scholarly journals Cuffless blood pressure monitoring from a wristband with calibration-free algorithms for sensing location based on bio-impedance sensor array and autoencoder

2022 ◽  
Vol 12 (1) ◽  
Author(s):  
Bassem Ibrahim ◽  
Roozbeh Jafari
Keyword(s):  
Blood Pressure ◽  
Form Factor ◽  
Sensor Array ◽  
Pulse Signal ◽  
Wearable Devices ◽  
Activity Measurement ◽  
Arterial Pulse ◽  
Monitoring Method ◽  
Adaptive Boosting ◽  
Calibration Methods

AbstractContinuous monitoring of blood pressure (BP) is essential for the prediction and the prevention of cardiovascular diseases. Cuffless BP methods based on non-invasive sensors integrated into wearable devices can translate blood pulsatile activity into continuous BP data. However, local blood pulsatile sensors from wearable devices suffer from inaccurate pulsatile activity measurement based on superficial capillaries, large form-factor devices and BP variation with sensor location which degrade the accuracy of BP estimation and the device wearability. This study presents a cuffless BP monitoring method based on a novel bio-impedance (Bio-Z) sensor array built in a flexible wristband with small-form factor that provides a robust blood pulsatile sensing and BP estimation without calibration methods for the sensing location. We use a convolutional neural network (CNN) autoencoder that reconstructs an accurate estimate of the arterial pulse signal independent of sensing location from a group of six Bio-Z sensors within the sensor array. We rely on an Adaptive Boosting regression model which maps the features of the estimated arterial pulse signal to systolic and diastolic BP readings. BP was accurately estimated with average error and correlation coefficient of 0.5 ± 5.0 mmHg and 0.80 for diastolic BP, and 0.2 ± 6.5 mmHg and 0.79 for systolic BP, respectively.

scholarly journals Validation Study to Determine the Accuracy of Central Blood Pressure Measurement Using the Sphygmocor Xcel Cuff Device

Hypertension ◽  
2020 ◽  
Vol 76 (1) ◽  
pp. 244-250 ◽  
Author(s):  
Martin G. Schultz ◽  
Dean S. Picone ◽  
Matthew K. Armstrong ◽  
J. Andrew Black ◽  
Nathan Dwyer ◽  
...  
Keyword(s):  
Blood Pressure ◽  
Form Factor ◽  
Mean Arterial Pressure ◽  
Arterial Pressure ◽  
Validation Study ◽  
Blood Pressure Measurement ◽  
Mean Difference ◽  
Total N ◽  
Acceptable Error ◽  
Calibration Methods

Numerous devices purport to measure central (aortic) blood pressure (BP) as distinct from conventional brachial BP. This validation study aimed to determine the accuracy of the Sphygmocor Xcel cuff device (AtCor Medical, CardieX, Sydney, Australia) for measuring central BP. 296 patients (mean age 61±12 years) undergoing coronary angiography had simultaneous measurement of invasive central BP and noninvasive cuff-derived central BP using the Xcel cuff device (total n=558 individual comparisons). A subsample (n=151) also had invasive brachial BP measured. Methods were undertaken according to the Artery Society recommendations, and several calibration techniques to derive central systolic BP (SBP) were examined. Minimum acceptable error was ≤5±≤8 mm Hg. Central SBP was significantly underestimated, and with wide variability, when using the default calibration of brachial-cuff SBP and diastolic BP (DBP; mean difference±SD, −7.7±11.0 mm Hg). Similar variability was observed using other calibration methods (cuff 33% form-factor mean arterial pressure and DBP, −4.4±11.5 mm Hg; cuff 40% form-factor mean arterial pressure and DBP, 4.7±11.9 mm Hg; cuff oscillometric mean arterial pressure and DBP, −18.2±12.1 mm Hg). Only calibration with invasive central integrated mean arterial pressure and DBP was within minimal acceptable error (3.3±7.5 mm Hg). The difference between brachial-cuff SBP and invasive central SBP was 3.3±10.7 mm Hg. A subsample analysis to determine the accuracy of central-to-brachial SBP amplification showed this to be overestimated by the Xcel cuff device (mean difference 4.3±9.1 mm Hg, P =0.02). Irrespective of cuff calibration technique, the Sphygmocor Xcel cuff device does not meet the Artery Society accuracy criteria for noninvasive measurement of central BP.

