scholarly journals Non-Invasive Hemodynamics Monitoring System Based on Electrocardiography via Deep Convolutional Autoencoder

Sensors ◽  
2021 ◽  
Vol 21 (18) ◽  
pp. 6264
Author(s):  
Muammar Sadrawi ◽  
Yin-Tsong Lin ◽  
Chien-Hung Lin ◽  
Bhekumuzi Mathunjwa ◽  
Ho-Tsung Hsin ◽  
...  
Keyword(s):  
Blood Pressure ◽  
Mean Squared Error ◽  
Massachusetts General Hospital ◽  
Venous Pressure ◽  
Absolute Error ◽  
Cerebral Hemodynamics ◽  
Non Invasive ◽  
Arterial Blood ◽  
Convolutional Autoencoder ◽  
The Mean

This study evaluates cardiovascular and cerebral hemodynamics systems by only using non-invasive electrocardiography (ECG) signals. The Massachusetts General Hospital/Marquette Foundation (MGH/MF) and Cerebral Hemodynamic Autoregulatory Information System Database (CHARIS DB) from the PhysioNet database are used for cardiovascular and cerebral hemodynamics, respectively. For cardiovascular hemodynamics, the ECG is used for generating the arterial blood pressure (ABP), central venous pressure (CVP), and pulmonary arterial pressure (PAP). Meanwhile, for cerebral hemodynamics, the ECG is utilized for the intracranial pressure (ICP) generator. A deep convolutional autoencoder system is applied for this study. The cross-validation method with Pearson’s linear correlation (R), root mean squared error (RMSE), and mean absolute error (MAE) are measured for the evaluations. Initially, the ECG is used to generate the cardiovascular waveform. For the ABP system—the systolic blood pressure (SBP) and diastolic blood pressures (DBP)—the R evaluations are 0.894 ± 0.004 and 0.881 ± 0.005, respectively. The MAE evaluations for SBP and DBP are, respectively, 6.645 ± 0.353 mmHg and 3.210 ± 0.104 mmHg. Furthermore, for the PAP system—the systolic and diastolic pressures—the R evaluations are 0.864 ± 0.003 mmHg and 0.817 ± 0.006 mmHg, respectively. The MAE evaluations for systolic and diastolic pressures are, respectively, 3.847 ± 0.136 mmHg and 2.964 ± 0.181 mmHg. Meanwhile, the mean CVP evaluations are 0.916 ± 0.001, 2.220 ± 0.039 mmHg, and 1.329 ± 0.036 mmHg, respectively, for R, RMSE, and MAE. For the mean ICP evaluation in cerebral hemodynamics, the R and MAE evaluations are 0.914 ± 0.003 and 2.404 ± 0.043 mmHg, respectively. This study, as a proof of concept, concludes that the non-invasive cardiovascular and cerebral hemodynamics systems can be potentially investigated by only using the ECG signal.

scholarly journals An Estimation Method of Continuous Non-Invasive Arterial Blood Pressure Waveform Using Photoplethysmography: A U-Net Architecture-Based Approach

Sensors ◽  
2021 ◽  
Vol 21 (5) ◽  
pp. 1867
Author(s):  
Tasbiraha Athaya ◽  
Sunwoong Choi
Keyword(s):  
Blood Pressure ◽  
Cardiovascular Health ◽  
Estimation Method ◽  
Absolute Error ◽  
British Hypertension Society ◽  
Non Invasive ◽  
Arterial Blood ◽  
Blood Pressure Waveform ◽  
The Mean ◽  
Mimic Iii

Blood pressure (BP) monitoring has significant importance in the treatment of hypertension and different cardiovascular health diseases. As photoplethysmogram (PPG) signals can be recorded non-invasively, research has been highly conducted to measure BP using PPG recently. In this paper, we propose a U-net deep learning architecture that uses fingertip PPG signal as input to estimate arterial BP (ABP) waveform non-invasively. From this waveform, we have also measured systolic BP (SBP), diastolic BP (DBP), and mean arterial pressure (MAP). The proposed method was evaluated on a subset of 100 subjects from two publicly available databases: MIMIC and MIMIC-III. The predicted ABP waveforms correlated highly with the reference waveforms and we have obtained an average Pearson’s correlation coefficient of 0.993. The mean absolute error is 3.68 ± 4.42 mmHg for SBP, 1.97 ± 2.92 mmHg for DBP, and 2.17 ± 3.06 mmHg for MAP which satisfy the requirements of the Association for the Advancement of Medical Instrumentation (AAMI) standard and obtain grade A according to the British Hypertension Society (BHS) standard. The results show that the proposed method is an efficient process to estimate ABP waveform directly using fingertip PPG.

