performance[J]. IEEE journal of biomedical and health informatics 24(9):2701–2710. (PMID: 10.1109/JBHI.2020.2974425320785707429345)
Krishnan R, Natarajan B, Warren S (2008) Analysis and detection of motion artifact in photoplethysmographic data using higher order statistics, vol 2008 . 2008 IEEE International Conference on Acoustics, Speech and Signal Processing. IEEE, pp 613-616 https://doi.org/10.1109/ICASSP.2008.4517684.
Mashhadi MB, Asadi E, Eskandari M (2015) Heart rate tracking using wrist-type photoplethysmographic (PPG) signals during physical exercise with simultaneous accelerometry[J]. IEEE Signal Processing Letters 23(2):227–231. (PMID: 10.1109/LSP.2015.2509868)
Huang N, Selvaraj N (2020) Robust PPG-based ambulatory heart rate tracking algorithm, vol 2020. 2020 42nd Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC), pp 5929–5934 https://doi.org/10.1109/EMBC44109.2020.9175346.
Huang J, Dai J, Li Y (2023) Research on PCA-Kmeans++ clustering algorithm considering spatiotemporal dimension, vol 2023. 2023 2nd International Conference on 3D Immersion, Interaction and Multi-sensory Experiences (ICDIIME), Madrid, Spain, pp 195-201 https://doi.org/10.1109/ICDIIME59043.2023.00042.
Wu J, Luo S, Chen S, Lyu Y, Ji X Meng Y, Wang D (2024) Continuous authentication via wrist photoplethysmogram: an extensive study. IEEE Transactions on Mobile Computing (early access) https://doi.org/10.1109/TMC.2024.3443208.
Ribeiro L, Hu X, Oliveira HP, Pereira T (2023) AI-based models to predict the heart rate using PPG and accelerometer signals during physical exercise, 2023. 2023 IEEE International Conference on Bioinformatics and Biomedicine (BIBM), Istanbul, Turkiye, pp 4398-4403 https://doi.org/10.1109/BIBM58861.2023.10385871.
Bai S, Kolter JZ, Koltun V (2018) An empirical evaluation of generic convolutional and recurrent networks for sequence modeling. https://doi.org/10.48550/arXiv.1803.01271.
Kanoga S, Hoshino T, Kamei S (2023) Comparison of seven shallow and deep regressors in continuous blood pressure and heart rate estimation using single-channel photoplethysmograms under three evaluation cases[J]. Biomedical Signal Processing and Control 85:105029. (PMID: 10.1016/j.bspc.2023.105029)
Grant Information:
82272113 the National Nature Science Foundation of China; 61871360 the National Nature Science Foundation of China
Contributed Indexing:
Keywords: Bidirectional long short-term memory; Convolutional neural network; Deep neural network; Frequency domain kurtosis; Heart rate estimation; Photoplethysmogram
Entry Date(s):
Date Created: 20250603 Date Completed: 20251120 Latest Revision: 20251120
Update Code:
20251121
DOI:
10.1007/s11517-025-03379-x
PMID:
40461925
Database:
MEDLINE
Weitere Informationen
This paper proposes a novel approach for exercise heart rate (HR) estimation by integrating PPG quality assessment with deep neural network (DNN) modeling. A frequency-domain kurtosis (kurtF) metric is introduced to identify high-quality PPG samples, optimizing DNN training data and mitigating motion artifacts. An E-K scatter plot is used to visualize sample quality distribution, aiding dataset variability analysis. The proposed DNN model, designed for single-channel PPG input, demonstrates strong HR estimation performance on public datasets, achieving a mean absolute error (MAE) values of 3.76 bpm (PPG_DaLiA) and 3.18 bpm (IEEE-Training). Theoretical analysis and experimental validation confirm that prioritizing high-quality samples enhances model stability, accuracy, and generalizability. Additionally, a dataset quality analysis method is introduced to facilitate comparative assessments. The kurtF metric and quality-driven sample selection strategy provide a robust framework for improving HR estimation, even in data-limited scenarios. This study underscores the importance of integrating sample quality assessment into HR estimation workflows, paving the way for more accurate and reliable PPG-based HR monitoring during exercise.
(© 2025. International Federation for Medical and Biological Engineering.)
Declarations. Competing interests: The authors declare no competing interests.