Central blood pressure monitoring method oriented to wearable devices

2016 ◽  
Vol 6 (3) ◽  
pp. 197-204 ◽  
Author(s):  
Germán Fierro ◽  
Fernando Silveira ◽  
Ricardo Armentano
Keyword(s):  
Blood Pressure ◽  
Blood Pressure Monitoring ◽  
Wearable Devices ◽  
Central Blood Pressure ◽  
Pressure Monitoring ◽  
Monitoring Method

Monitoring the Cardiovascular Changes of a Rabbit Caused by Phenylephrine via a Microfluidic-Based Tactile Sensor

2019 ◽  
Author(s):  
Dan Wang ◽  
Frank A. Lattanzio ◽  
Mario C. Rodriguez ◽  
Zhili Hao
Keyword(s):  
Blood Pressure ◽  
Elastic Modulus ◽  
Pulse Signal ◽  
High Sensitivity ◽  
Tactile Sensor ◽  
Arterial Pulse ◽  
Record Blood Pressure ◽  
Baseline Values ◽  
Low Sensitivity ◽  
Model Based Analysis

Abstract In this work, a microfluidic-based tactile sensor was investigated for monitoring changes in the cardiovascular (CV) system of a rabbit caused by phenylephrine. The sensor was fixed on the front right leg of an anesthetized rabbit to measure the arterial pulse signal. Phenylephrine, as a vasoconstrictor, was used to introduce CV changes of the rabbit. Two sensors, one with high sensitivity and the other with low sensitivity, were tested on their suitability for measuring the pulse signals of the rabbit. The sensor with low sensitivity generated clear pulse signals and was further used to monitor the CV changes of the rabbit caused by phenylephrine. An automated sphygmomanometer and an ECG were used to record blood pressure and heart rate for comparison. Three low-dose injections of phenylephrine were sequentially performed on the rabbit. Through model-based analysis of the measured pulse signals, arterial elastic modulus, arterial radius and pulse wave velocity (PWV) were obtained. As compared with the baseline values measured before injection, injections of phenylephrine caused an increase in mean blood pressure (MAP) recorded by the medical instruments, and a decrease in arterial radius (increase in peripheral vascular resistance (PVR)) and an increase in arterial elastic modulus and PWV captured by the tactile sensor. Thus, the tactile sensor was proven to be feasible for monitoring the changes in the CV system caused by phenylephrine.

scholarly journals Calibration methods of new device for human blood pressure measurement

2019 ◽  
Vol 201 ◽  
pp. 06003
Author(s):  
Łukasz Stala ◽  
Krzysztof Tomczuk
Keyword(s):  
Blood Pressure ◽  
Pressure Measurement ◽  
Diastolic Pressure ◽  
Blood Pressure Measurement ◽  
New Device ◽  
Reference Device ◽  
Calibration Methods ◽  
Pressure Maximum ◽  
Parallel Measurement ◽  
Pneumatic Sensor

The paper presents two methods of determining calibration curve of a new device for blood pressure measurement. The device was developed at Wrocław University of Science and Technology. First method is based on parallel measurement of systolic and diastolic pressure measurement with use of reference device such as sphygmomanometer and researched new device with pneumatic sensor equipped with voltage type output. Obtained data (systolic ps and diastolic pd pressure, maximum us and minimum ud voltage) was then used to determine individual pressure-voltage characteristic of the device, which can be represented as a linear equation. Second method is based on substitution of experimentally proved coefficient b with its analytical equivalent extracted from mathematical model of described pneumatic sensor. Described methods were verified experimentally and compared. Metrological parameters of the device were designated.

scholarly journals A Unified Calibration Paradigm for a Better Cuffless Blood Pressure Estimation with Modes of Elastic Tube and Vascular Elasticity

2021 ◽  
Vol 2021 ◽  
pp. 1-12
Author(s):  
Jiang Shao ◽  
Ping Shi ◽  
Sijung Hu
Keyword(s):  
Blood Pressure ◽  
Elastic Tube ◽  
Least Square ◽  
Calibration Methods ◽  
Significant Difference ◽  
Vascular Elasticity ◽  
Point To Point ◽  
Blood Pressure Estimation ◽  
Cuffless Blood Pressure ◽  
Selection Of

Although two modes of elastic tube (ET) and vascular elasticity (VE) have been well explored for cuffless continuous blood pressure (BP) monitoring estimation, the initial calibration with these two models could be derived from different mathematical mechanisms for BP estimation. The study is aimed at evaluating the performance of VE and ET models by means of an advanced point-to-point (aPTP) pairing calibration. The cuff BPs were only taken up while the signals of PPG and ECG were synchronously acquired from individual subjects. Two popular VE models together with one representative ET model were designated to study aPTP as a unified assessment criterion. The VE model has demonstrated the stronger correlation r of 0.89 and 0.86 of SBP and DBP, respectively, and the lower estimated BP error of − 0.01 ± 5.90 (4.55) mmHg and 0.04 ± 4.40 (3.38) mmHg of SBP and DBP, respectively, than the ET model. With the ET model, there is a significant difference between the methods of conventional least-square (LS) calibration and aPTP calibration ( p < 0.05 ). These results showed that the VE model surpasses the ET model under the same uniform calibration. The outcome has been unveiled that the selection of initial calibration methods was vital to work out diastolic BP with the ET model. The study revealed an evident fact about initial sensitivity between the modes of different BP estimation and initial calibration.