Vasomotor waves of arterial blood pressure after cessation of cardiopulmonary bypass in man

Perfusion ◽  
1990 ◽  
Vol 5 (4) ◽  
pp. 261-266
Author(s):  
V. Vainionpää ◽  
A. Hollme'n ◽  
J. Timisjärvi
Keyword(s):  
Blood Pressure ◽  
Cardiopulmonary Bypass ◽  
Arterial Pressure ◽  
Venous Pressure ◽  
Systemic Arterial Pressure ◽  
Heart Patients ◽  
Arterial Blood ◽  
Pulmonary Arterial ◽  
The Mean ◽  
Epicardial Pacing

The occurrence of vasomotor waves during cardiopulmonary bypass (CPB) is a recognized phenomenon. The lesser known oscillation of arterial pressure after cessation of CPB was observed in 18 open-heart patients. The duration of an oscillatory wave was 13.5±5.0 seconds, the amplitude 6.1 ±2.6mmNg and the mean arterial pressure 76.5± 10.7mmHg. Inter-and also intraindividual variations in frequency and amplitude of the oscillation, however, did occur. In 13 patients, this oscillation occurred during ventricular epicardial pacing. The oscillation continued until the end of the operation in eight patients; in others, the oscillation was of shorter duration. An oscillation of pulmonary arterial pressure (PAP) was simultaneously observed in nine patients (eight with pacemaker) and central venous pressure (CVP) oscillation in eight patients (all with pacemaker). The duration of a wave was the same as in systemic arterial pressure and the amplitudes were 1.5-3.0mmHg in PAP and 1.0-2.0mmHg in CVP. These arterial vasomotor waves, seen here after CPB, largely resemble those observed during perfusion in man and also the Mayerwaves explored in experimental animals. The pacing rhythm seems to favourthe appearance of those blood pressure oscillations.

Haemodynamic Monitoring in Theatre

2003 ◽  
Vol 4 (1) ◽  
pp. 10-16 ◽  
Author(s):  
Heidi Clinton
Keyword(s):  
Blood Pressure ◽  
Pulmonary Artery ◽  
Central Venous Pressure ◽  
Venous Pressure ◽  
Haemodynamic Monitoring ◽  
Pressure Monitoring ◽  
Central Venous ◽  
Noninvasive Measurements ◽  
Non Invasive ◽  
Arterial Blood

AbstractThe number of devices available to monitor the haemodynamic status of patients is increasing. Practitioners need to be aware of the non-invasive and invasive methods available in order to care for their patients safely and effectively. This article reviews a number of noninvasive measurements of haemodynamic function, in addition to invasive methods such as arterial blood pressure, central venous pressure and pulmonary artery pressure monitoring. It is argued that using these methods in combination provides a comprehensive haemodynamic assessment.

scholarly journals Tracking blood pressure changes by means of non-invasive intermittent blood pressure measurements in clinical application

2021 ◽  
Vol 7 (2) ◽  
pp. 815-818
Author(s):  
Robert Huhle ◽  
Thorsten Richter ◽  
Marcelo Gama de Abreu
Keyword(s):  
Blood Pressure ◽  
Arterial Blood Pressure ◽  
Pressure Range ◽  
Tracking Error ◽  
Correlation Coefficients ◽  
Non Invasive ◽  
Arterial Blood ◽  
The Mean ◽  
Blood Pressure Range ◽  
Pressure Changes

