Calibration-free blood pressure estimation based on a convolutional neural network

In this study, we conducted research on a deep learning-based blood pressure (BP) estimation model suitable for wearable environments. To measure BP while wearing a wearable watch, it needs to be considered that computing power for signal processing is limited and the input signals are subject to noise interference. Therefore, we employed a convolutional neural network (CNN) as the BP estimation model and utilized time-series electrocardiogram (ECG) and photoplethysmogram (PPG) signals, which are quantifiable in a wearable context. We generated periodic input signals and used differential and thresholding methods to decrease noise in the preprocessing step. We then applied a max-pooling technique with filter sizes of 2 × 1 and 5 × 1 within a 3-layer convolutional neural network to estimate BP. Our method was trained, validated, and tested using 2.4 million data samples from 49 patients in the intensive care unit. These samples, totaling 3.1 GB were obtained from the publicly accessible MIMIC database. As a result of a test with 480,000 data samples, the average root mean square error in BP estimation was 3.41, 5.80, and 2.78 mm Hg in the prediction of pulse pressure, systolic BP (SBP), and diastolic BP (DBP), respectively. The cumulative error percentage less than 5 mm Hg was 68% and 93% for SBP and DBP, respectively. In addition, the cumulative error percentage less than 15 mm Hg was 98% and 99% for SBP and DBP. Subsequently, we evaluated the impact of changes in input signal length (1 cycle vs. 30 s) and the introduction of noise on BP estimation results. The experimental results revealed that the length of the input signal did not significantly affect the performance of CNN-based analysis. When estimating BP using noise-added ECG signals, the mean absolute error (MAE) for SBP and DBP was 9.72 and 6.67 mm Hg, respectively. Meanwhile, when using noise-added PPG signals, the MAE for SBP and DBP was 26.85 and 14.00 mm Hg, respectively. Therefore, this study confirmed that using ECG signals rather than PPG signals is advantageous for noise reduction in a wearable environment. Besides, short sampling frames without calibration can be effective as input signals. Furthermore, it demonstrated that using a model suitable for information extraction rather than a specialized deep learning model for sequential data can yield satisfactory results in BP estimation.

Non-Invasive Heart Failure Evaluation Using Machine Learning Algorithms

Heart failure is a prevalent cardiovascular condition with significant health implications, necessitating effective diagnostic strategies for timely intervention. This study explores the potential of continuous monitoring of non-invasive signals, specifically integrating photoplethysmogram (PPG) and electrocardiogram (ECG), for enhancing early detection and diagnosis of heart failure. Leveraging a dataset from the MIMIC-III database, encompassing 682 heart failure patients and 954 controls, our approach focuses on continuous, non-invasive monitoring. Key features, including the QRS interval, RR interval, augmentation index, heart rate, systolic pressure, diastolic pressure, and peak-to-peak amplitude, were carefully selected for their clinical relevance and ability to capture cardiovascular dynamics. This feature selection not only highlighted important physiological indicators but also helped reduce computational complexity and the risk of overfitting in machine learning models. The use of these features in training machine learning algorithms led to a model with impressive accuracy (98%), sensitivity (97.60%), specificity (96.90%), and precision (97.20%). Our integrated approach, combining PPG and ECG signals, demonstrates superior performance compared to single-signal strategies, emphasizing its potential in early and precise heart failure diagnosis. The study also highlights the importance of continuous monitoring with wearable technology, suggesting a significant stride forward in non-invasive cardiovascular health assessment. The proposed approach holds promise for implementation in hardware systems to enable continuous monitoring, aiding in early detection and prevention of critical health conditions.

A Novel Respiratory Rate Estimation Algorithm from Photoplethysmogram Using Deep Learning Model

Respiratory rate (RR) is a critical vital sign that can provide valuable insights into various medical conditions, including pneumonia. Unfortunately, manual RR counting is often unreliable and discontinuous. Current RR estimation algorithms either lack the necessary accuracy or demand extensive window sizes. In response to these challenges, this study introduces a novel method for continuously estimating RR from photoplethysmogram (PPG) with a reduced window size and lower processing requirements. To evaluate and compare classical and deep learning algorithms, this study leverages the BIDMC and CapnoBase datasets, employing the Respiratory Rate Estimation (RRest) toolbox. The optimal classical techniques combination on the BIDMC datasets achieves a mean absolute error (MAE) of 1.9 breaths/min. Additionally, the developed neural network model utilises convolutional and long short-term memory layers to estimate RR effectively. The best-performing model, with a 50% train-test split and a window size of 7 s, achieves an MAE of 2 breaths/min. Furthermore, compared to other deep learning algorithms with window sizes of 16, 32, and 64 s, this study's model demonstrates superior performance with a smaller window size. The study suggests that further research into more precise signal processing techniques may enhance RR estimation from PPG signals.

Investigating the impact of smoking habits through photoplethysmography analysis

Smoking is widely recognized as a significant risk factor in the progression of arterial stiffness and cardiovascular diseases. Valuable information related to cardiac arrhythmias and heart function can be obtained by analyzing biosignals such as the electrocardiogram (ECG) and the photoplethysmogram (PPG). The PPG signal is a non-invasive optical technique that can be used to evaluate the changes in blood volume, and thus it can be linked to the health of the vascular system.Objective. In this study, the impact of three smoking habits-cigarettes, shisha, and electronic cigarettes (e-cigarettes)-on the features of the PPG signal were investigated.Approach. The PPG signals are measured for 45 healthy smokers before, during, and after the smoking session and then processed to extract the morphological features. Quantitative statistical techniques were used to analyze the PPG features and provide the most significant features of the three smoking habits. The impact of smoking is observed through significant changes in the features of the PPG signal, indicating blood volume instability.Main results. The results revealed that the three smoking habits influence the characteristics of the PPG signal significantly, which presentseven after 15 min of smoking. Among them, shisha has the greatest impact on PPG features, particularly on heart rate, systolic time, augmentation index, and peak pulse interval change. In contrast, e-cigarettes have the least effect on PPG features. Interestingly, smoking electronic cigarettes, which many participants use as a substitute for traditional cigarettes when attempting to quit smoking, has nearly a comparable effect to regular smoking.Significance. The findings suggest that individuals who smoke shisha are more likely to develop cardiovascular diseases at an earlier age compared to those who have other smoking habits. Understanding the variations in the PPG signal caused by smoking can aid in the early detection of cardiovascular disorders and provide insight into cardiac conditions. This ultimately contributes to the prevention of the development of cardiovascular diseases and the development of a health screening system.

Evaluation of the Photoplethysmogram-Based Deep Learning Model for Continuous Respiratory Rate Estimation in Surgical Intensive Care Unit

The respiratory rate (RR) is a significant indicator to evaluate a patient's prognosis and status; however, it requires specific instrumentation or estimates from other monitored signals. A photoplethysmogram (PPG) is extensively used in clinical environments as well as in intensive care units (ICUs) to primarily monitor peripheral circulation while capturing indirect information about intrathoracic pressure changes. This study aims to apply and evaluate several deep learning models using a PPG for the continuous and accurate estimation of the RRs of patients. The dataset was collected twice for 2 min each in 100 patients aged 18 years and older from the surgical intensive care unit of a tertiary referral hospital. The BIDMC and CapnoBase public datasets were also analyzed. The collected dataset was preprocessed and split according to the 5-fold cross-validation. We used seven deep learning models, including our own Dilated Residual Neural Network, to check how accurately the RR estimates match the ground truth using the mean absolute error (MAE). As a result, when validated using the collected dataset, our model showed the best results with a 1.2628 ± 0.2697 MAE on BIDMC and RespNet and with a 3.1268 ± 0.6363 MAE on our dataset, respectively. In conclusion, RR estimation using PPG-derived models is still challenging and has many limitations. However, if there is an equal amount of data from various breathing groups to train, we expect that various models, including our Dilated ResNet model, which showed good results, can achieve better results than the current ones.

