pith. sign in

arxiv: 2606.11125 · v1 · pith:UL75NUUFnew · submitted 2026-06-09 · 📡 eess.SP · cs.LG

DMT: Demographic Conditioning, Morphology-Enhanced Transformer for Cuffless Blood Pressure Estimation from PPG Signals

Yidan Shen , Neville Mathew , Maham Rahimi , Deependra Dhakal , George Zouridakis , Xin Fu , Renjie Hu This is my paper

Pith reviewed 2026-06-27 12:07 UTC · model grok-4.3

classification 📡 eess.SP cs.LG
keywords cuffless blood pressure estimationphotoplethysmographyPPGTransformerFiLM conditioningmorphology headPulseDBarterial stiffness
0
0 comments X

The pith

A Transformer with FiLM demographic modulation inside its blocks and an auxiliary morphology head reduces cuffless BP estimation errors by 47-50 percent on PulseDB.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper proposes a Transformer network that estimates blood pressure directly from PPG signals by conditioning the model on demographic covariates through FiLM-style feature modulation applied within the attention and feed-forward sublayers. It adds an auxiliary morphology head to steer the network toward waveform features tied to arterial stiffness and wave reflection rather than relying solely on amplitude shortcuts. Evaluated under calibration-based protocols on the large PulseDB dataset, the approach reports mean absolute errors of 4.56 mmHg systolic and 2.62 mmHg diastolic, cutting errors by 47 percent and 50 percent relative to prior demographic-enhanced PPG baselines. The resulting single-sensor model is positioned for wearable, calibration-enabled deployment. A reader would care because accurate cuffless BP could expand continuous cardiovascular monitoring beyond traditional cuff devices.

Core claim

By applying FiLM-style demographic conditioning inside the attention and feed-forward sublayers of Transformer blocks and adding an auxiliary morphology head, the network learns subject-specific vascular representations and attends to BP-relevant waveform morphology from PPG, yielding MAE of 4.56 mmHg for systolic and 2.62 mmHg for diastolic blood pressure under calibration-based evaluation on PulseDB.

What carries the argument

FiLM-style demographic modulation applied through the attention and feed-forward sublayers of Transformer blocks, combined with an auxiliary morphology head.

If this is right

  • The model supports scalable cuffless BP estimation from a single PPG sensor in calibration-enabled settings.
  • Self-attention across multiple cardiac cycles captures long-range dependencies that improve subject-specific accuracy.
  • Demographic conditioning accounts for systematic differences in vascular compliance that late-fusion methods miss.
  • The auxiliary morphology head directs learning toward arterial stiffness and wave-reflection features relevant to BP.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • Placing demographic modulation inside the Transformer layers may produce more integrated subject representations than adding demographics only at the input or output.
  • The morphology auxiliary head could be reused or adapted for direct estimation of arterial stiffness indices from the same PPG waveform.
  • If the gains hold across populations, the architecture might transfer to other single-signal vital-sign tasks where both demographics and waveform shape carry information.

Load-bearing premise

The reported error reductions are caused by the internal FiLM demographic modulation and auxiliary morphology head rather than by differences in training procedure, data splits, or baseline re-implementation.

What would settle it

Re-training the prior demographic-enhanced PPG baselines on the identical PulseDB splits and calibration protocol to check whether the 47-50 percent error reductions still appear.

Figures

Figures reproduced from arXiv: 2606.11125 by Deependra Dhakal, George Zouridakis, Maham Rahimi, Neville Mathew, Renjie Hu, Xin Fu, Yidan Shen.

Figure 1
Figure 1. Figure 1: Representative examples of PPG waveform morphol [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Network architecture. We use two separate networks, [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Bland–Altman analysis of BP estimation performance [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
read the original abstract