Improved Vibration-Model-Based Analysis for Estimation of Arterial Parameters From Noninvasively Measured Arterial Pulse Signals

2020 ◽  
Author(s):  
Md Mahfuzur Rahman ◽  
Najmin Ara Sultana ◽  
Linda Vahala ◽  
Leryn Reynolds ◽  
Zhili Hao
Keyword(s):  
Arterial Wall ◽  
Pulse Wave ◽  
Pulse Signal ◽  
Arterial Pulse ◽  
Post Exercise ◽  
First Order ◽  
Model Based ◽  
Vibration Model ◽  
Key Features ◽  
Model Based Analysis

Abstract With the goal of achieving consistence in interpretation of an arterial pulse signal between its vibration model and its hemodynamic relations and improving its physiological implications in our previous study, this paper presents an improved vibration-model-based analysis for estimation of arterial parameters: elasticity (E), viscosity (η), and radius (r0) at diastolic blood pressure (DBP) of the arterial wall, from a noninvasively measured arterial pulse signal. The arterial wall is modeled as a unit-mass vibration model, and its spring stiffness (K) and damping coefficient (D) are related to arterial parameters. Key features of a measured pulse signal and its first-order and second-order derivatives are utilized to estimate the values of K and D. These key features are then utilized in hemodynamic relations, where their interpretation is consistent with the vibration model, to estimate the value of r0 from K and D. Consequently, E, η, and pulse wave velocity (PWV) are also estimated from K and D. The improved vibration-model-based analysis was conducted on pulse signals of a few healthy subjects measured under two conditions: at-rest and immediately post-exercise. With E, r0, and PWV at-rest as baseline, their changes immediately post-exercise were found to be consistent with the related findings in the literature. Thus, this improved vibration-model-based analysis is validated and contributes to estimation of arterial parameters with better physiological implications, as compared with its previous counterpart.

Advance Monitoring of Blood Pressure and Respiratory Rate Using De-Noising Auto Encoder

2021 ◽  
Author(s):  
Seung-Ho Park ◽  
Kyoung-Su Park
Keyword(s):  
Blood Pressure ◽  
Respiratory Rate ◽  
Vital Sign ◽  
Superficial Temporal Artery ◽  
Vital Signs ◽  
Pulse Signal ◽  
Temporal Artery ◽  
Sinus Arrhythmia ◽  
Vital Signs Monitoring ◽  
Vital Sign Measurement

Abstract As the importance of continuous vital signs monitoring increases, the need for wearable devices to measure vital sign is increasing. In this study, the device is designed to measure blood pressure (BP), respiratory rate (RR), and heartrate (HR) with one sensor. The device is in earphone format and is manufactured as wireless type using Arduino-based bluetooth module. The device measures pulse signal in the Superficial temporal artery using Photoplethysmograghy (PPG) sensor. The device uses the Auto Encoder to remove noise caused by movement, etc., contained in the pulse signal. Extract the feature from the pulse signal and use them for the vital sign measurement. The device is measured using Slope transit time (STT) method for BP and Respiratory sinus arrhythmia (RSA) method for RR. Finally, the accuracy is determined by comparing the vital signs measured through the device with the reference vital signs measured simultaneously.

An overview of cardiovascular disease infection: A comparative analysis of boosting algorithms and some single based classifiers

2020 ◽  
Vol 36 (4) ◽  
pp. 1189-1198
Author(s):  
Nureni Olawale Adeboye ◽  
Olawale Victor Abimbola
Keyword(s):  
Blood Pressure ◽  
Early Detection ◽  
Model Building ◽  
World Health ◽  
Gradient Boosting ◽  
Adaptive Boosting ◽  
Modern Age ◽  
Boosting Algorithms ◽  
Health Organization ◽  
Official Data

Machine learning is a branch of artificial intelligence that helps machines learn from observational data without being explicitly programmed and its methods have been found to be very useful in the modern age for medical diagnosis and for early detection of diseases. According to the World Health Organization, 12 million deaths occur annually due to heart-related diseases. Thus, its early detection and treatment are of interest. This research introduces a better way of improving the timely prediction of cardiovascular diseases in suspected patients by comparing the efficiency of two boosting algorithms with four (4) other single based classifiers on cardiovascular official data. The best model was selected based on performances of 5 different evaluation metrics. From the results, Adaptive boosting is seen to outperform all other algorithms with a classification accuracy of 74.2%, closely followed by gradient boosting. However, gradient boosting was chosen as an acceptable technique because it trains faster than Adaboost with a better precision of 74.9% compared to 74.7% exhibited by Adaboost. Thus boosting algorithms are better predictors compared to single based classifiers with factors of age, systolic blood pressure, weight, cholesterol, height, and diastolic blood pressure as the major contributors to the model building.

Sign in / Sign up

Export Citation Format

Share Document