Evaluation of the effect of photoplethysmograms on workers' exposure to methyl bromide using second derivative

Methyl bromide (MB) is worldwide the only effective fumigant heavily used for quarantine pre-shipment treatment and has a critical use exemption for soil fumigations due to its excellent permeability and insecticidal effect. However, MB should be replaced as it is an an ozone-depleting substance and also highly toxic to humans. Recently, MB has been shown to be hazardous even for asymptomatic workers, affecting their central and autonomic nervous systems. However, the effects of MB exposure on vascular health have not been explored. This study aimed to determine whether MB affects the arterial system of asymptomatic workers. We measured the second derivative of the photoplethysmogram (SDPTG) indices, which are indicators of vascular load and aging, and urinary bromide ion (Br-) concentrations in 44 fumigators (study group) and 20 inspectors (control group) before and after fumigation. In fumigators, the mean values of post-work SDPTG indices (b/a, c/a, d/a, e/a, and SDPTG aging index) and Br- levels were significantly changed compared to their pre-work values (p < 0.05), indicating a negative effect on their cardiovascular health. In contrast, SDPTG indices and Br- levels in inspectors did not show any differences before and after work. All SDPTG indices except c/a showed significant correlations with Br- levels in all individuals (p < 0.05). In conclusion, the Br- levels and SDPTG indices of fumigators varied after MB work, and they experienced negative effects on their health despite being asymptomatic.

Fully convolutional neural network and PPG signal for arterial blood pressure waveform estimation

Objective. The quality of the arterial blood pressure (ABP) waveform is crucial for predicting the value of blood pressure. The ABP waveform is predicted through experiments, and then Systolic blood pressure (SBP), Diastolic blood pressure, (DBP), and Mean arterial pressure (MAP) information are estimated from the ABP waveform.Approach. To ensure the quality of the predicted ABP waveform, this paper carefully designs the network structure, input signal, loss function, and structural parameters. A fully convolutional neural network (CNN) MultiResUNet3+ is used as the core architecture of ABP-MultiNet3+. In addition to performing Kalman filtering on the original photoplethysmogram (PPG) signal, its first-order derivative and second-order derivative signals are used as ABP-MultiNet3+ enter. The model's loss function uses a combination of mean absolute error (MAE) and means square error (MSE) loss to ensure that the predicted ABP waveform matches the reference waveform.Main results. The proposed ABP-MultiNet3+ model was tested on the public MIMIC II databases, MAE of MAP, DBP, and SBP was 1.88 mmHg, 3.11 mmHg, and 4.45 mmHg, respectively, indicating a small model error. It experiment fully meets the standards of the AAMI standard and obtains level A in the DBP and MAP prediction standard test under the BHS standard. For SBP prediction, it obtains level B in the BHS standard test. Although it does not reach level A, it has a certain improvement compared with the existing methods.Significance. The results show that this algorithm can achieve sleeveless blood pressure estimation, which may enable mobile medical devices to continuously monitor blood pressure and greatly reduce the harm caused by Cardiovascular disease (CVD).

Green-Solvent-Processed High-Performance Broadband Organic Photodetectors

Solution-processed organic photodetectors with broadband activity have been demonstrated with an environmentally benign solvent, ortho-xylene (o-xylene), as the processing solvent. The organic photodetectors employ a wide band gap polymer donor PBDB-T and a narrow band gap small-molecule non-fullerene acceptor CO1-4F, both dissolvable in o-xylene at a controlled temperature. The o-xylene-processed devices have shown external quantum efficiency of up to 70%, surpassing the counterpart processed with chlorobenzene. With a well-suppressed dark current, the device can also present a high specific detectivity of over 1012 Jones at -2 V within practical operation frequencies and is applicable for photoplethysmography with its fast response. These results further highlight the potential of green-solvent-processed organic photodetectors as a high-performing alternative to their counterparts processed in toxic chlorinated solvents without compromising the excellent photosensing performance.

A Deep Learning Approach for Atrial Fibrillation Classification Using Multi-Feature Time Series Data from ECG and PPG

Atrial fibrillation is a prevalent cardiac arrhythmia that poses significant health risks to patients. The use of non-invasive methods for AF detection, such as Electrocardiogram and Photoplethysmogram, has gained attention due to their accessibility and ease of use. However, there are challenges associated with ECG-based AF detection, and the significance of PPG signals in this context has been increasingly recognized. The limitations of ECG and the untapped potential of PPG are taken into account as this work attempts to classify AF and non-AF using PPG time series data and deep learning. In this work, we emploted a hybrid deep neural network comprising of 1D CNN and BiLSTM for the task of AF classification. We addressed the under-researched area of applying deep learning methods to transmissive PPG signals by proposing a novel approach. Our approach involved integrating ECG and PPG signals as multi-featured time series data and training deep learning models for AF classification. Our hybrid 1D CNN and BiLSTM model achieved an accuracy of 95% on test data in identifying atrial fibrillation, showcasing its strong performance and reliable predictive capabilities. Furthermore, we evaluated the performance of our model using additional metrics. The precision of our classification model was measured at 0.88, indicating its ability to accurately identify true positive cases of AF. The recall, or sensitivity, was measured at 0.85, illustrating the model's capacity to detect a high proportion of actual AF cases. Additionally, the F1 score, which combines both precision and recall, was calculated at 0.84, highlighting the overall effectiveness of our model in classifying AF and non-AF cases.

Automated Multi-Wavelength Quality Assessment of Photoplethysmography Signals Using Modulation Spectrum Shape Features

Photoplethysmography (PPG) is used to measure blood volume changes in the microvascular bed of tissue. Information about these changes along time can be used for estimation of various physiological parameters, such as heart rate variability, arterial stiffness, and blood pressure, to name a few. As a result, PPG has become a popular biological modality and is widely used in wearable health devices. However, accurate measurement of various physiological parameters requires good-quality PPG signals. Therefore, various signal quality indexes (SQIs) for PPG signals have been proposed. These metrics have usually been based on statistical, frequency, and/or template analyses. The modulation spectrogram representation, however, captures the second-order periodicities of a signal and has been shown to provide useful quality cues for electrocardiograms and speech signals. In this work, we propose a new PPG quality metric based on properties of the modulation spectrum. The proposed metric is tested using data collected from subjects while they performed various activity tasks contaminating the PPG signals. Experiments on this multi-wavelength PPG dataset show the combination of proposed and benchmark measures significantly outperforming several benchmark SQIs with improvements of 21.3% BACC (balanced accuracy) for green, 21.6% BACC for red, and 19.0% BACC for infrared wavelengths, respectively, for PPG quality detection tasks. The proposed metrics also generalize for cross-wavelength PPG quality detection tasks.

Photoplethysmogram Biometric Authentication Using a 1D Siamese Network

In the head-mounted display environment for experiencing metaverse or virtual reality, conventional input devices cannot be used, so a new type of nonintrusive and continuous biometric authentication technology is required. Since the wrist wearable device is equipped with a photoplethysmogram sensor, it is very suitable for use for nonintrusive and continuous biometric authentication purposes. In this study, we propose a one-dimensional Siamese network biometric identification model using a photoplethysmogram. To maintain the unique characteristics of each person and reduce noise in preprocessing, we adopted a multicycle averaging method without using a bandpass or low-pass filter. In addition, to verify the effectiveness of the multicycle averaging method, the number of cycles was changed and the results were compared. Genuine and impostor data were used to verify the biometric identification. We used the one-dimensional Siamese network to verify the similarity between the classes and found that the method with five overlapping cycles was the most effective. Tests were conducted on the overlapping data of five single-cycle signals and excellent identification results were observed, with an AUC score of 0.988 and an accuracy of 0.9723. Thus, the proposed biometric identification model is time-efficient and shows excellent security performance, even in devices with limited computational capabilities, such as wearable devices. Consequently, our proposed method has the following advantages compared with previous works. First, the effect of noise reduction and information preservation through multicycle averaging was experimentally verified by varying the number of photoplethysmogram cycles. Second, by analyzing authentication performance through genuine and impostor matching analysis based on a one-dimensional Siamese network, the accuracy that is not affected by the number of enrolled subjects was derived.

