Blood Pressure Signals Research Articles

Evening binge alcohol disrupts cardiovagal tone and baroreflex function during polysomnographic sleep.

Binge alcohol consumption is associated with increased cardiovascular risk. The effects of evening binge alcohol consumption (i.e. 4-5 beverages within 2 h) on the vagal components of HRV and cardiovagal baroreflex sensitivity (cvBRS) during sleep remain largely equivocal. The present study examined the effects of evening binge alcohol consumption on nocturnal cardiac vagal tone and baroreflex sensitivity during stage N2, slow wave (SWS), and rapid eye movement (REM) sleep. We hypothesized that evening binge drinking would reduce HRV and cvBRS in each sleep stage. Following a familiarization night within the laboratory, twenty-three participants were examined following a night of binge alcohol consumption and a fluid control (randomized, crossover design). A quality nocturnal beat-to-beat blood pressure signal was obtained in both conditions in 16 participants (seven men, nine women; 25 ± 1 years). Binge drinking reduced both the high frequency (HF) and time-domain components (i.e. pNN50 and RMSSD) of HRV in stage N2 sleep, SWS, and REM. In addition, cvBRS up-up (vagal activation) was reduced following binge alcohol consumption in stage N2 (21 ± 3 vs. 15 ± 3 ms/mmHg, p = 0.035) and REM (15[11-28] vs. 11[9-18] ms/mmHg, p = 0.009). Binge alcohol consumption reduced cvBRS down-down (vagal withdrawal) in stage N2 (23 ± 2 vs. 14 ± 2 ms/mmHg, p < 0.001), SWS (20[14-30] vs. 14[9-17] ms/mmHg, p = 0.022), and REM (14[11-24] vs. 10[7-15] ms/mmHg, p = 0.006). Evening binge alcohol consumption disrupts cardiac vagal tone and baroreflex function during nearly all sleep stages. These findings provide mechanistic insight into the potential role of binge drinking and alcohol abuse on cardiovascular risk. Alcohol and Neural Cardiovascular Control in Binge Drinkers, www.clinicaltrials.gov/ct2/show/NCT03567434, NCT03567434.

The FMM Approach to Analyze Biomedical Signals: Theory, Software, Applications and Future

Oscillatory systems arise in the different biological and medical fields. Mathematical and statistical approaches are fundamental to deal with these processes. The Frequency Modulated Mobiüs approach (FMM), reviewed in this paper, is one of these approaches. Little known as it has been recently developed, it solves a variety of exciting questions with real data; some of them, such as the decomposition of the signal into components and their multiple uses, are of general application, others are specific. Among the exciting specific applications is the automatic interpretation of the electrocardiogram signal. In this paper, a summary of the theoretical, statistical and computational properties of the FMM approach are revised. Additionally, as a novelty, the FMM approach’s usefulness for the analysis of blood pressure signals is shown. For the latter, a new robust estimation algorithm is proposed using FMM models with restrictions. The paper ends with a view about challenges for the future.

Evening Binge Alcohol Disrupts Cardiovagal Tone and Baroreflex Function During Polysomnographic Sleep Stages