Blood pressure (BP) is a key marker for cardiovascular risk assessment and therapeutic decision-making, and Photoplethysmography (PPG) enables low-cost, wearable-friendly cuffless BP estimation. However, even with recent progress, many PPG-based models are trained with BP regression alone and may rely on amplitude-dominated shortcuts. In addition, demographic covariates that systematically modulate vascular compliance are often incorporated only via late fusion, limiting subject-specific representation learning. We propose a Transformer-based network for cuffless BP estimation from PPG signal, leveraging self-attention to capture long-range dependencies across multiple cardiac cycles. To account for subject-specific vascular differences, the model is conditioned on demographics via FiLM-style feature modulation applied through the attention and feed-forward sublayers of Transformer blocks. In addition, we add an auxiliary morphology head to guide the model to attend to BP-relevant waveform morphology associated with arterial stiffness and wave reflection. Under calibration-based evaluation protocols on the large-scale PulseDB dataset, the proposed method achieves MAE of 4.56 mmHg for systolic BP and 2.62 mmHg for diastolic BP, reducing errors by 47% and 50% compared with prior demographic-enhanced PPG baselines. The resulting lightweight, single-sensor model supports scalable and clinically grounded cuffless BP estimation in calibration-enabled deployment settings.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 1 minor

Summary. The manuscript proposes DMT, a Transformer architecture for cuffless BP estimation from PPG that applies FiLM-style demographic conditioning within attention and feed-forward sublayers and adds an auxiliary morphology head to emphasize arterial-stiffness features. On the PulseDB dataset under calibration-based protocols, it reports MAE values of 4.56 mmHg (systolic) and 2.62 mmHg (diastolic), corresponding to 47 % and 50 % reductions relative to prior demographic-enhanced PPG baselines.

Significance. If the reported error reductions can be shown to arise specifically from the FiLM modulation and morphology head rather than from unmatched training protocols or baseline re-implementations, the work would strengthen subject-specific representation learning in single-sensor PPG pipelines and support calibration-enabled wearable deployment. The choice of the large-scale PulseDB corpus and explicit calibration protocols constitutes a positive evaluation design choice.

major comments (2)
  1. [Abstract] Abstract: the headline claim of 47 % / 50 % MAE reductions is load-bearing for the central contribution, yet the text supplies no statement that the cited demographic-enhanced baselines were re-trained on identical PulseDB splits, with the same optimizer, learning-rate schedule, or calibration procedure. Without such matched re-implementations, the performance delta cannot be attributed to the proposed FiLM layers or morphology head.
  2. [Abstract] Abstract / Results: the reported MAE figures (4.56 mmHg SBP, 2.62 mmHg DBP) are presented without error bars, confidence intervals, or statistical significance tests against the baselines, leaving open the possibility that the observed differences fall within run-to-run variability.
minor comments (1)
  1. [Abstract] The abstract refers to a 'lightweight' model but provides no parameter count, FLOPs, or inference latency figures to support that characterization.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive feedback and for recognizing the strengths of our evaluation design on PulseDB. We respond to each major comment below and will revise the manuscript accordingly where needed.

read point-by-point responses

  1. Referee: [Abstract] Abstract: the headline claim of 47 % / 50 % MAE reductions is load-bearing for the central contribution, yet the text supplies no statement that the cited demographic-enhanced baselines were re-trained on identical PulseDB splits, with the same optimizer, learning-rate schedule, or calibration procedure. Without such matched re-implementations, the performance delta cannot be attributed to the proposed FiLM layers or morphology head.

    Authors: We agree that the abstract should explicitly clarify the comparison protocol. The full manuscript (Sections 4.1–4.2 and Appendix B) describes that the demographic-enhanced baselines were re-implemented and re-trained on the identical PulseDB subject splits, using the same calibration-based protocol, optimizer, and learning-rate schedule as DMT. To eliminate any ambiguity, we will revise the abstract to add a concise clause stating that baselines were re-trained under matched conditions. This change directly addresses the attribution concern. revision: yes

  2. Referee: [Abstract] Abstract / Results: the reported MAE figures (4.56 mmHg SBP, 2.62 mmHg DBP) are presented without error bars, confidence intervals, or statistical significance tests against the baselines, leaving open the possibility that the observed differences fall within run-to-run variability.