COVID-19 Detection Using Photoplethysmography and Neural Networks

The early identification of microvascular changes in patients with Coronavirus Disease 2019 (COVID-19) may offer an important clinical opportunity. This study aimed to define a method, based on deep learning approaches, for the identification of COVID-19 patients from the analysis of the raw PPG signal, acquired with a pulse oximeter. To develop the method, we acquired the PPG signal of 93 COVID-19 patients and 90 healthy control subjects using a finger pulse oximeter. To select the good quality portions of the signal, we developed a template-matching method that excludes samples corrupted by noise or motion artefacts. These samples were subsequently used to develop a custom convolutional neural network model. The model accepts PPG signal segments as input and performs a binary classification between COVID-19 and control samples. The proposed model showed good performance in identifying COVID-19 patients, achieving 83.86% accuracy and 84.30% sensitivity (hold-out validation) on test data. The obtained results indicate that photoplethysmography may be a useful tool for microcirculation assessment and early recognition of SARS-CoV-2-induced microvascular changes. In addition, such a noninvasive and low-cost method is well suited for the development of a user-friendly system, potentially applicable even in resource-limited healthcare settings.

DDM-HSA: Dual Deterministic Model-Based Heart Sound Analysis for Daily Life Monitoring

A sudden cardiac event in patients with heart disease can lead to a heart attack in extreme cases. Therefore, prompt interventions for the particular heart situation and periodic monitoring are critical. This study focuses on a heart sound analysis method that can be monitored daily using multimodal signals acquired with wearable devices. The dual deterministic model-based heart sound analysis is designed in a parallel structure that uses two bio-signals (PCG and PPG signals) related to the heartbeat, enabling more accurate heart sound identification. The experimental results show promising performance of the proposed Model III (DDM-HSA with window and envelope filter), which had the highest performance, and S1 and S2 showed average accuracy (unit: %) of 95.39 (±2.14) and 92.55 (±3.74), respectively. The findings of this study are anticipated to provide improved technology to detect heart sounds and analyze cardiac activities using only bio-signals that can be measured using wearable devices in a mobile environment.

The Accuracy of Wrist-Worn Photoplethysmogram-Measured Heart and Respiratory Rates in Abdominal Surgery Patients: Observational Prospective Clinical Validation Study

BACKGROUND

Postoperative deterioration is often preceded by abnormal vital parameters. Therefore, vital parameters of postoperative patients are routinely measured by nursing staff. Wrist-worn sensors could potentially provide an alternative tool for the measurement of vital parameters in low-acuity settings. These devices would allow more frequent or even continuous measurements of vital parameters without relying on time-consuming manual measurements, provided their accuracy in this clinical population is established.

OBJECTIVE

This study aimed to assess the accuracy of heart rate (HR) and respiratory rate (RR) measures obtained via a wearable photoplethysmography (PPG) wristband in a cohort of postoperative patients.

METHODS

The accuracy of the wrist-worn PPG sensor was assessed in 62 post-abdominal surgery patients (mean age 55, SD 15 years; median BMI 34, IQR 25-40 kg/m²). The wearable obtained HR and RR measurements were compared to those of the reference monitor in the postanesthesia or intensive care unit. Bland-Altman and Clarke error grid analyses were performed to determine agreement and clinical accuracy.

RESULTS

Data were collected for a median of 1.2 hours per patient. With a coverage of 94% for HR and 34% for RR, the device was able to provide accurate measurements for the large majority of the measurements as 98% and 93% of the measurements were within 5 bpm or 3 rpm of the reference signal. Additionally, 100% of the HR and 98% of the RR measurements were clinically acceptable on Clarke error grid analysis.

CONCLUSIONS

The wrist-worn PPG device is able to provide measurements of HR and RR that can be seen as sufficiently accurate for clinical applications. Considering the coverage, the device was able to continuously monitor HR and report RR when measurements of sufficient quality were obtained.

TRIAL REGISTRATION

ClinicalTrials.gov NCT03923127; https://www.clinicaltrials.gov/ct2/show/NCT03923127.

Influence of Photoplethysmogram Signal Quality on Pulse Arrival Time during Polysomnography

Intervals of low-quality photoplethysmogram (PPG) signals might lead to significant inaccuracies in estimation of pulse arrival time (PAT) during polysomnography (PSG) studies. While PSG is considered to be a "gold standard" test for diagnosing obstructive sleep apnea (OSA), it also enables tracking apnea-related nocturnal blood pressure fluctuations correlated with PAT. Since the electrocardiogram (ECG) is recorded synchronously with the PPG during PSG, it makes sense to use the ECG signal for PPG signal-quality assessment. (1) Objective: to develop a PPG signal-quality assessment algorithm for robust PAT estimation, and investigate the influence of signal quality on PAT during various sleep stages and events such as OSA. (2) Approach: the proposed algorithm uses R and T waves from the ECG to determine approximate locations of PPG pulse onsets. The MESA database of 2055 PSG recordings was used for this study. (3) Results: the proportions of high-quality PPG were significantly lower in apnea-related oxygen desaturation (matched-pairs rc = 0.88 and rc = 0.97, compared to OSA and hypopnea, respectively, when p < 0.001) and arousal (rc = 0.93 and rc = 0.98, when p < 0.001) than in apnea events. The significantly large effect size of interquartile ranges of PAT distributions was between low- and high-quality PPG (p < 0.001, rc = 0.98), and regular and irregular pulse waves (p < 0.001, rc = 0.74), whereas a lower quality of the PPG signal was found to be associated with a higher interquartile range of PAT across all subjects. Suggested PPG signal quality-based PAT evaluation reduced deviations (e.g., rc = 0.97, rc = 0.97, rc = 0.99 in hypopnea, oxygen desaturation, and arousal stages, respectively, when p < 0.001) and allowed obtaining statistically larger differences between different sleep stages and events. (4) Significance: the implemented algorithm has the potential to increase the robustness of PAT estimation in PSG studies related to nocturnal blood pressure monitoring.

Forehead monitoring of heart rate in neonatal intensive care

Heart rate is an extremely important physiological parameter to measure in critically unwell infants, as it is the main physiological marker that changes in response to a change in infant condition. Heart rate is routinely measured peripherally on a limb with a pulse oximeter. However, when infants are critically unwell, the blood supply to these peripheries is reduced in preference for central perfusion of vital organs such as the brain and heart. Measurement of heart rate with a reflection mode photoplethysmogram (PPG) sensor on the forehead could help minimise this problem and make it easier for other important medical equipment, such as cannulas, to be placed on the limbs. This study compares heart rates measured with a forehead-based PPG sensor against a wrist-based PPG sensor in 19 critically unwell infants in neonatal intensive care collecting 198 h of data. The two heart rates were compared using positive percentage agreement, Spearman's correlation coefficient and Bland-Altman analysis. The forehead PPG sensor showed good agreement with the wrist-based PPG sensor with limits of agreement of 8.44 bpm, bias of -0.22 bpm; positive percentage agreement of 98.87%; and Spearman's correlation coefficient of 0.9816. The analysis demonstrates that the forehead is a reliable alternative location for measuring vital signs using the PPG.

Variations in the Second Derivative of a Photoplethysmogram with Age in Healthy Korean Adults

Second derivative of photoplethysmogram (SDPTG) indices correlate with aging and vascular health. The trend of SDPTG indices with age has not yet been studied in the Korean population. Various SDPTG indices were measured in 300 healthy Koreans (150 men and 150 women), aged 19−69 years, stratified into five age groups consisting of 60 people (30 men and 30 women) in each age group from their 20s to 60s. The values of the SDPTG indices clearly showed distinct variations with age in healthy Korean groups (p < 0.001 for all indices). b/a increased linearly with age (y = 0.0045x − 0.803), as did SDPTG aging index (SDPTG-AI) (y = 0.0162x − 1.1389). c/a decreased linearly with age (y = −0.0044x + 0.1017), as did d/a (y = −0.0062x + 0.034) and e/a (y = −0.001x + 0.2002). A significant sex difference was shown in b/a and e/a ratios and SDPTG-AI (p < 0.001 for all indices); women had a higher b/a ratio (−0.55 ± 0.14 versus −0.65 ± 0.13) and SDPTG-AI (−0.33 ± 0.3 versus −0.52 ± 0.33) and a lower e/a ratio (0.13 ± 0.06 versus 0.18 ± 0.07) than men. A linear regression model of diverse SDPTG indices was provided according to the age of healthy Koreans, which may be valuable in preventing diseases related to vascular conditions by estimating the degradation of arterial function.