Evening alcohol consumption has been shown to attenuate nocturnal cardiovagal baroreflex sensitivity (cvBRS) and vagal tone, but the impact of alcohol on cvBRS during specific polysomnographic sleep stages remains unclear. We hypothesized that binge alcohol consumption would decrease nocturnal cardiovagal tone and cvBRS, particularly during slow wave (SWS) and rapid eye movement (REM) sleep due to their reported impact on cardiovascular health. Sixteen volunteers (9 women, 7 men; 25±1 years; 28±1 kg/m2) participated in this study. Utilizing a randomized, crossover design, participants arrived to the laboratory for two testing nights (alcohol vs. fluid control) separated by one month. The alcohol condition simulated a binge drinking episode by consuming a 4-5 drink equivalent within two hours. The alcohol dose was based on National Institute of Alcohol Abuse & Alcoholism (NIAAA) definitions of binge drinking, which includes differences based upon body weight and sex (1g/kg men, 0.85g/kg women). Overnight polysomnography (Natus, Middleton, WI), two-lead electrocardiogram (ECG), and dual-finger plethysmography (Finapres NOVA, Netherlands) were recorded during an 8-hour in-laboratory sleep acquisition period. The nocturnal ECG and continuous blood pressure signals were imported into custom software (WinCPRS, Absolute Aliens, Finland) for heart rate variability (HRV) and cvBRS analyses. A ten-minute period of stable sleep was selected in each participant for each sleep stage, and was verified by a board-certified sleep physician. Each segment of stable sleep was free of any scorable arousal, apneic event, and limb movement. Cardiovagal activity was assessed via fast-Fourier transform and quantified as high frequency (HF; 0.15-0.4 Hz), and cvBRS was assessed using up-up (vagal activation) and down-down (vagal withdrawal) sequence analyses. Evening binge alcohol consumption reduced HF during SWS (1429±279 vs. 698±146 ms2, p>0.001) and REM (1114±238 vs. 599±151 ms2, p=0.022), but not stage II (1619±453 vs. 901±258 ms2, p=0.137). Evening binge alcohol consumption decreased cvBRS during stage II (up-up, 21±3 vs. 15±3 ms/mmHg, p=0.035; down-down, 23±2 vs. 14±2 ms/mmHg, p=0.001) and REM sleep (up-up, 21±4 vs. 15±3 ms/mmHg, p=0.008; down-down, 19±3 vs. 12±2 ms/mmHg, p=0.008), but did not alter cvBRS during SWS (up-up, 21±3 vs. 15±3 ms/mmHg, p=0.224; down-down, 22±2 vs. 17±4 ms/mmHg, p=0.225). Taken together, our results indicate that the deleterious cardiovascular effects of binge alcohol consumption may be present across multiple sleep stages, but particularly during stable REM sleep.

Detection Method and System of the Human Body Characteristic Index Based on TCM.

As traditional Chinese medicine (TCM) has gained more and more recognition in the world, Chinese medicine has also played its important role. However, traditional Chinese medicine equipment is relatively deficient, with insufficient functions and low degree of digitalization. For example, existing auscultation equipment can obtain few human characteristic indicators, which is difficult to meet the needs of reference in traditional Chinese medicine diagnosis. Based on this, this paper designed a human body characteristic index detection system based on the principle of traditional Chinese medicine, which includes respiratory and heartbeat signal acquisition device, meridian and acupoint signal acquisition device, temperature signal acquisition device, pulse and blood pressure acquisition device, processing module, keyword module, and output module. The respiratory and heartbeat signal acquisition device is used to collect the respiratory and heartbeat signal of human body. Meridian acupoint signal acquisition device is used to collect human meridian acupoint radio signals. The temperature signal acquisition device is used to collect the infrared temperature light wave signal of human body. Pulse and blood pressure acquisition devices are used to collect pulse and blood pressure signals. The processing module is used to obtain one or more human body characteristic indicators according to one or more of the respiration and heartbeat signals, meridians and acupoints signals, temperature signals, pulse, and blood pressure, including Qi and blood characteristic indicators, viscera and six meridian characteristic indicators, and temperature characteristic indicators. The keyword corresponding module is used to obtain the corresponding keyword representing the physiological state information of human body according to the one or more human body characteristic indicators. The output module is used to output the human body characteristic index and the key words. It includes the key words of Qi and blood state information, the key words of viscera state information, the key words of Qi and blood state information, etc. The system can be used for serious disease screening, chronic disease management, and risk early warning.