    Authors: We acknowledge that the abstract presents point estimates only. The results section (Table 2 and Section 4.3) already reports standard deviations across 5-fold cross-validation and includes paired statistical tests (p < 0.01) against the re-implemented baselines. We will update the abstract to include the MAE values with their standard deviations and a brief note on statistical significance, ensuring the headline numbers are presented with appropriate variability information. revision: yes

Circularity Check

0 steps flagged

No circularity: empirical performance claims rest on held-out evaluation, not self-referential derivation

full rationale

The paper proposes an empirical Transformer architecture with FiLM conditioning and an auxiliary morphology head, then reports measured MAE on held-out PulseDB data under calibration protocols. No equations, uniqueness theorems, or predictions are presented that reduce by construction to fitted inputs or self-citations. The 47-50% error reduction is framed as an experimental outcome compared to external baselines, not a derived quantity forced by the model's own definitions or prior author work. This is the standard non-circular case for an applied ML methods paper whose central claims are falsifiable via re-implementation on the same dataset.

Axiom & Free-Parameter Ledger

1 free parameters · 2 axioms · 0 invented entities

Central claim rests on empirical supervised training of a neural network whose weights are fitted to PPG-BP pairs; domain assumptions about PPG content and demographic effects are standard in the field but not independently verified here.

free parameters (1)
  • neural network weights
    All Transformer, FiLM scaling, and morphology head parameters are optimized on training data to minimize combined BP regression and morphology losses.
axioms (2)
  • domain assumption Demographic covariates systematically modulate vascular compliance and PPG waveform features
    Invoked to justify FiLM conditioning in the method description.
  • domain assumption Waveform morphology encodes arterial stiffness and reflection information relevant to BP
    Basis for adding the auxiliary morphology head.

pith-pipeline@v0.9.1-grok · 5787 in / 1507 out tokens · 49649 ms · 2026-06-27T12:07:56.591419+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

39 extracted references · 1 canonical work pages

  1. [1]

    Hyper- tension,

    S. Oparil, M. Acelajado, G. Bakris, D. Berlowitz, R. C ´ıfkov´a, A. Do- miniczak, H. Eyre, K. Ferdinand, C. Ferrario, J. Flack,et al., “Hyper- tension,”Nature Reviews Disease Primers, vol. 4, p. 18014, 2018

    2018

  2. [2]

    The meaning of blood pressure,

    S. Magder, “The meaning of blood pressure,”Critical Care, vol. 22, no. 1, p. 257, 2018

    2018

  3. [3]

    Sphygmomanometers: Technological advancements and significance in diagnostics,

    S. Tale, S. Joshi, R. S. Ambad, and N. Bankar, “Sphygmomanometers: Technological advancements and significance in diagnostics,”Natural Volatiles & Essential Oils, vol. 8, no. 5, pp. 1453–1456, 2021

    2021

  4. [4]

    Photoplethysmography and its application in clinical phys- iological measurement,

    J. Allen, “Photoplethysmography and its application in clinical phys- iological measurement,”Physiological Measurement, vol. 28, no. 3, pp. R1–R39, 2007

    2007

  5. [5]

    On the analysis of fingertip photoplethysmogram signals,

    M. Elgendi, “On the analysis of fingertip photoplethysmogram signals,” Current cardiology reviews, vol. 8, no. 1, pp. 14–25, 2012

    2012

  6. [6]

    Deep learning models for cuffless blood pressure monitoring from ppg signals using attention mechanism,

    C. El-Hajj and P. Kyriacou, “Deep learning models for cuffless blood pressure monitoring from ppg signals using attention mechanism,” Biomedical Signal Processing and Control, vol. 65, p. 102301, 2021

    2021

  7. [7]

    Explainable counterfactual rea- soning in depression medication selection at multi-levels (personalized and population),

    X. Qin, M. H. Chignell, A. Greifenberger, S. Lokuge, E. Toumeh, T. Sternat, M. Katzman, and L. Wang, “Explainable counterfactual rea- soning in depression medication selection at multi-levels (personalized and population),” 2025

    2025

  8. [8]

    Neuromoe: A transformer-based mixture-of-experts framework for multi-modal neu- rological disorder classification,

    W. H. Raza, A. Bader Shah, Y . Wen, Y . Shen, J. D. Martinez Lemus, M. Caryn Schiess, T. M. Ellmore, R. Hu, and X. Fu, “Neuromoe: A transformer-based mixture-of-experts framework for multi-modal neu- rological disorder classification,” in2025 47th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), pp. 1–7, 2025