Noninvasive Monitoring of Simulated Hemorrhage and Whole Blood Resuscitation

Hemorrhage is the leading cause of preventable death from trauma. Accurate monitoring of hemorrhage and resuscitation can significantly reduce mortality and morbidity but remains a challenge due to the low sensitivity of traditional vital signs in detecting blood loss and possible hemorrhagic shock. Vital signs are not reliable early indicators because of physiological mechanisms that compensate for blood loss and thus do not provide an accurate assessment of volume status. As an alternative, machine learning (ML) algorithms that operate on an arterial blood pressure (ABP) waveform have been shown to provide an effective early indicator. However, these ML approaches lack physiological interpretability. In this paper, we evaluate and compare the performance of ML models trained on nine ABP-derived features that provide physiological insight, using a database of 13 human subjects from a lower-body negative pressure (LBNP) model of progressive central hypovolemia and subsequent progressive restoration to normovolemia (i.e., simulated hemorrhage and whole blood resuscitation). Data were acquired at multiple repressurization rates for each subject to simulate varying resuscitation rates, resulting in 52 total LBNP collections. This work is the first to use a single ABP-based algorithm to monitor both simulated hemorrhage and resuscitation. A gradient-boosted regression tree model trained on only the half-rise to dicrotic notch (HRDN) feature achieved a root-mean-square error (RMSE) of 13%, an R² of 0.82, and area under the receiver operating characteristic curve of 0.97 for detecting decompensation. This single-feature model's performance compares favorably to previously reported results from more-complex black box machine learning models. This model further provides physiological insight because HRDN represents an approximate measure of the delay between the ABP ejected and reflected wave and therefore is an indication of cardiac and peripheral vascular mechanisms that contribute to the compensatory response to blood loss and replacement.

Lightweight End-to-End Deep Learning Solution for Estimating the Respiration Rate from Photoplethysmogram Signal

Respiratory ailments are a very serious health issue and can be life-threatening, especially for patients with COVID. Respiration rate (RR) is a very important vital health indicator for patients. Any abnormality in this metric indicates a deterioration in health. Hence, continuous monitoring of RR can act as an early indicator. Despite that, RR monitoring equipment is generally provided only to intensive care unit (ICU) patients. Recent studies have established the feasibility of using photoplethysmogram (PPG) signals to estimate RR. This paper proposes a deep-learning-based end-to-end solution for estimating RR directly from the PPG signal. The system was evaluated on two popular public datasets: VORTAL and BIDMC. A lightweight model, ConvMixer, outperformed all of the other deep neural networks. The model provided a root mean squared error (RMSE), mean absolute error (MAE), and correlation coefficient (R) of 1.75 breaths per minute (bpm), 1.27 bpm, and 0.92, respectively, for VORTAL, while these metrics were 1.20 bpm, 0.77 bpm, and 0.92, respectively, for BIDMC. The authors also showed how fine-tuning a small subset could increase the performance of the model in the case of an out-of-distribution dataset. In the fine-tuning experiments, the models produced an average R of 0.81. Hence, this lightweight model can be deployed to mobile devices for real-time monitoring of patients.

Photoplethysmogram Recording Length: Defining Minimal Length Requirement from Dynamical Characteristics

Photoplethysmography is a widely used technique to noninvasively assess heart rate, blood pressure, and oxygen saturation. This technique has considerable potential for further applications—for example, in the field of physiological and mental health monitoring. However, advanced applications of photoplethysmography have been hampered by the lack of accurate and reliable methods to analyze the characteristics of the complex nonlinear dynamics of photoplethysmograms. Methods of nonlinear time series analysis may be used to estimate the dynamical characteristics of the photoplethysmogram, but they are highly influenced by the length of the time series, which is often limited in practical photoplethysmography applications. The aim of this study was to evaluate the error in the estimation of the dynamical characteristics of the photoplethysmogram associated with the limited length of the time series. The dynamical properties were evaluated using recurrence quantification analysis, and the estimation error was computed as a function of the length of the time series. Results demonstrated that properties such as determinism and entropy can be estimated with an error lower than 1% even for short photoplethysmogram recordings. Additionally, the lower limit for the time series length to estimate the average prediction time was computed.

Cuffless and Touchless Measurement of Blood Pressure from Ballistocardiogram Based on a Body Weight Scale

Currently, in terms of reducing the infection risk of the COVID-19 virus spreading all over the world, the development of touchless blood pressure (BP) measurement has potential benefits. The pulse transit time (PTT) has a high relation with BP, which can be measured by electrocardiogram (ECG) and photoplethysmogram (PPG). The ballistocardiogram (BCG) reflects the mechanical vibration (or displacement) caused by the heart contraction/relaxation (or heart beating), which can be measured from multiple degrees of the body. The goal of this study is to develop a cuffless and touchless BP-measurement method based on a commercial weight scale combined with a PPG sensor when measuring body weight. The proposed method was that the PTTBCG-PPGT was extracted from the BCG signal measured by a weight scale, and the PPG signal was measured from the PPG probe placed at the toe. Four PTT models were used to estimate BP. The reference method was the PTTECG-PPGF extracted from the ECG signal and PPG signal measured from the PPG probe placed at the finger. The standard BP was measured by an electronic blood pressure monitor. Twenty subjects were recruited in this study. By the proposed method, the root-mean-square error (ERMS) of estimated systolic blood pressure (SBP) and diastolic blood pressure (DBP) are 6.7 ± 1.60 mmHg and 4.8 ± 1.47 mmHg, respectively. The correlation coefficients, r2, of the proposed model for the SBP and DBP are 0.606 ± 0.142 and 0.284 ± 0.166, respectively. The results show that the proposed method can serve for cuffless and touchless BP measurement.

An Overview of the Sensors for Heart Rate Monitoring Used in Extramural Applications

This work presents an overview of the main strategies that have been proposed for non-invasive monitoring of heart rate (HR) in extramural and home settings. We discuss three categories of sensing according to what physiological effect is used to measure the pulsatile activity of the heart, and we focus on an illustrative sensing modality for each of them. Therefore, electrocardiography, photoplethysmography, and mechanocardiography are presented as illustrative modalities to sense electrical activity, mechanical activity, and the peripheral effect of heart activity. In this paper, we describe the physical principles underlying the three categories and the characteristics of the different types of sensors that belong to each class, and we touch upon the most used software strategies that are currently adopted to effectively and reliably extract HR. In addition, we investigate the strengths and weaknesses of each category linked to the different applications in order to provide the reader with guidelines for selecting the most suitable solution according to the requirements and constraints of the application.

Establishing best practices in photoplethysmography signal acquisition and processing

Photoplethysmography is now widely utilised by clinical devices such as pulse oximeters, and wearable devices such as smartwatches. It holds great promise for health monitoring in daily life. This editorial considers whether it would be possible and beneficial to establish best practices for photoplethysmography signal acquisition and processing. It reports progress made towards this, balanced with the challenges of working with a diverse range of photoplethysmography device designs and intended applications, each of which could benefit from different approaches to signal acquisition and processing. It concludes that there are several potential benefits to establishing best practices. However, it is not yet clear whether it is possible to establish best practices which hold across the range of photoplethysmography device designs and applications.

Using noninvasive adjusted pulse transit time for tracking beat-to-beat systolic blood pressure during ventricular arrhythmia

Tracking beat-to-beat blood pressure noninvasively during ventricular arrhythmia (VA) is of great importance but rarely reported. The goal of our study was to investigate the potential utility of the adjusted pulse transit time (APTT) to track beat-to-beat femoral systolic blood pressure (SBP) during VA. Patients who underwent radiofrequency ablation for arrhythmias at Fuwai Hospital were enrolled. Electrocardiograms (ECGs), finger photoplethysmograms, and femoral arterial blood pressure were recorded simultaneously during VA. The APTT was calculated as the ratio between the square of the conventional pulse transit time (cPTT) and the RR interval of the ECG waveform. Forty-five patients were enrolled in our study, and 22,849 beats were collected during their VA. The inverse of the APTT showed a good correlation with femoral SBP during VA (r = 0.70 ± 0.18). The APTT-derived SBP demonstrated acceptable accuracy in terms of the mean difference ± standard deviation (-0.01 ± 10.54 mmHg) from the invasive femoral SBP. The area under the receiver operating characteristic (ROC) curve for the ability of the APTT to detect ≥30% decreases in femoral SBP was 0.903 (95% confidential interval, 0.895-0.911). In addition, the APTT performed better than the cPTT and RR interval in the above analysis (all P < 0.05). Therefore, the APTT has acceptable accuracy in tracking beat-to-beat femoral SBP and could detect substantially decreased femoral SBP. These findings indicate that the APTT may be a promising noninvasive surrogate for invasive femoral SBP during VA. A multiparameter model combining APTT and other parameters is needed to further improve the accuracy.