Simulation of ECG, blood pressure and ballistocardiographic signals

The blood flow in human arterial system can be considered as a fluid dynamics problem. Simulation of blood flow will provide a better understanding of the physiology of human body. Simulation studies of blood flow in the diseased condition can help to diagnose the health problem easily and also have many applications in the areas such as surgical planning and design of medical devices. This paper presents a synthetic electrocardiogram (ECG), blood pressure signals (BP) and ballistocardiographic signal (BCG). Dynamical models of electrocardiogram and cardiovascular system are important in medicine because they can be used as approximation of the real patient. An example is the Windkessel model, which is often used for simulation. ECG, BP and BCG signals can be generated with different sampling frequencies, with different noise levels, with different shapes, filters etc. The paper is based on real data (Real data and identification methods can be used to create models), which are then used for models based on coupled oscillators. Models of the above-mentioned signals are generated by a microcontroller, which allows easy control and adjustment of the output signal and other experiments. The presented paper describes a device that was developed and used for educational purposes, especially for biomedical engineering.

A novel art of continuous noninvasive blood pressure measurement

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

Antenatal betamethasone redistributes central blood flows and preferentially augments right ventricular output and pump function in preterm fetal lambs.

The glucocorticosteroid betamethasone, which is routinely administered prior to anticipated preterm birth to enhance maturation of the lungs and the cardiovascular system, has diverse fetal regional blood flow effects ranging from increased pulmonary flow to decreased cerebral flow. The aim of this study was to test the hypothesis that these diverse effects reflect alterations in major central flow patterns that are associated with complementary shifts in left ventricular (LV) and right ventricular (RV) pumping performance. Studies were performed in anesthetized preterm fetal lambs (gestation = 127 ± 1 days, term = 147 days) with (n = 14) or without (n = 12) preceding betamethasone treatment via maternal intramuscular injection. High-fidelity central arterial blood pressure and flow signals were obtained to calculate LV and RV outputs and total hydraulic power. Betamethasone therapy was accompanied by 1) increased RV, but not LV, output; 2) a greater RV than LV increase in total power; 3) a redistribution of LV output away from the fetal upper body region and toward the lower body and placenta; 4) a greater proportion of RV output passing to the lungs, and a lesser proportion to the lower body and placenta; and 5) a change in the relative contribution of venous streams to ventricular filling, with the LV having increased pulmonary venous and decreased foramen ovale components, and the RV having lesser superior vena caval and greater inferior vena caval portions. Taken together, these findings suggest that antenatal betamethasone produces a widespread redistribution of central arterial and venous flows in the fetus, accompanied by a preferential rise in RV pumping performance.

Coupling between Blood Pressure and Subarachnoid Space Width Oscillations during Slow Breathing.

The precise mechanisms connecting the cardiovascular system and the cerebrospinal fluid (CSF) are not well understood in detail. This paper investigates the couplings between the cardiac and respiratory components, as extracted from blood pressure (BP) signals and oscillations of the subarachnoid space width (SAS), collected during slow ventilation and ventilation against inspiration resistance. The experiment was performed on a group of 20 healthy volunteers (12 females and 8 males; BMI kg/m; age years). We analysed the recorded signals with a wavelet transform. For the first time, a method based on dynamical Bayesian inference was used to detect the effective phase connectivity and the underlying coupling functions between the SAS and BP signals. There are several new findings. Slow breathing with or without resistance increases the strength of the coupling between the respiratory and cardiac components of both measured signals. We also observed increases in the strength of the coupling between the respiratory component of the BP and the cardiac component of the and vice versa. Slow breathing synchronises the SAS oscillations, between the brain hemispheres. It also diminishes the similarity of the coupling between all analysed pairs of oscillators, while inspiratory resistance partially reverses this phenomenon. BP–SAS and SAS–BP interactions may reflect changes in the overall biomechanical characteristics of the brain.

Entropy Analysis of COVID-19 Cardiovascular Signals.