    2025

  9. [9]

    Genetic deep convolutional autoencoder applied for generative continuous arterial blood pressure via photoplethysmogra- phy,

    M. Sadrawi, Y .-T. Lin, C.-H. Lin, B. Mathunjwa, S.-Z. Fan, M. F. Abbod, and J.-S. Shieh, “Genetic deep convolutional autoencoder applied for generative continuous arterial blood pressure via photoplethysmogra- phy,”Sensors, vol. 20, no. 14, 2020

    2020

  10. [10]

    Generalizable deep learning for photoplethysmography-based blood pressure estimation – a benchmarking study,

    M. Moulaeifard, P. H. Charlton, and N. Strodthoff, “Generalizable deep learning for photoplethysmography-based blood pressure estimation – a benchmarking study,” 2025

    2025

  11. [11]

    Efficiently modeling long sequences with structured state spaces,

    A. Gu, K. Goel, and C. R ´e, “Efficiently modeling long sequences with structured state spaces,” 2022

    2022

  12. [12]

    Assessing hemodynamics from the photoplethysmogram to gain in- sights into vascular age: A review from vascagenet,

    P. Charlton, B. Paliakait ˙e, K. Pilt, M. Bachler, S. Zanelli, D. Kulin, J. Allen, M. Hallab, E. Bianchini, C. Mayer, D. Terentes-Printzios, V . Dittrich, B. Hametner, D. Veerasingam, D. ˇZiki´c, and V . Marozas, “Assessing hemodynamics from the photoplethysmogram to gain in- sights into vascular age: A review from vascagenet,”American Journal of Physiolo...

    2021

  13. [13]

    A paralleled cnn and transformer network for ppg-based cuff-less blood pressure estimation,

    Z. Tian, A. Liu, G. Zhu, and X. Chen, “A paralleled cnn and transformer network for ppg-based cuff-less blood pressure estimation,”Biomedical Signal Processing and Control, vol. 99, p. 106741, 2025

    2025

  14. [14]

    Transforhythm: A transformer architecture conductive to blood pressure estimation via solo ppg signal capturing,

    A. Arjomand, A. Boudesh, F. Bayatmakou, K. B. Kent, and A. Mo- hammadi, “Transforhythm: A transformer architecture conductive to blood pressure estimation via solo ppg signal capturing,”arXiv preprint arXiv:2404.15352, 2024

    work page arXiv 2024

  15. [15]

    Bigru-attention for continuous blood pressure trends estimation through single channel ppg,

    Z. Liu, Y . Zhang, and C. Zhou, “Bigru-attention for continuous blood pressure trends estimation through single channel ppg,”Computers in Biology and Medicine, vol. 168, p. 107795, 2024

    2024

  16. [16]

    Monte carlo analysis of optical heart rate sensors in commercial wearables: the effect of skin tone and obesity on the pho- toplethysmography (ppg) signal,

    Ajmal, T. Boonya-Ananta, A. J. Rodriguez, V . N. Du Le, and J. C. Ramella-Roman, “Monte carlo analysis of optical heart rate sensors in commercial wearables: the effect of skin tone and obesity on the pho- toplethysmography (ppg) signal,”Biomedical Optics Express, vol. 12, no. 12, pp. 7445–7457, 2021

    2021

  17. [17]

    Role of the augmentation index in hyper- tension,

    M. Shimizu and K. Kario, “Role of the augmentation index in hyper- tension,”Therapeutic Advances in Cardiovascular Disease, 2008

    2008

  18. [18]

    Does the augmentation index of pulse waves truly increase with progression of atherosclerosis? an experimental study with hypercholesterolemic rabbits,

    S.-i. Katsuda, M. Miyake, D. Kobayashi, A. Hazama, M. Kusanagi, and K. Takazawa, “Does the augmentation index of pulse waves truly increase with progression of atherosclerosis? an experimental study with hypercholesterolemic rabbits,”American Journal of Hypertension, vol. 26, no. 3, pp. 311–317, 2013

    2013

  19. [19]

    An optimal filter for short photoplethysmogram signals,

    Y . Liang, Z. Chen, R. Ward, and M. Elgendi, “An optimal filter for short photoplethysmogram signals,”Scientific Data, vol. 5, p. 180076, 2018