Continuous 24-h Photoplethysmogram Monitoring Enables Detection of Atrial Fibrillation

Aim: Atrial fibrillation (AF) detection is challenging because it is often asymptomatic and paroxysmal. We evaluated continuous photoplethysmogram (PPG) for signal quality and detection of AF.

Methods: PPGs were recorded using a wrist-band device in 173 patients (76 AF, 97 sinus rhythm, SR) for 24 h. Simultaneously recorded 3-lead ambulatory ECG served as control. The recordings were split into 10-, 20-, 30-, and 60-min time-frames. The sensitivity, specificity, and F1-score of AF detection were evaluated for each time-frame. AF alarms were generated to simulate continuous AF monitoring. Sensitivities, specificities, and positive predictive values (PPVs) of the alarms were evaluated. User experiences of PPG and ECG recordings were assessed. The study was registered in the Clinical Trials database (NCT03507335).

Results: The quality of PPG signal was better during night-time than in daytime (67.3 ± 22.4% vs. 30.5 ± 19.4%, p < 0.001). The 30-min time-frame yielded the highest F1-score (0.9536), identifying AF correctly in 72/76 AF patients (sensitivity 94.7%), only 3/97 SR patients receiving a false AF diagnosis (specificity 96.9%). The sensitivity and PPV of the simulated AF alarms were 78.2 and 97.2% at night, and 49.3 and 97.0% during the daytime. 82% of patients were willing to use the device at home.

Conclusion: PPG wrist-band provided reliable AF identification both during daytime and night-time. The PPG data’s quality was better at night. The positive user experience suggests that wearable PPG devices could be feasible for continuous rhythm monitoring.

Age-Related Changes in the Characteristics of the Elderly Females Using the Signal Features of an Earlobe Photoplethysmogram

Non-invasive measurement of physiological parameters and indicators, specifically among the elderly, is of utmost importance for personal health monitoring. In this study, we focused on photoplethysmography (PPG), and developed a regression model that calculates variables from the second (SDPPG) and third (TDPPG) derivatives of the PPG pulse that can observe the inflection point of the pulse wave measured by a wearable PPG device. The PPG pulse at the earlobe was measured for 3 min in 84 elderly Korean women (age: 71.19 ± 6.97 years old). Based on the PPG-based cardiovascular function, we derived additional variables from TDPPG, in addition to the aging variable to predict the age. The Aging Index (AI) from SDPPG and Sum of TDPPG variables were calculated in the second and third differential forms of PPG. The variables that significantly correlated with age were c/a, Tac, AI of SDPPG, sum of TDPPG, and correlation coefficient ‘r’ of the model. In multiple linear regression analysis, the r value of the model was 0.308, and that using deep learning on the model was 0.839. Moreover, the possibility of improving the accuracy of the model using supervised deep learning techniques, rather than the addition of datasets, was confirmed.

Estimating Resting HRV during fMRI: A Comparison between Laboratory and Scanner Environment

Heart rate variability (HRV) is regularly assessed in neuroimaging studies as an indicator of autonomic, emotional or cognitive processes. In this study, we investigated the influence of a loud and cramped environment during magnetic resonance imaging (MRI) on resting HRV measures. We compared recordings during functional MRI sessions with recordings in our autonomic laboratory (LAB) in 101 healthy subjects. In the LAB, we recorded an electrocardiogram (ECG) and a photoplethysmogram (PPG) over 15 min. During resting state functional MRI, we acquired a PPG for 15 min. We assessed anxiety levels before the scanning in each subject. In 27 participants, we performed follow-up sessions to investigate a possible effect of habituation. We found a high intra-class correlation ranging between 0.775 and 0.996, indicating high consistency across conditions. We observed no systematic influence of the MRI environment on any HRV index when PPG signals were analyzed. However, SDNN and RMSSD were significantly higher when extracted from the PPG compared to the ECG. Although we found a significant correlation of anxiety and the decrease in HRV from LAB to MRI, a familiarization session did not change the HRV outcome. Our results suggest that psychological factors are less influential on the HRV outcome during MRI than the methodological choice of the cardiac signal to analyze.

A Continuous Blood Pressure Estimation Method Using Photoplethysmography by GRNN-Based Model

Compared with diastolic blood pressure (DBP) and systolic blood pressure (SBP), the blood pressure (BP) waveform contains richer physiological information that can be used for disease diagnosis. However, most models based on photoplethysmogram (PPG) signals can only estimate SBP and DBP and are susceptible to noise signals. We focus on estimating the BP waveform rather than discrete BP values. We propose a model based on a generalized regression neural network to estimate the BP waveform, SBP and DBP. This model takes the raw PPG signal as input and BP waveform as output. The SBP and DBP are extracted from the estimated BP waveform. In addition, the model contains encoders and decoders, and their role is to be responsible for the conversion between the time domain and frequency domain of the waveform. The prediction results of our model show that the mean absolute error is 3.96 ± 5.36 mmHg for SBP and 2.39 ± 3.28 mmHg for DBP, the root mean square error is 5.54 for SBP and 3.45 for DBP. These results fulfill 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 model can effectively estimate the BP waveform only using the raw PPG signal.

Classification of Gamers Using Multiple Physiological Signals: Distinguishing Features of Internet Gaming Disorder

The proliferating and excessive use of internet games has caused various comorbid diseases, such as game addiction, which is now a major social problem. Recently, the American Psychiatry Association classified “Internet gaming disorder (IGD)” as an addiction/mental disorder. Although many studies have been conducted on the diagnosis, treatment, and prevention of IGD, screening studies for IGD are still scarce. In this study, we classified gamers using multiple physiological signals to contribute to the treatment and prevention of IGD. Participating gamers were divided into three groups based on Young’s Internet Addiction Test score and average game time as follows: Group A, those who rarely play games; Group B, those who enjoy and play games regularly; and Group C, those classified as having IGD. In our game-related cue-based experiment, we obtained self-reported craving scores and multiple physiological data such as electrooculogram (EOG), photoplethysmogram (PPG), and electroencephalogram (EEG) from the users while they watched neutral (natural scenery) or stimulating (gameplay) videos. By analysis of covariance (ANCOVA), 13 physiological features (vertical saccadic movement from EOG, standard deviation of N-N intervals, and PNN50 from PPG, and many EEG spectral power indicators) were determined to be significant to classify the three groups. The classification was performed using a 2-layers feedforward neural network. The fusion of three physiological signals showed the best result compared to other cases (combination of EOG and PPG or EEG only). The accuracy was 0.90 and F-1 scores were 0.93 (Group A), 0.89 (Group B), and 0.88 (Group C). However, the subjective self-reported scores did not show a significant difference among the three groups by ANCOVA analysis. The results indicate that the fusion of physiological signals can be an effective method to objectively classify gamers.

Estimation of Continuous Blood Pressure from PPG via a Federated Learning Approach

Ischemic heart disease is the highest cause of mortality globally each year. This puts a massive strain not only on the lives of those affected, but also on the public healthcare systems. To understand the dynamics of the healthy and unhealthy heart, doctors commonly use an electrocardiogram (ECG) and blood pressure (BP) readings. These methods are often quite invasive, particularly when continuous arterial blood pressure (ABP) readings are taken, and not to mention very costly. Using machine learning methods, we develop a framework capable of inferring ABP from a single optical photoplethysmogram (PPG) sensor alone. We train our framework across distributed models and data sources to mimic a large-scale distributed collaborative learning experiment that could be implemented across low-cost wearables. Our time-series-to-time-series generative adversarial network (T2TGAN) is capable of high-quality continuous ABP generation from a PPG signal with a mean error of 2.95 mmHg and a standard deviation of 19.33 mmHg when estimating mean arterial pressure on a previously unseen, noisy, independent dataset. To our knowledge, this framework is the first example of a GAN capable of continuous ABP generation from an input PPG signal that also uses a federated learning methodology.