Abstract Considering accuracy/precision cut-offs of 5 ± 8 mmHg and cut-off values for inter-class correlation coefficients (ICC=0.37...1, from DIN EN ISO 81060-2), absolute and relative errors in time independent measurement of blood pressure changes with non-invasive intermittent devices (NiBP) are derived mathematically for mean arterial blood pressure range of 40-180 mmHg. As a clinically relevant value for change of arterial blood pressure 20% of the baseline blood pressure is considered. The mean ratio between the change of BP measured by the NiBP and measured by the invasive reference device (TE%) were proposed as quality measure for the evaluation of NiBP device tracking capability. The proposed measure TE%is theoretically independent of absolute accuracy but depends on precision and ICC of a device. NiBP devices show considerable maximum TE% of 41% in tracking mean blood pressure changes respectively. In 10% of the measurements in the low blood pressure range TE% exceeding 100%. The mean 50th/90th TE% percentile over the whole blood pressure range were 25/61%, respectively. Furthermore, TE% was relatively insensitive to assumed blood pressure range but sensitive to ICC. NiBP devices have high relative error in tracking blood pressure changes that make those devices not well-suited for tracking blood pressure changes. The proposed tracking error allows the definition of reasonable accuracy/precision requirements of NBP devices.

Blood pressure prediction by exploiting informative features from ICU patients’ ECG and PPG signals under a heterogeneous ensemble learning framework

2020 ◽  
Author(s):  
Keke Qin ◽  
Guobiao Xu ◽  
Jun Huang
Keyword(s):  
Blood Pressure ◽  
Ensemble Learning ◽  
Absolute Error ◽  
Physiological Signal ◽  
Intensive Care Patients ◽  
Learning Framework ◽  
Non Invasive ◽  
Multiple Regression Models ◽  
The Mean ◽  
Heterogeneous Ensemble

Abstract BackgroundAlthough invasive methods are currently used to monitor blood pressure (BP) for intensive care patients, accurate and timely non-invasive BP monitoring in non-invasive way is still significant. Yet, physiological signal data of patients is irregular, with more noise and abnormal patterns included, making accurate and stable prediction challenging. The traditional BP measurement methods are cuff-based, and the prediction accuracy and stability of the machine learning based cuff-less prediction model needs to be further improved. Additionally, data must be cleaned and effective features must be grubbed from the irregular signals, which is a prerequisite for model training.ResultsIn the present study, we proposed a novel heterogeneous ensemble learning BP prediction (ELBP) model, where: 1) Related features are systematically extracted and selected for systolic, diastolic and mean BP prediction tasks; 2)Then, multiple regression models are trained and then are weighted for final prediction, wherein the weights are learned from data; 3) Hyper-parameters of each model are optimised using Bayesian optimisation based on cross-validation. We experimentally verified the ELBP effectiveness, the mean absolute error of ELBP is 1.802 mmHg, 3.936 mmHg and 3.121 mmHg for diastolic, systolic and mean BP respectively on mimic-1, and 2.722 mmHg, 5.039 mmHg and 3.812 mmHg respectively on mimic-2. Further experiments demonstrated that ELBP performance is superior to state-of-the-art algorithms on seven evaluation metrics.ConclusionIn conclusion, BP prediction precision can be further improved by integrating multiple learners appropriately, and this study is valuable in promoting BP prediction in practical application.

scholarly journals Comparison of Doppler ultrasonic and oscillometric devices (with or without proprietary optimisations) for non-invasive blood pressure measurement in conscious cats

2020 ◽  
pp. 1098612X2093240
Author(s):  
Petra Cerna ◽  
Panos E Archontakis ◽  
Hester OK Cheuk ◽  
Danièlle A Gunn-Moore
Keyword(s):  
Blood Pressure ◽  
Blood Pressure Measurement ◽  
Reference Intervals ◽  
Invasive Blood Pressure ◽  
Non Invasive ◽  
Arterial Blood ◽  
Oscillometric Device ◽  
Significant Difference ◽  
Systolic Arterial Blood Pressure ◽  
The Mean