The world has faced a coronavirus outbreak, which, in addition to lung complications, has caused other serious problems, including cardiovascular. There is still no explanation for the mechanisms of coronavirus that trigger dysfunction of the cardiac autonomic nervous system (ANS). We believe that the complex mechanisms that change the status of ANS could only be solved by advanced multidimensional analysis of many variables, obtained both from the original cardiovascular signals and from laboratory analysis and detailed patient history. The aim of this paper is to analyze different measures of entropy as potential dimensions of the multidimensional space of cardiovascular data. The measures were applied to heart rate and systolic blood pressure signals collected from 116 patients with COVID-19 and 77 healthy controls. Methods that indicate a statistically significant difference between patients with different levels of infection and healthy controls will be used for further multivariate research. As a result, it was shown that a statistically significant difference between healthy controls and patients with COVID-19 was shown by sample entropy applied to integrated transformed probability signals, common symbolic dynamics entropy, and copula parameters. Statistical significance between serious and mild patients with COVID-19 can only be achieved by cross-entropies of heart rate signals and systolic pressure. This result contributes to the hypothesis that the severity of COVID-19 disease is associated with ANS disorder and encourages further research.

Beat-to-beat blood pressure variability: an early predictor of disease and cardiovascular risk.

Blood pressure (BP) varies on the long, short and very-short term. Owing to the hidden physiological and pathological information present in BP time-series, increasing interest has been given to the study of continuous, beat-to-beat BP variability (BPV) using invasive and noninvasive methods. Different linear and nonlinear parameters of variability are employed in the characterization of BP signals in health and disease. Although linear parameters of beat-to-beat BPV are mainly measures of dispersion, such as standard deviation (SD), nonlinear parameters of BPV quantify the degree of complexity/irregularity- using measures of entropy or self-similarity/correlation. In this review, we summarize the value of linear and nonlinear parameters in reflecting different information about the pathophysiology of changes in beat-to-beat BPV independent of or superior to mean BP. We then provide a comparison of the relative power of linear and nonlinear parameters of beat-to-beat BPV in detecting early and subtle differences in various states. The practical advantage and utility of beat-to-beat BPV monitoring support its incorporation into routine clinical practices.

A New Framework to Estimate Breathing Rate From Electrocardiogram, Photoplethysmogram, and Blood Pressure Signals

Breathing Rate (BR) is a key physiological parameter measured in a wide range of clinical settings. However, it is still widely measured manually. In this paper, a novel framework is proposed to estimate the BR from an electrocardiogram (ECG), a photoplethysmogram (PPG), or a blood pressure (BP) signal. The framework uses Empirical Mode Decomposition (EMD) and Discrete Wavelet Transform (DWT) methods to extract respiratory signals, taking advantage of both time and frequency domain information. An Extended Kalman Filter (EKF), incorporating a Signal Quality Index (SQI), enabled our method to achieve acceptable performance even for significantly distorted periods of the signals. Using state vector fusion, the output signals are combined and finally the BR is estimated. The framework was tested on two publicly available clinical databases: the MIT-BIH Polysomnographic and BIDMC databases. Performance was evaluated using the mean absolute percentage error (MAPE). The results indicated high accuracy: MAPEs on the two databases of 3.9% and 3.6% for ECG signals, 6.0% for PPG, and 5.0% for BP signals. The results also indicated high robustness to noise down to 0dB. Therefore, this framework may have utility for BR monitoring in high noise settings.

Retracted] A Novel Smart Healthcare Monitoring System Using Machine Learning and the Internet of Things

The Internet of Things (IoT) has enabled the invention of smart health monitoring systems. These health monitoring systems can track a person’s mental and physical wellness. Stress, anxiety, and hypertension are key causes of many physical and mental disorders. Age‐related problems such as stress, anxiety, and hypertension necessitate specific attention in this setting. Stress, anxiety, and blood pressure monitoring can prevent long‐term damage by detecting problems early. This will increase the quality of life and reduce caregiver stress and healthcare costs. Determine fresh technology solutions for real‐time stress, anxiety, and blood pressure monitoring using discreet wearable sensors and machine learning approaches. This study created an automated artefact detection method for BP and PPG signals. It was proposed to automatically remove outlier points generated by movement artefacts from the blood pressure signal. Next, eleven features taken from the oscillometric waveform envelope were utilised to analyse the relationship between diastolic blood pressure (SBP) and systolic blood pressure (DBP). This paper validates a proposed computational method for estimating blood pressure. The proposed architecture leverages sophisticated regression to predict systolic and diastolic blood pressure values from PPG signal characteristics.