    2018

  20. [20]

    Skin optical properties in the obese and their relation to body mass index: a review,

    A. J. Rodriguezet al., “Skin optical properties in the obese and their relation to body mass index: a review,”Journal of Biomedical Optics, vol. 27, no. 3, p. 030902, 2022

    2022

  21. [21]

    Effects of aging on changing arterial compliance and left ventricular load in a northern chinese urban community,

    A. P. Avolio, S. G. Chen, R. P. Wang, C. L. Zhang, M. F. Li, and M. F. O’Rourke, “Effects of aging on changing arterial compliance and left ventricular load in a northern chinese urban community,”Circulation, vol. 68, no. 1, pp. 50–58, 1983

    1983

  22. [22]

    Pulsedb: A large, cleaned dataset based on mimic-iii and vitaldb for benchmark- ing cuff-less blood pressure estimation methods,

    W. Wang, P. Mohseni, K. L. Kilgore, and L. Najafizadeh, “Pulsedb: A large, cleaned dataset based on mimic-iii and vitaldb for benchmark- ing cuff-less blood pressure estimation methods,”Frontiers in Digital Health, vol. V olume 4 - 2022, 2023

    2022

  23. [23]

    Pulse arrival time is not an adequate surrogate for pulse transit time as a marker of blood pressure,

    G. Zhang, M. Gao, D. Xu, N. B. Olivier, and R. Mukkamala, “Pulse arrival time is not an adequate surrogate for pulse transit time as a marker of blood pressure,”Journal of Applied Physiology, vol. 111, no. 6, pp. 1681–1686, 2011. PMID: 21960657

    2011

  24. [24]

    Wearable continuous blood pressure monitoring devices based on pulse wave transit time and pulse arrival time: A review,

    Z.-B. Zhou, T.-R. Cui, D. Li, J.-M. Jian, Z. Li, S.-R. Ji, X. Li, J.-D. Xu, H.-F. Liu, Y . Yang, and T.-L. Ren, “Wearable continuous blood pressure monitoring devices based on pulse wave transit time and pulse arrival time: A review,”Materials, vol. 16, no. 6, 2023

    2023

  25. [25]

    Novel deep convolutional neural network for cuff-less blood pressure measurement using ecg and ppg signals,

    C. Yan, Z. Li, W. Zhao, J. Hu, D. Jia, H. Wang, and T. You, “Novel deep convolutional neural network for cuff-less blood pressure measurement using ecg and ppg signals,” in2019 41st Annual International Confer- ence of the IEEE Engineering in Medicine and Biology Society (EMBC), pp. 1917–1920, IEEE, 2019

    1917

  26. [26]

    Multitask learning: A knowledge-based source of inductive bias,

    R. Caruana, “Multitask learning: A knowledge-based source of inductive bias,” inMachine learning: Proceedings of the tenth international conference, pp. 41–48, 1993

    1993

  27. [27]

    Mlp-bp: A novel framework for cuffless blood pressure measurement with ppg and ecg signals based on mlp-mixer neural networks,

    B. Huang, W. Chen, C.-L. Lin, C.-F. Juang, and J. Wang, “Mlp-bp: A novel framework for cuffless blood pressure measurement with ppg and ecg signals based on mlp-mixer neural networks,”Biomedical Signal Processing and Control, vol. 73, p. 103404, 2022

    2022

  28. [28]

    Non-invasive arterial blood pressure measurement and spo2 estimation using ppg signal: a deep learning framework,

    Y . Chu, K. Tang, Y . C. Hsu, T. Huang, D. Wang, W. Li, S. I. Savitz, X. Jiang, and S. Shams, “Non-invasive arterial blood pressure measurement and spo2 estimation using ppg signal: a deep learning framework,”BMC Medical Informatics and Decision Making, vol. 23, no. 1, p. 131, 2023

    2023

  29. [29]

    A review of deep learning methods for photoplethysmography data,

    G. Nie, J. Zhu, G. Tang, D. Zhang, S. Geng, Q. Zhao, and S. Hong, “A review of deep learning methods for photoplethysmography data,” 2024