Objectives This study compared Doppler and oscillometric (PetMAP+) devices (with or without proprietary optimisations) for the non-invasive measurement of blood pressure in conscious cats. Methods Twenty-three cats were enrolled; however, five were excluded as fewer than five measurements were obtained for each assessment. All measurements were obtained according to American College of Veterinary Internal Medicine consensus guidelines. Oscillometric device modes A and B were operated according to the manufacturer’s guidelines. Doppler and oscillometric devices were used alternately as the first device. Results Systolic arterial blood pressure (SAP) measurements were obtained by Doppler (SAPd) and oscillometry; the mean of each set of five values was used for statistical analysis. There was a significant difference between SAPd and SAP measurements in oscillometric modes A ( P <0.001) and B ( P <0.001). While both modes measured SAP higher than SAPd, B had a smaller bias (+15.72 mmHg) and narrower limits of agreement (LOA). There was also a significant difference between SAPd and mean arterial pressure (MAP) on oscillometric modes A ( P = 0.002) and B ( P <0.001). Both modes’ MAP readings were lower than SAPd and oscillometric A MAP was closer to SAPd (–14.94 mmHg), with a smaller bias and narrower LOA. Conclusions and relevance The findings support that Doppler and oscillometric devices cannot be used interchangeably, with or without proprietary optimisations. Methodology should always be taken into account and reference intervals (RIs) need to be defined for the different methodologies. Until methodology-specific RIs are published, definitive diagnosis of hypertension and sub-staging of patients with kidney disease according to the International Renal Interest Society guidelines remains challenging.

scholarly journals Non-invasive continuous blood pressure monitoring (ClearSight™ system) during shoulder surgery in the beach chair position: a prospective self-controlled study

2020 ◽  
Vol 20 (1) ◽  
Author(s):  
Konrad Chachula ◽  
Florian Lieb ◽  
Florian Hess ◽  
Joellen Welter ◽  
Nicole Graf ◽  
...  
Keyword(s):  
Blood Pressure ◽  
Blood Pressure Monitoring ◽  
Shoulder Surgery ◽  
Controlled Study ◽  
Pressure Monitoring ◽  
Non Invasive ◽  
Arterial Blood ◽  
Beach Chair Position ◽  
The Mean ◽  
Beach Chair

Abstract Background The beach chair position that is commonly used in shoulder surgery is associated with relative hypovolemia, which leads to a reduction in arterial blood pressure. The effects of patient positioning on the accuracy of non-invasive continuous blood pressure monitoring with the ClearSight™ system (CS-BP; Edwards Lifesciences, Irvine CA, USA) have not been studied extensively. Our research aim was to assess agreement levels between CS-BP measurements with traditional blood pressure monitoring techniques. Methods For this prospective self-controlled study, we included 20 consecutively treated adult patients undergoing elective shoulder surgery in the beach chair position. We performed Bland-Altman analyses to determine agreement levels between blood pressure values from CS-BP and standard non-invasive (NIBP) methods. Perioperative measurements were done in both the supine (as reference) and beach chair surgical positions. Additionally, we compared invasive blood pressure (IBP) measurements with both the non-invasive methods (CS-BP and NIBP) in a sub-group of patients (n = 10) who required arterial blood pressure monitoring. Results We analyzed 229 data points (116 supine, 113 beach chair) from the entire cohort; per patient measurements were based on surgical length (range 3–9 supine, 2–10 beach chair). The mean difference (±SD; 95% limits of agreement) in the mean arterial pressure (MAP) between CS-BP and NIBP was − 0.9 (±11.0; − 24.0–22.2) in the beach chair position and − 4.9 mmHg (±11.8; − 28.0–18.2) when supine. In the sub-group, the difference between CS-BP and IBP in the beach chair position was − 1.6 mmHg (±16.0; − 32.9–29.7) and − 2.8 mmHg (±15.3; − 32.8–27.1) in the supine position. Between NIBP and IBP, we detected a difference of 3.0 mmHg (±9.1; − 20.8–14.7) in the beach chair position, and 4.6 mmHg (±13.3; − 21.4–30.6) in the supine position. Conclusions We found clinically acceptable mean differences in MAP measurements between the ClearSight™ and non-invasive oscillometric blood pressure systems when patients were in either the supine or beach chair position. For all comparisons of the monitoring systems and surgical positions, the standard deviations and limits of agreement were wide. Trial registration This study was prospectively registered at the German Clinical Trial Register (www.DRKS.de; DRKS00013773). Registered 26/01/2018.

scholarly journals Non-invasive cuff-less blood pressure estimation using a hybrid deep learning model

2021 ◽  
Vol 53 (2) ◽  
Author(s):  
Sen Yang ◽  
Yaping Zhang ◽  
Siu-Yeung Cho ◽  
Ricardo Correia ◽  
Stephen P. Morgan
Keyword(s):  
Machine Learning ◽  
Blood Pressure ◽  
Deep Learning ◽  
Absolute Error ◽  
Learning Model ◽  
Physiological Measurement ◽  
British Hypertension Society ◽  
Non Invasive ◽  
Blood Pressure Estimation ◽  
Deep Learning Model