Development of a Finger-Ring-Shaped Hybrid Smart Stethoscope for Automatic S1 and S2 Heart Sound Identification

Cardiac auscultation is one of the most popular diagnosis approaches to determine cardiovascular status based on listening to heart sounds with a stethoscope. However, heart sounds can be masked by visceral sounds such as organ movement and breathing, and a doctor's level of experience can more seriously affect the accuracy of auscultation results. To improve the accuracy of auscultation, and to allow nonmedical staff to conduct cardiac auscultation anywhere and anytime, a hybrid-type personal smart stethoscope with an automatic heart sound analysis function is presented in this paper. The device was designed with a folding finger-ring shape that can be worn on the finger and placed on the chest to measure photoplethysmogram (PPG) signals and acquire the heart sound simultaneously. The measured heart sounds are detected as phonocardiogram (PCG) signals, and the boundaries of the heart sound variation and the peaks of the PPG signal are detected in preprocessing by an advanced Shannon entropy envelope. According to the relationship between PCG and PPG signals, an automatic heart sound analysis algorithm based on calculating the time interval between the first and second heart sounds (S1, S2) and the peak of the PPG was developed and implemented via the manufactured prototype device. The prototype device underwent accuracy and usability testing with 20 young adults, and the experimental results showed that the proposed smart stethoscope could satisfactorily collect the heart sounds and PPG signals. In addition, within the developed algorithm, the device was as accurate in start-points of heart sound detection as professional physiological signal-acquisition systems. Furthermore, the experimental results demonstrated that the device was able to identify S1 and S2 heart sounds automatically with high accuracy.

Assessment of Non-Invasive Blood Pressure Prediction from PPG and rPPG Signals Using Deep Learning

Exploiting photoplethysmography signals (PPG) for non-invasive blood pressure (BP) measurement is interesting for various reasons. First, PPG can easily be measured using fingerclip sensors. Second, camera based approaches allow to derive remote PPG (rPPG) signals similar to PPG and therefore provide the opportunity for non-invasive measurements of BP. Various methods relying on machine learning techniques have recently been published. Performances are often reported as the mean average error (MAE) on the data which is problematic. This work aims to analyze the PPG- and rPPG based BP prediction error with respect to the underlying data distribution. First, we train established neural network (NN) architectures and derive an appropriate parameterization of input segments drawn from continuous PPG signals. Second, we use this parameterization to train NNs with a larger PPG dataset and carry out a systematic evaluation of the predicted blood pressure. The analysis revealed a strong systematic increase of the prediction error towards less frequent BP values across NN architectures. Moreover, we tested different train/test set split configurations which underpin the importance of a careful subject-aware dataset assignment to prevent overly optimistic results. Third, we use transfer learning to train the NNs for rPPG based BP prediction. The resulting performances are similar to the PPG-only case. Finally, we apply different personalization techniques and retrain our NNs with subject-specific data for both the PPG-only and rPPG case. Whilst the particular technique is less important, personalization reduces the prediction errors significantly.

Recent Research for Unobtrusive Atrial Fibrillation Detection Methods Based on Cardiac Dynamics Signals: A Survey

Atrial fibrillation (AF) is the most common cardiac arrhythmia. It tends to cause multiple cardiac conditions, such as cerebral artery blockage, stroke, and heart failure. The morbidity and mortality of AF have been progressively increasing over the past few decades, which has raised widespread concern about unobtrusive AF detection in routine life. The up-to-date non-invasive AF detection methods include electrocardiogram (ECG) signals and cardiac dynamics signals, such as the ballistocardiogram (BCG) signal, the seismocardiogram (SCG) signal and the photoplethysmogram (PPG) signal. Cardiac dynamics signals can be collected by cushions, mattresses, fabrics, or even cameras, which is more suitable for long-term monitoring. Therefore, methods for AF detection by cardiac dynamics signals bring about extensive attention for recent research. This paper reviews the current unobtrusive AF detection methods based on the three cardiac dynamics signals, summarized as data acquisition and preprocessing, feature extraction and selection, classification and diagnosis. In addition, the drawbacks and limitations of the existing methods are analyzed, and the challenges in future work are discussed.

Deep learning enables sleep staging from photoplethysmogram for patients with suspected sleep apnea

Study Objectives

Accurate identification of sleep stages is essential in the diagnosis of sleep disorders (e.g. obstructive sleep apnea [OSA]) but relies on labor-intensive electroencephalogram (EEG)-based manual scoring. Furthermore, long-term assessment of sleep relies on actigraphy differentiating only between wake and sleep periods without identifying specific sleep stages and having low reliability in identifying wake periods after sleep onset. To address these issues, we aimed to develop an automatic method for identifying the sleep stages from the photoplethysmogram (PPG) signal obtained with a simple finger pulse oximeter.

Methods

PPG signals from the diagnostic polysomnographies of susptected OSA patients (n = 894) were utilized to develop a combined convolutional and recurrent neural network. The deep learning model was trained individually for three-stage (wake/NREM/REM), four-stage (wake/N1+N2/N3/REM), and five-stage (wake/N1/N2/N3/REM) classification of sleep.

Results

The three-stage model achieved an epoch-by-epoch accuracy of 80.1% with Cohen’s κ of 0.65. The four- and five-stage models achieved 68.5% (κ = 0.54), and 64.1% (κ = 0.51) accuracies, respectively. With the five-stage model, the total sleep time was underestimated with a mean bias error (SD) of of 7.5 (55.2) minutes.

Conclusion

The PPG-based deep learning model enabled accurate estimation of sleep time and differentiation between sleep stages with a moderate agreement to manual EEG-based scoring. As PPG is already included in ambulatory polygraphic recordings, applying the PPG-based sleep staging could improve their diagnostic value by enabling simple, low-cost, and reliable monitoring of sleep and help assess otherwise overlooked conditions such as REM-related OSA.

Cuffless blood pressure estimation methods: physiological model parameters versus machine-learned features

Objective.For the first time in the literature, this paper investigates some crucial aspects of blood pressure (BP) monitoring using photoplethysmogram (PPG) and electrocardiogram (ECG). In general, the proposed approaches utilize two types of features: parameters extracted from physiological models or machine-learned features. To provide an overview of the different feature extraction methods, we assess the performance of these features and their combinations. We also explore the importance of the ECG waveform. Although ECG contains critical information, most models merely use it as a time reference. To take this one step further, we investigate the effect of its waveform on the performance.Approach.We extracted 27 commonly used physiological parameters in the literature. In addition, convolutional neural networks (CNNs) were deployed to define deep-learned representations. We applied the CNNs to extract two different feature sets from the PPG segments alone and alongside corresponding ECG segments. Then, the extracted feature vectors and their combinations were fed into various regression models to evaluate our hypotheses.Main results.We performed our evaluations using data collected from 200 subjects. The results were analyzed by the mean difference t-test and graphical methods. Our results confirm that the ECG waveform contains important information and helps us to improve accuracy. The comparison of the physiological parameters and machine-learned features also reveals the superiority of machine-learned representations. Moreover, our results highlight that the combination of these feature sets does not provide any additional information.Significance.We conclude that CNN feature extractors provide us with concise and precise representations of ECG and PPG for BP monitoring.

A clinical set-up for noninvasive blood pressure monitoring using two photoplethysmograms and based on convolutional neural networks

Blood pressure is a reliable indicator of many cardiac arrhythmias and rheological problems. This study proposes a clinical set-up using conventional monitoring systems to estimate systolic and diastolic blood pressures continuously based on two photoplethysmogram signals (PPG) taken from the earlobe and toe. Several amendments were applied to conventional clinical monitoring devices to construct our project plan. We used two monitors to acquire two PPGs, one ECG, and invasive blood pressure as the reference to evaluate the estimation accuracy. One of the most critical requirements was the synchronization of the acquired signals that was accomplished by using ECG as the time reference. Following data acquisition and preparation procedures, the performance of each PPG signal alone and together was investigated using deep convolutional neural networks. The proposed architecture was evaluated on 32 records acquired from 14 patients after cardiovascular surgery. The results showed a better performance for toe PPG in comparison with earlobe PPG. Moreover, they indicated the algorithm accuracy improves if both signals are applied together to the network. According to the British Hypertension Society standards, the results achieved grade A for both blood pressure measurements. The mean and standard deviation of estimation errors were +0.3 ± 4.9 and +0.1 ± 3.2 mmHg for systolic and diastolic BPs, respectively. Since the method is based on conventional monitoring equipment and provides a high estimation consistency, it can be considered as a possible alternative for inconvenient invasive BP monitoring in clinical environments.