Wave-shape oscillatory model for nonstationary periodic time series analysis

The oscillations observed in many time series, particularly in biomedicine, exhibit morphological variations over time. These morphological variations are caused by intrinsic or extrinsic changes to the state of the generating system, henceforth referred to as dynamics. To model these time series (including and specifically pathophysiological ones) and estimate the underlying dynamics, we provide a novel wave-shape oscillatory model. In this model, time-dependent variations in cycle shape occur along a manifold called the wave-shape manifold. To estimate the wave-shape manifold associated with an oscillatory time series, study the dynamics, and visualize the time-dependent changes along the wave-shape manifold, we propose a novel algorithm coined Dynamic Diffusion map (DDmap) by applying the well-established diffusion maps (DM) algorithm to the set of all observed oscillations. We provide a theoretical guarantee on the dynamical information recovered by the DDmap algorithm under the proposed model. Applying the proposed model and algorithm to arterial blood pressure (ABP) signals recorded during general anesthesia leads to the extraction of nociception information. Applying the wave-shape oscillatory model and the DDmap algorithm to cardiac cycles in the electrocardiogram (ECG) leads to ectopy detection and a new ECG-derived respiratory signal, even when the subject has atrial fibrillation.

Non-Contact Blood Pressure Measurement Based on IPPG

Blood pressure is an important physiological parameter to reflect human vital signs. In order to achieve the non-contact dynamic blood pressure acquisition based on ordinary optical camera, a theoretical understanding of the functional relationship between blood pressure and pulse wave signal conduction time. And through imaging photoelectric plethysmography (IPPG), pulse wave signal conduction time of forehead and hand was obtained with ordinary optical camera. First, the pulse wave conduction time was obtained by recording the video with an ordinary optical camera. Second, real-time blood pressure values were collected. Finally, based on the relationship between blood pressure and pulse wave conduction time, a non-contact blood pressure measurement prediction model was obtained through neural network fitting. So that noncontact blood pressure measurement with optical camera could be completed. The method in this paper has several advantages, such as low requirements on measuring equipment, low cost, and simple operation. It can let people get rid of the discomfort caused by measuring equipment such as cuff and can measure blood pressure at any time. The predicted blood pressure results were compared with an Omron wrist electronic sphygmomanometer. The calculated error of systolic blood pressure is –9.28%~3.16%, and the error of diastolic blood pressure is –9.84~4.35%.

Study on the Correlation between OBF and Age Based on Analysis of Time Domain and Frequency Domain

Many studies indicate that the old people are prone to high blood pressure with the age growing. Elevated blood pressure can lead to the changes of hemodynamic parameters in the blood circulation system, and then cause abnormal changes in the physiological activities of the endothelium, heart and so on. These Characteristic of physiological rhythm reflect the changes of ocular blood pressure and ocular blood flow signals. Thus, we could study on the correlation between ocular blood flow (OBF) and age by analyzing the variety of age on hemodynamic. The paper sets out to fully examine the time domain and frequency domain relationship between OBF and age. Time domain: the paper introduces the thigh cuff method to produce sudden drops in arterial blood pressure (ABP) and study the evolution of the immediate OBF response with LSFG. The LSFG (Laser Speckle Flow graph) allows for quantitative estimation of blood flow in the optic nerve head (ONH), and it is the only noninvasive method that allows sufficient temporal resolution to measure these rapid changes. The paper studies on the relationship between time-domain parameter Kr(L) (The slope of DAR descending curve), Kr(R) (The slope of DAR up curve), Tr(L) (The amount of time it takes to the maximal BF or BP decrease), Tr(R) (The amount of time it takes to the maximal BF or BP increase) and age, respectively. Kr(R), Kr(R), Tr(R) are shown significantly influenced by age changing. Frequency domain: in the case of without any factor, the harmonic content of the envelope waveform of the blood flow velocity in the ophthalmic artery is analyzed in aging and OBF. Cronbach-α reliabilities of the calculated frequency coefficient (FC) and frequency index (FI) are 0.91 and 0.92, respectively. The FI of the ophthalmic artery allows an assessment of the ocular circulation in consideration of age and OBF. The order the age, the smaller the FC, FI in OBF.