AbstractConventional blood pressure (BP) measurement methods have different drawbacks such as being invasive, cuff-based or requiring manual operations. There is significant interest in the development of non-invasive, cuff-less and continual BP measurement based on physiological measurement. However, in these methods, extracting features from signals is challenging in the presence of noise or signal distortion. When using machine learning, errors in feature extraction result in errors in BP estimation, therefore, this study explores the use of raw signals as a direct input to a deep learning model. To enable comparison with the traditional machine learning models which use features from the photoplethysmogram and electrocardiogram, a hybrid deep learning model that utilises both raw signals and physical characteristics (age, height, weight and gender) is developed. This hybrid model performs best in terms of both diastolic BP (DBP) and systolic BP (SBP) with the mean absolute error being 3.23 ± 4.75 mmHg and 4.43 ± 6.09 mmHg respectively. DBP and SBP meet the Grade A and Grade B performance requirements of the British Hypertension Society respectively.

scholarly journals On the Pathology of the dropsy produced by obstruction of the superior and inferior venœ and the portal vein.— Preliminary communication

1907 ◽  
Vol 79 (532) ◽  
pp. 267-283 ◽  
Keyword(s):  
Blood Pressure ◽  
Arterial Pressure ◽  
Portal Vein ◽  
Mean Arterial Blood Pressure ◽  
Normal Limit ◽  
The Body ◽  
Diastolic Filling ◽  
Arterial Blood ◽  
Preliminary Communication ◽  
The Mean

In August, 1903, I published a paper in the ‘Journal of Pathology’(1) in which I demonstrated a method experimentally producing uncompensated hear disease in an animal, which was compatible with life. This method consisted in diminishing the size of the pericardial sac by stitches, so that the diastolic filling of the heart was impeded. The main symptoms of this condition were dropsy and diminution in the amount of urine excreted. As the immediate result of this interference with the action of the heart, there occurred a rise of pressure throughout the whole systemic venous system extending as far back as the capillaries, and a fall of the mean arterial blood-pressure. Further, I found that the pressure in all the veins fell to the normal limit again within the space of about one hour, and that subsequently when dropsy was being produced, the vanous pressure in all parts of the body was normal, and the arterial pressure had almost recovered itself.

scholarly journals Continuous Blood Pressure Estimation Using Exclusively Photopletysmography by LSTM-Based Signal-to-Signal Translation

Sensors ◽  
2021 ◽  
Vol 21 (9) ◽  
pp. 2952
Author(s):  
Latifa Nabila Harfiya ◽  
Ching-Chun Chang ◽  
Yung-Hui Li
Keyword(s):  
Blood Pressure ◽  
Absolute Error ◽  
Estimation Methods ◽  
British Hypertension Society ◽  
Arterial Blood ◽  
Proposed Model ◽  
Pressure Estimation ◽  
Effective Signal ◽  
Blood Pressure Estimation ◽  
Continuous Blood Pressure

Monitoring continuous BP signal is an important issue, because blood pressure (BP) varies over days, minutes, or even seconds for short-term cases. Most of photoplethysmography (PPG)-based BP estimation methods are susceptible to noise and only provides systolic blood pressure (SBP) and diastolic blood pressure (DBP) prediction. Here, instead of estimating a discrete value, we focus on different perspectives to estimate the whole waveform of BP. We propose a novel deep learning model to learn how to perform signal-to-signal translation from PPG to arterial blood pressure (ABP). Furthermore, using a raw PPG signal only as the input, the output of the proposed model is a continuous ABP signal. Based on the translated ABP signal, we extract the SBP and DBP values accordingly to ease the comparative evaluation. Our prediction results achieve average absolute error under 5 mmHg, with 70% confidence for SBP and 95% confidence for DBP without complex feature engineering. These results fulfill the standard from Association for the Advancement of Medical Instrumentation (AAMI) and the British Hypertension Society (BHS) with grade A. From the results, we believe that our model is applicable and potentially boosts the accuracy of an effective signal-to-signal continuous blood pressure estimation.

Sign in / Sign up

Export Citation Format

Share Document