Recurrence Plot and Machine Learning for Signal Quality Assessment of Photoplethysmogram in Mobile Environment

The purpose of this study was to develop a machine learning model that could accurately evaluate the quality of a photoplethysmogram based on the shape of the photoplethysmogram and the phase relevance in a pulsatile waveform without requiring complicated pre-processing. Photoplethysmograms were recorded for 76 participants (5 min for each participant). All recorded photoplethysmograms were segmented for each beat to obtain a total of 49,561 pulsatile segments. These pulsatile segments were manually labeled as ‘good’ and ‘poor’ classes and converted to a two-dimensional phase space trajectory image using a recurrence plot. The classification model was implemented using a convolutional neural network with a two-layer structure. As a result, the proposed model correctly classified 48,827 segments out of 49,561 segments and misclassified 734 segments, showing a balanced accuracy of 0.975. Sensitivity, specificity, and positive predictive values of the developed model for the test dataset with a ‘poor’ class classification were 0.964, 0.987, and 0.848, respectively. The area under the curve was 0.994. The convolutional neural network model with recurrence plot as input proposed in this study can be used for signal quality assessment as a generalized model with high accuracy through data expansion. It has an advantage in that it does not require complicated pre-processing or a feature detection process.

Novel Analgesic Index for Postoperative Pain Assessment Based on a Photoplethysmographic Spectrogram and Convolutional Neural Network: Observational Study

Background

Although commercially available analgesic indices based on biosignal processing have been used to quantify nociception during general anesthesia, their performance is low in conscious patients. Therefore, there is a need to develop a new analgesic index with improved performance to quantify postoperative pain in conscious patients.

Objective

This study aimed to develop a new analgesic index using photoplethysmogram (PPG) spectrograms and a convolutional neural network (CNN) to objectively assess pain in conscious patients.

Methods

PPGs were obtained from a group of surgical patients for 6 minutes both in the absence (preoperatively) and in the presence (postoperatively) of pain. Then, the PPG data of the latter 5 minutes were used for analysis. Based on the PPGs and a CNN, we developed a spectrogram–CNN index for pain assessment. The area under the curve (AUC) of the receiver-operating characteristic curve was measured to evaluate the performance of the 2 indices.

Results

PPGs from 100 patients were used to develop the spectrogram–CNN index. When there was pain, the mean (95% CI) spectrogram–CNN index value increased significantly—baseline: 28.5 (24.2-30.7) versus recovery area: 65.7 (60.5-68.3); P<.01. The AUC and balanced accuracy were 0.76 and 71.4%, respectively. The spectrogram–CNN index cutoff value for detecting pain was 48, with a sensitivity of 68.3% and specificity of 73.8%.

Conclusions

Although there were limitations to the study design, we confirmed that the spectrogram–CNN index can efficiently detect postoperative pain in conscious patients. Further studies are required to assess the spectrogram–CNN index’s feasibility and prevent overfitting to various populations, including patients under general anesthesia.

Trial Registration

Clinical Research Information Service KCT0002080; https://cris.nih.go.kr/cris/search/search_result_st01.jsp?seq=6638

Towards Continuous and Ambulatory Blood Pressure Monitoring: Methods for Efficient Data Acquisition for Pulse Transit Time Estimation

We developed a prototype for measuring physiological data for pulse transit time (PTT) estimation that will be used for ambulatory blood pressure (BP) monitoring. The device is comprised of an embedded system with multimodal sensors that streams high-throughput data to a custom Android application. The primary focus of this paper is on the hardware–software codesign that we developed to address the challenges associated with reliably recording data over Bluetooth on a resource-constrained platform. In particular, we developed a lossless compression algorithm that is based on optimally selective Huffman coding and Huffman prefixed coding, which yields virtually identical compression ratios to the standard algorithm, but with a 67–99% reduction in the size of the compression tables. In addition, we developed a hybrid software–hardware flow control method to eliminate microcontroller (MCU) interrupt-latency related data loss when multi-byte packets are sent from the phone to the embedded system via a Bluetooth module at baud rates exceeding 115,200 bit/s. The empirical error rate obtained with the proposed method with the baud rate set to 460,800 bit/s was identically equal to 0%. Our robust and computationally efficient physiological data acquisition system will enable field experiments that will drive the development of novel algorithms for PTT-based continuous BP monitoring.

Acquiring Wearable Photoplethysmography Data in Daily Life: The PPG Diary Pilot Study †

The photoplethysmogram (PPG) signal is widely measured by smart watches and fitness bands for heart rate monitoring. New applications of the PPG are also emerging, such as to detect irregular heart rhythms, track infectious diseases, and monitor blood pressure. Consequently, datasets of PPG signals acquired in daily life are valuable for algorithm development. The aim of this pilot study was to assess the feasibility of acquiring PPG data in daily life. A single subject was asked to wear a wrist-worn PPG sensor six days a week for four weeks, and to keep a diary of daily activities. The sensor was worn for 75.0% of the time, signals were acquired for 60.6% of the time, and signal quality was high for 30.5% of the time. This small pilot study demonstrated the feasibility of acquiring PPG data during daily living. Key lessons were learnt for future studies: (i) devices which are waterproof and require charging less frequently may provide signals for a greater proportion of the time; (ii) data should either be stored on the device or streamed via a reliable connection to a second device for storage; (iii) it may be beneficial to acquire signals during the night or during periods of low activity to achieve high signal quality; and (iv) there are several promising areas for PPG algorithm development including the design of pulse wave analysis techniques to track changes in cardiovascular properties in daily life.

Determination of hypertension disease using chirp z-transform and statistical features of optimal band-pass filtered short-time photoplethysmography signals

Hypertension is the condition where the normal blood pressure is high. This situation is manifested by the high pressure of the blood in the vein towards the vessel wall. Hypertension mostly affects the brain, kidneys, eyes, arteries and heart. Therefore, the diagnosis of this common disease is important. It may take days, weeks or even months for diagnosis. Often a device, called a blood pressure holter, is connected to the person for 24 or 48 h and the person's blood pressure is recorded at certain intervals. Diagnosis can be made by the specialist physician considering these results. In recent years, various physiological measurement techniques have been used to accelerate this time-consuming diagnostic phase and intelligent models have been proposed. One of these techniques is photopletesmography (PPG). In this study, a model for the detection of hypertension disease in individuals was proposed using chirp z-transform and statistical features (total band power, autoregressive model parameters, standard deviation of signal's derivative and zero crossing rate) of optimal band-pass filtered short-time PPG signals. The proposed method was successfully applied to 657 PPG trials, which each of them had only 2.1 s signal length and achieved a classification accuracy rate of 77.52% on the test data. The results showed that the diagnosis of hypertension can be performed effectively by chirp z-transform and statistical features and support vector machine classifier using optimal frequency range of 1.4-6 Hz.

Continuous Blood Pressure Estimation From Electrocardiogram and Photoplethysmogram During Arrhythmias

Objective

Continuous blood pressure (BP) provides valuable information for the disease management of patients with arrhythmias. The traditional intra-arterial method is too invasive for routine healthcare settings, whereas cuff-based devices are inferior in reliability and comfortable for long-term BP monitoring during arrhythmias. The study aimed to investigate an indirect method for continuous and cuff-less BP estimation based on electrocardiogram (ECG) and photoplethysmogram (PPG) signals during arrhythmias and to test its reliability for the determination of BP using invasive BP (IBP) as reference.

Methods

Thirty-five clinically stable patients (15 with ventricular arrhythmias and 20 with supraventricular arrhythmias) who had undergone radiofrequency ablation were enrolled in this study. Their ECG, PPG, and femoral arterial IBP signals were simultaneously recorded with a multi-parameter monitoring system. Fifteen features that have the potential ability in indicating beat-to-beat BP changes during arrhythmias were extracted from the ECG and PPG signals. Four machine learning algorithms, decision tree regression (DTR), support vector machine regression (SVR), adaptive boosting regression (AdaboostR), and random forest regression (RFR), were then implemented to develop the BP models.

