HGCTNet: Handcrafted Feature-Guided CNN and Transformer Network for Wearable Cuffless Blood Pressure Measurement DOI
Zengding Liu, Ye Li, Yuan‐Ting Zhang

et al.

IEEE Journal of Biomedical and Health Informatics, Journal Year: 2024, Volume and Issue: 28(7), P. 3882 - 3894

Published: April 30, 2024

Biosignals collected by wearable devices, such as electrocardiogram and photoplethysmogram, exhibit redundancy global temporal dependencies, posing a challenge in extracting discriminative features for blood pressure (BP) estimation. To address this challenge, we propose HGCTNet, handcrafted feature-guided CNN transformer network cuffless BP measurement based on devices. By leveraging convolutional operations self-attention mechanisms, design CNN-Transformer hybrid architecture to learn from biosignals that capture both local information dependencies. Then, introduce attention module utilizes extracted query vectors eliminate redundant within the learned features. Finally, feature fusion integrates features, demographics enhance model performance. We validate our approach using two large datasets: CAS-BP dataset Aurora-BP dataset. Experimental results demonstrate HGCTNet achieves an estimation error of 0.9 $\pm$ 6.5 mmHg diastolic (DBP) 0.7 8.3 systolic (SBP) On dataset, corresponding errors are notation="LaTeX">$-$ 0.4 7.0 DBP 8.6 SBP. Compared current state-of-the-art approaches, reduces mean absolute SBP 10.68% 9.84% These highlight potential improving performance measurements. The source code available at https://github.com/zdzdliu/HGCTNet.

Language: Английский

A novel art of continuous noninvasive blood pressure measurement DOI Creative Commons
Jürgen Fortin,

Dorothea E. Rogge,

Christian Fellner

et al.

Nature Communications, Journal Year: 2021, Volume and Issue: 12(1)

Published: March 2, 2021

Abstract Wearable sensors to continuously measure blood pressure and derived cardiovascular variables have the potential revolutionize patient monitoring. Current wearable methods analyzing time components (e.g., pulse transit time) still lack clinical accuracy, whereas existing technologies for direct measurement are too bulky. Here we present an innovative art of continuous noninvasive hemodynamic monitoring (CNAP2GO). It directly measures by using a volume control technique could be used small integrated in finger-ring. As software prototype, CNAP2GO showed excellent performance comparison with invasive reference measurements 46 patients having surgery. The resulting pulsatile signal carries information derive cardiac output other variables. We show that can self-calibrate miniaturized approaches. potentially constitutes breakthrough flow both ambulatory in-hospital settings.

Language: Английский

Citations

69
Virtual management of hypertension: lessons from the COVID-19 pandemic–International Society of Hypertension position paper endorsed by the World Hypertension League and European Society of Hypertension DOI Open Access
Nadia Khan, George S. Stergiou, Stefano Omboni

et al.

Journal of Hypertension, Journal Year: 2022, Volume and Issue: 40(8), P. 1435 - 1448

Published: May 17, 2022

The coronavirus disease 2019 pandemic caused an unprecedented shift from in person care to delivering healthcare remotely. To limit infectious spread, patients and providers rapidly adopted distant evaluation with online or telephone-based diagnosis management of hypertension. It is likely that virtual chronic diseases including hypertension will continue some form into the future. purpose International Society Hypertension's (ISH) position paper provide practical guidance on improve its blood pressure control based currently available evidence international experts' opinion for nonpregnant adults. Virtual represents provision services at a distance communication conducted between providers, users their circle care. This statement provides consensus on: selecting monitoring devices, accurate home assessments, patient education virtually, health behavior modification, medication adjustment long-term monitoring. We further recommendations modalities assessment across spectrum resource availability ability.