    2024

  30. [30]

    Non-invasive continuous blood pressure estimation from single-channel ppg based on a temporal convolutional network integrated with an attention mechanism,

    D. Dai, Z. Ji, and H. Wang, “Non-invasive continuous blood pressure estimation from single-channel ppg based on a temporal convolutional network integrated with an attention mechanism,”Applied Sciences, vol. 14, no. 14, p. 6061, 2024

    2024

  31. [31]

    Advancing cuffless blood pressure estimation: A ppg-based multi-task learning model for enhanced feature extraction and fusion,

    H. Xiao, A. Zhao, W. Song, T. Liu, L. Long, Y . Li, and H. Li, “Advancing cuffless blood pressure estimation: A ppg-based multi-task learning model for enhanced feature extraction and fusion,”Biomedical Signal Processing and Control, vol. 95, p. 106378, 2024

    2024

  32. [32]

    Bench- marking and enhancing ppg-based cuffless blood pressure estimation methods,

    N. Mathew, Y . Shen, R. Hu, M. Rahimi, and G. Zouridakis, “Bench- marking and enhancing ppg-based cuffless blood pressure estimation methods,” inProceedings of the 19th International Joint Conference on Biomedical Engineering Systems and Technologies - BIOSIGNALS, pp. 45–55, INSTICC, SciTePress, 2026

    2026

  33. [33]

    Photoplethysmography signal processing and synthesis,

    E. Mejia, J. Allen, K. Budiha, C. El-Hajj, P. Kyriacou, and P. Charlton, “Photoplethysmography signal processing and synthesis,”Photoplethys- mographyJ., Eds.; Elsevier, 2021

    2021

  34. [34]

    Clustered photoplethysmogram pulse wave shapes and their associations with clinical data,

    S. Zanelli, K. Eveilleau, P. H. Charlton, M. Ammi, M. Hallab, and M. A. El Yacoubi, “Clustered photoplethysmogram pulse wave shapes and their associations with clinical data,”Frontiers in Physiology, vol. 14, p. 1176753, 2023

    2023

  35. [35]

    Second derivative of the finger arterial pressure waveform: an insight into dynamics of the peripheral arterial pressure pulse,

    J. Simek, D. Wichterle, V . Melenovsky, J. Malik, S. Svacina, and J. Widimsky, “Second derivative of the finger arterial pressure waveform: an insight into dynamics of the peripheral arterial pressure pulse,” Physiological research, vol. 54, no. 5, p. 505, 2005

    2005

  36. [36]

    Utility of second derivative of the finger photoplethysmogram for the estimation of the risk of coronary heart disease in the general population,

    T. Otsuka, T. Kawada, M. Katsumata, and C. Ibuki, “Utility of second derivative of the finger photoplethysmogram for the estimation of the risk of coronary heart disease in the general population,”Circulation Journal, vol. 70, no. 3, pp. 304–310, 2006

    2006

  37. [37]

    A multi- stage deep neural network model for blood pressure estimation using photoplethysmogram signals,

    J. Esmaelpoor, M. H. Moradi, and A. Kadkhodamohammadi, “A multi- stage deep neural network model for blood pressure estimation using photoplethysmogram signals,”Computers in Biology and Medicine, vol. 120, p. 103719, 2020

    2020

  38. [38]

    Ability of a 24-h ambulatory cuffless blood pressure monitoring device to track blood pressure changes in clinical practice,

    F. C. Derendinger, A. S. Vischer, P. Krisai, T. Socrates, C. Schumacher, M. Mayr, and T. Burkard, “Ability of a 24-h ambulatory cuffless blood pressure monitoring device to track blood pressure changes in clinical practice,”Journal of Hypertension, vol. 42, no. 4, pp. 662–671, 2024

    2024

  39. [39]

    Guidance for the interpretation of continual cuffless blood pressure data for the diagnosis and management of hypertension,

    J. Sola, R. Delgado-Gonzalo, R. S. Padwal, G. Parati, A. Bechlioulis, and G. S. Stergiou, “Guidance for the interpretation of continual cuffless blood pressure data for the diagnosis and management of hypertension,” Frontiers in Medical Technology, vol. 4, p. 899143, 2022

    2022