A Manchester-OOK Visible Light Communication System for Patient Monitoring in Intensive Care Units

Visible Light Communication (VLC) is an interesting solution in Intensive Care (IC) medical environments to, among others, prevent the spread of emerging diseases. This work presents the design of a low-cost and stable VLC transceiver system in addition to practical characterizations and experimental performance evaluations for application in IC environments. The application focuses on the transmission of Manchester-based on-off keying signals, in a proof-of-concept apparatus that makes use of the Eye Opening Penalty (EOP) metric to illustrate a comprehensive picture of the potentials of the evaluated VLC system. Parameters such as Light-Emitting Diode (LED) bias current, modulation frequency, line-of-sight link distance, and signal pattern were analyzed here. Experimental results show an outstanding behavior considering an LED bias current of 400 mA, a signal frequency of 1 MHz and a VLC transmission link length of 2.5 m. To validate our system, vital parameters as heart rate, oxygen saturation, pulse rate, respiration rate, temperature and non-intrusive blood pressure signals were transmitted for VLC links of 1.5, 5 and 15 m, which were received by a multi-parameter monitor, with EOP values of 0.89, 0.96 and 2.67 dB, respectively.

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

Hypertension or high blood pressure is a major health problem worldwide and primary risk factor for cardiovascular disease. Blood pressure is one of the four vital signs that provides important information regarding patients’ cardiovascular system conditions. Continuous and regular blood pressure monitoring is essential for early diagnosis and prevention of cardiovascular disease. Considering the existing invasive or cuff-based blood pressuring monitoring techniques in clinical practice, several studies have identified motivation and advantages of a new non-invasive and cuffless blood pressuring measurement technique using Photoplethysmogram (PPG) signals. In this study, we propose several systolic and diastolic blood pressure estimation models using recurrent neural networks with bidirectional connections and attention mechanism utilising only PPG signals. The models were evaluated on PPG and blood pressure signals derived from the Multiparameter Intelligent Monitoring in Intensive Care II database. In the process, 22 characteristic features were extracted from the PPG waveform followed by various dimensionality reduction techniques to eliminate redundancies and reduce computational complexity. The proposed models were evaluated on both the 22-feature set and the reduced input feature vector, respectively. The models were compared with four machine learning techniques commonly used in the literature. Experimental results demonstrated that the proposed models could capture the non-linear relationship between the PPG features and blood pressure with high accuracy and outperformed the conventional machine learning methods on both datasets. The results for all the proposed models were acceptable by the global standards set by the Association for the Advancement of Medical Instrumentation for cuffless blood pressure estimation.

Predicting hypotension in the ICU using noninvasive physiological signals

Hypotension frequently occurs in Intensive Care Units (ICU), and its early prediction can improve the outcome of patient care. Trends observed in signals related to blood pressure (BP) are critical in predicting future events. Unfortunately, the invasive measurement of BP signals is neither comfortable nor feasible in all bed settings. In this study, we investigate the performance of machine-learning techniques in predicting hypotensive events in ICU settings using physiological signals that can be obtained noninvasively. We show that noninvasive mean arterial pressure (NIMAP) can be simulated by down-sampling the invasively measured MAP. This enables us to investigate the effect of BP measurement frequency on the algorithm's performance by training and testing the algorithm on a large dataset provided by the MIMIC III database. This study shows that having NIMAP information is essential for adequate predictive performance. The proposed predictive algorithm can flag hypotension with a sensitivity of 84%, positive predictive value (PPV) of 73%, and F1-score of 78%. Furthermore, the predictive performance of the algorithm improves by increasing the frequency of BP sampling.