Language: Английский

Citations

41
Emerging sensing and modeling technologies for wearable and cuffless blood pressure monitoring DOI Creative Commons
Lei Zhao, Cunman Liang, Yan Huang

et al.

npj Digital Medicine, Journal Year: 2023, Volume and Issue: 6(1)

Published: May 22, 2023

Abstract Cardiovascular diseases (CVDs) are a leading cause of death worldwide. For early diagnosis, intervention and management CVDs, it is highly desirable to frequently monitor blood pressure (BP), vital sign closely related during people’s daily life, including sleep time. Towards this end, wearable cuffless BP extraction methods have been extensively researched in recent years as part the mobile healthcare initiative. This review focuses on enabling technologies for monitoring platforms, covering both emerging flexible sensor designs algorithms. Based signal type, sensing devices classified into electrical, optical, mechanical sensors, state-of-the-art material choices, fabrication methods, performances each type briefly reviewed. In model review, contemporary algorithmic estimation beat-to-beat measurements continuous waveform introduced. Mainstream approaches, such pulse transit time-based analytical models machine learning compared terms their input modalities, features, implementation algorithms, performances. The sheds light interdisciplinary research opportunities combine latest innovations processing fields achieve new generation measurement with improved wearability, reliability, accuracy.

Language: Английский

Citations

39
Personalized Machine Learning-Coupled Nanopillar Triboelectric Pulse Sensor for Cuffless Blood Pressure Continuous Monitoring DOI

Chuanjie Yao,

Tiancheng Sun,

Shuang Huang

et al.

ACS Nano, Journal Year: 2023, Volume and Issue: 17(23), P. 24242 - 24258

Published: Nov. 20, 2023

A wearable system that can continuously track the fluctuation of blood pressure (BP) based on pulse signals is highly desirable for treatments cardiovascular diseases, yet sensitivity, reliability, and accuracy remain challenging. Since correlations waveforms to BP are individualized due diversity patients' physiological characteristics, sensors universal designs algorithms often fail derive accurately when applied individual patients. Herein, a triboelectric sensor biomimetic nanopillar layer was developed coupled with Personalized Machine Learning (ML) provide accurate continuous monitoring BP. Flexible conductive nanopillars as were fabricated through soft lithography replication cicada wing, which could effectively enhance sensor's output performance detect weak signal characteristics waveform derivation. The personalized Partial Least-Squares Regression (PLSR) ML unknown reasonable accuracy, avoiding issue variability encountered by General PLSR or formula algorithms. cuffless intelligent design endow this ML-sensor promising platform care hypertensive

Language: Английский

Citations

25
Recent Progress in Wearable Self‐Powered Biomechanical Sensors: Mechanisms and Applications DOI
Shaotong Zhang, Xiang Lin, Ji Wan

et al.

Advanced Materials Technologies, Journal Year: 2024, Volume and Issue: 9(21)

Published: Feb. 7, 2024

Abstract Biomechanical signals, such as strain variations of the skin, vibrations chest and throat, well motions limbs, hold immense significance in healthcare monitoring, disease diagnosis, human‐machine interface. Examples span from monitoring blood pressure pulse waves for atherosclerosis to distinguishing between metatarsalgia patients healthy individuals by tracking their walking postures, voiceprint recognition hearing aid technology based on vibration sensing. Wearable biomechanical sensors play a crucial role providing valuable insights into one's health condition physiological features. However, development high‐performance capable prolonged poses challenges. Traditional batteries have limited lifespan pose difficulty replacement. Using self‐powered devices measurement signals represents an attractive solution tackle issues caused batteries. This review focuses mechanisms wearable sensors, delves recent advancements applications, covering areas cardiovascular system acoustic detection, human motion tracking, many others associated with biomechanics. A concluding section outlines potential future prospects this evolving field materials biomedical research.

Language: Английский

Citations

15
Cuffless Blood Pressure in clinical practice: challenges, opportunities and current limits. DOI Creative Commons
Benoît Henry,

Maxime Merz,

Harry Hoang

et al.

Blood Pressure, Journal Year: 2024, Volume and Issue: 33(1)

Published: Jan. 21, 2024

Background: Cuffless blood pressure measurement technologies have attracted significant attention for their potential to transform cardiovascular monitoring.

Language: Английский

Citations

10
Using machine learning models for cuffless blood pressure estimation with ballistocardiogram and impedance plethysmogram DOI Creative Commons
Shing-Hong Liu, Yao Sun,

Bo-Yan Wu

et al.

Frontiers in Digital Health, Journal Year: 2025, Volume and Issue: 7

Published: Feb. 21, 2025

Introduction Blood pressure (BP) serves as a crucial parameter in the management of three prevalent chronic diseases, hypertension, cardiovascular and cerebrovascular diseases. However, conventional sphygmomanometer, utilizing cuff, is unsuitable for approach mobile health (mHealth). Methods Cuffless blood measurement, which eliminates need considered promising avenue. This method based on relationship between pulse arrival time (PAT) parameters BP. In this study, transit (PTT) was derived from ballistocardiograms (BCG) impedance plethysmograms (IPG) obtained weight-fat scale. study aims to address two challenges using deep learning machine technologies: first, identifying BCG IPG signals with good quality, then extracting PTT them estimate A stacked model comprising one-dimensional convolutional neural network (1D CNN) gated recurrent unit (GRU) proposed classify quality signals. Seven parameters, including calibration-based calibration-free heart rate (HR), were examined BP random forest (RF) XGBoost models. Seventeen healthy subjects participated their elevated through exercise. digital sphygmomanometer employed measure reference values. Our methodology validated data collected our custom-made device. Results The results demonstrated signal classification accuracy 0.989. Furthermore, five-fold cross-validation, Pearson correlation coefficients 0.953 ± 0.007 0.935 achieved systolic (SBP) diastolic (DBP) estimations, respectively. mean absolute differences (MADs) calculated 3.54 0.34 2.57 0.17 mmHg SBP DBP, Discussion significantly improved cuffless indicating its potential integration into scales an unconstrained device effective utilization mHealth applications.

Language: Английский

Citations

1
Advances in Non-Invasive Blood Pressure Monitoring DOI Creative Commons

Xina Quan,

Junjun Liu,

Thomas Roxlo

et al.

Sensors, Journal Year: 2021, Volume and Issue: 21(13), P. 4273 - 4273

Published: June 22, 2021

This paper reviews recent advances in non-invasive blood pressure monitoring and highlights the added value of a novel algorithm-based sensor which uses machine-learning techniques to extract values from shape pulse waveform. We report results preliminary studies on range patient populations discuss accuracy limitations this capacitive-based technology its potential application hospitals communities.

Language: Английский

Citations

50
A Review of Noninvasive Methodologies to Estimate the Blood Pressure Waveform DOI Creative Commons
Tasbiraha Athaya, Sunwoong Choi

Sensors, Journal Year: 2022, Volume and Issue: 22(10), P. 3953 - 3953

Published: May 23, 2022

Accurate estimation of blood pressure (BP) waveforms is critical for ensuring the safety and proper care patients in intensive units (ICUs) intraoperative hemodynamic monitoring. Normal cuff-based BP measurements can only provide systolic (SBP) diastolic (DBP). Alternatively, waveform be used to estimate a variety other physiological parameters provides additional information about patient’s health. As result, various techniques are being proposed accurately estimating waveforms. The purpose this review summarize current state knowledge regarding waveform, three methodologies (pressure-based, ultrasound-based, deep-learning-based) noninvasive research feasibility employing these strategies at home as well ICUs. Additionally, article will discuss physical concepts underlying both invasive measurements. We historical measurements, standard clinical procedures, more recent innovations Although technique has not been validated, it expected that precise, available near future due its enormous potential.

Language: Английский

Citations

37
Blood pressure monitoring via double sandwich-structured triboelectric sensors and deep learning models DOI
Ran Xu,

Fangyuan Luo,

Zhiming Lin

et al.

Nano Research, Journal Year: 2022, Volume and Issue: 15(6), P. 5500 - 5509

Published: March 15, 2022

Language: Английский

Citations

33