Fourier analysis: the simplification of a complex waveform into simple component sine waves of different amplitudes and frequencies

Fourier analysis is the simplification of a complex waveform into simple component sine waves of different amplitudes and frequencies. A discussion on Fourier analysis necessitates reiteration of the physics of waves. A wave is a series of repeating disturbances that propagate in space and time. Frequency: the number of oscillations, or cycles per second. It is measured in Hertz and denoted as 1/time or s-1. Fundamental frequency: the lowest frequency wave in a series. It is also known as the first harmonic. Every other wave in the series is an exact multiple of the fundamental frequency. Harmonic: whole number multiples of the fundamental frequency. Amplitude: the maximum disturbance or displacement from zero caused by the wave. This is the height of the wave. Period: time to complete one oscillation. Wavelength: physical length of one complete cycle. This can be between two crests or two troughs. The higher the frequency, the shorter the wavelength. Velocity: frequency x wavelength. Phase: displacement of one wave compared to another, described as 0°–360°. A sine wave is a simple wave. It can be depicted as the path of a point travelling round a circle at a constant speed, defined by the equation ‘y = sinx’. Combining sine waves of different frequency, amplitude and phase can yield any waveform, and, conversely, any wave can be simplified into its component sine waves. Fourier analysis is a mathematical method of analysing a complex periodic waveform to find its constituent frequencies (as sine waves). Complex waveforms can be analysed, with very simple results. Usually, few sine and cosine waves combine to create reasonably accurate representations of most waves. Fourier analysis finds its anaesthetic applications in invasive blood pressure, electrocardiogram (ECG) and electroencephalogram (EEG) signals, which are all periodic waveforms. It enables monitors to display accurate representations of these biological waveforms. Fourier analysis was developed by Joseph Fourier, a mathematician who analysed and altered periodic waveforms. It is done by computer programmes that plot the results of the analysis as a spectrum of frequencies with amplitude on the y-axis and frequency on the x-axis.

Abstract 16850: Machine Learning Based Estimation of Systolic and Diastolic Arterial Blood Pressure From the Electrocardiogram and Oxygen Saturation Waveforms

Introduction: Accurate, continuous blood pressure (BP) monitoring requires invasive equipment. We investigate the utility of a machine learning algorithm (ML) to estimate the systolic (sys) and diastolic (dia) BP from the electrocardiogram (ECG) and oxygen saturation (SpO2) signals. Methods: We identified over 2,000 hours (~9.2 M heart-beats), of 4 lead ECG, arterial BP, and SpO2 waveforms, from 82 random ICU patients, with various diagnoses. The ECG, BP and SpO2 waveforms were delineated using a wavelet-transform-based algorithm. Features extracted from the ECG and SpO2 signals were used as input to the ML model, while the sys/dia BP values of the corresponding BP signal (range: 0-250 mmHg), served as the gold standard. Using these features, we trained two independent Random Forest (RF) models to estimate the sys/dia BP values, and measured their performance over five-fold cross-validation. To identify the optimal window-length (# heart beats) in estimating sys/dia BP values, the entire process of extracting features and training/testing of RF models was performed for window-lengths of {1,5,6,7,8,9,10,11,12,13,14,15,20,30,50} heart-beats. Results: The root-mean-square error (RMSE) at each window-length is shown in the Figure. The lowest RMSE for both estimated sys/dia BP, 5.81±4.61 mmHg and 3.73±3.26 mmHg, respectively, was achieved at 14 heart-beats. The corresponding RMSE values using a one-beat length-window were 7.73±5.93 mmHg and 5.28±4.42 mmHg, respectively. This result indicates the ability to estimate the instantaneous BP values even on a beat-to-beat basis, with low-error. Conclusions: In comparison to prior studies, our novel ML model, to the best of our knowledge, has been trained over the largest amount of data. Our ML model can accurately estimate the sys/dia values of the arterial BP. Figure: RMSE values obtained while estimating the mean sys/dia BP, across different window-lengths.