Results

The results showed that the mean value ± standard deviation of root mean square error for the estimated systolic BP (SBP), diastolic BP (DBP) with the RFR model against the reference in all patients were 5.87 ± 3.13 and 3.52 ± 1.38 mmHg, respectively, which achieved the best performance among all the models. Furthermore, the mean error ± standard deviation of error between the estimated SBP and DBP with the RFR model against the reference in all patients were −0.04 ± 6.11 and 0.11 ± 3.62 mmHg, respectively, which complied with the Association for the Advancement of Medical Instrumentation and the British Hypertension Society (Grade A) standards.

Conclusion

The results indicated that the utilization of ECG and PPG signals has the potential to enable cuff-less and continuous BP estimation in an indirect way for patients with arrhythmias.

Derivation of Breathing Metrics From a Photoplethysmogram at Rest: Machine Learning Methodology

Background

There has been a recent increased interest in monitoring health using wearable sensor technologies; however, few have focused on breathing. The ability to monitor breathing metrics may have indications both for general health as well as respiratory conditions such as asthma, where long-term monitoring of lung function has shown promising utility.

Objective

In this paper, we explore a long short-term memory (LSTM) architecture and predict measures of interbreath intervals, respiratory rate, and the inspiration-expiration ratio from a photoplethysmogram signal. This serves as a proof-of-concept study of the applicability of a machine learning architecture to the derivation of respiratory metrics.

Methods

A pulse oximeter was mounted to the left index finger of 9 healthy subjects who breathed at controlled respiratory rates. A respiratory band was used to collect a reference signal as a comparison.

Results

Over a 40-second window, the LSTM model predicted a respiratory waveform through which breathing metrics could be derived with a bias value and 95% CI. Metrics included inspiration time (–0.16 seconds, –1.64 to 1.31 seconds), expiration time (0.09 seconds, –1.35 to 1.53 seconds), respiratory rate (0.12 breaths per minute, –2.13 to 2.37 breaths per minute), interbreath intervals (–0.07 seconds, –1.75 to 1.61 seconds), and the inspiration-expiration ratio (0.09, –0.66 to 0.84).

Conclusions

A trained LSTM model shows acceptable accuracy for deriving breathing metrics and could be useful for long-term breathing monitoring in health. Its utility in respiratory disease (eg, asthma) warrants further investigation.

Multimodal Photoplethysmography-Based Approaches for Improved Detection of Hypertension

Elevated blood pressure (BP) is a major cause of death, yet hypertension commonly goes undetected. Owing to its nature, it is typically asymptomatic until later in its progression when the vessel or organ structure has already been compromised. Therefore, noninvasive and continuous BP measurement methods are needed to ensure appropriate diagnosis and early management before hypertension leads to irreversible complications. Photoplethysmography (PPG) is a noninvasive technology with waveform morphologies similar to that of arterial BP waveforms, therefore attracting interest regarding its usability in BP estimation. In recent years, wearable devices incorporating PPG sensors have been proposed to improve the early diagnosis and management of hypertension. Additionally, the need for improved accuracy and convenience has led to the development of devices that incorporate multiple different biosignals with PPG. Through the addition of modalities such as an electrocardiogram, a final measure of the pulse wave velocity is derived, which has been proved to be inversely correlated to BP and to yield accurate estimations. This paper reviews and summarizes recent studies within the period 2010–2019 that combined PPG with other biosignals and offers perspectives on the strengths and weaknesses of current developments to guide future advancements in BP measurement. Our literature review reveals promising measurement accuracies and we comment on the effective combinations of modalities and success of this technology.

End-to-End Deep Learning Architecture for Continuous Blood Pressure Estimation Using Attention Mechanism

Blood pressure (BP) is a vital sign that provides fundamental health information regarding patients. Continuous BP monitoring is important for patients with hypertension. Various studies have proposed cuff-less BP monitoring methods using pulse transit time. We propose an end-to-end deep learning architecture using only raw signals without the process of extracting features to improve the BP estimation performance using the attention mechanism. The proposed model consisted of a convolutional neural network, a bidirectional gated recurrent unit, and an attention mechanism. The model was trained by a calibration-based method, using the data of each subject. The performance of the model was compared to the model that used each combination of the three signals, and the model with the attention mechanism showed better performance than other state-of-the-art methods, including conventional linear regression method using pulse transit time (PTT). A total of 15 subjects were recruited, and electrocardiogram, ballistocardiogram, and photoplethysmogram levels were measured. The 95% confidence interval of the reference BP was [86.34, 143.74] and [51.28, 88.74] for systolic BP (SBP) and diastolic BP (DBP), respectively. The R 2 values were 0.52 and 0.49, and the mean-absolute-error values were 4.06 ± 4.04 and 3.33 ± 3.42 for SBP and DBP, respectively. In addition, the results complied with global standards. The results show the applicability of the proposed model as an analytical metric for BP estimation.

Cuffless Single-Site Photoplethysmography for Blood Pressure Monitoring

One in three adults worldwide has hypertension, which is associated with significant morbidity and mortality. Consequently, there is a global demand for continuous and non-invasive blood pressure (BP) measurements that are convenient, easy to use, and more accurate than the currently available methods for detecting hypertension. This could easily be achieved through the integration of single-site photoplethysmography (PPG) readings into wearable devices, although improved reliability and an understanding of BP estimation accuracy are essential. This review paper focuses on understanding the features of PPG associated with BP and examines the development of this technology over the 2010-2019 period in terms of validation, sample size, diversity of subjects, and datasets used. Challenges and opportunities to move single-site PPG forward are also discussed.

SeisMote: A Multi-Sensor Wireless Platform for Cardiovascular Monitoring in Laboratory, Daily Life, and Telemedicine

This article presents a new wearable platform, SeisMote, for the monitoring of cardiovascular function in controlled conditions and daily life. It consists of a wireless network of sensorized nodes providing simultaneous multiple measures of electrocardiogram (ECG), acceleration, rotational velocity, and photoplethysmogram (PPG) from different body areas. A custom low-power transmission protocol was developed to allow the concomitant real-time monitoring of 32 signals (16 bit @200 Hz) from up to 12 nodes with a jitter in the among-node time synchronization lower than 0.2 ms. The BluetoothLE protocol may be used when only a single node is needed. Data can also be collected in the off-line mode. Seismocardiogram and pulse transit times can be derived from the collected data to obtain additional information on cardiac mechanics and vascular characteristics. The employment of the system in the field showed recordings without data gaps caused by transmission errors, and the duration of each battery charge exceeded 16 h. The system is currently used to investigate strategies of hemodynamic regulation in different vascular districts (through a multisite assessment of ECG and PPG) and to study the propagation of precordial vibrations along the thorax. The single-node version is presently exploited to monitor cardiac patients during telerehabilitation.

The intensity of oscillations of the photoplethysmographic waveform variability at frequencies 0.04-0.4 Hz is effective marker of hypertension and coronary artery disease in males

Background: It is believed that the intensity of oscillations in the photoplethysmographic waveform variability reflects the activity of vascular regulatory mechanisms. However, the relationship of such fluctuations with the state of health is poorly understood.Purpose: The aim of our study was to assess the possibility of using spectral indices that reflect the intensity of oscillations of the photoplethysmographic waveform variability at frequencies 0.04-0.4 Hz as markers of hypertension and coronary artery disease. We did not study women to exclude the influence of menopause and sex hormones on the results.Materials and Methods: We compared synchronous 10-minute records of finger photoplethysmogram and respiration at rest in 30 healthy males (48.8 ± 4.5 years; data presented as Mean ± SD) versus 30 patients with hypertension (aged 49.0 ± 4.3 years) versus 30 patients with stable coronary artery disease (49.2 ± 4.8 years). Percentages of high-frequency and low-frequency ranges in the total power of photoplethysmographic waveform variability spectrum (HF% and LF%), and LF/HF ratio were assessed.Results: HF% are subject to by 2- to 5-fold increase in hypertensive patients (p < .001) and up to an 8-fold increase in patients with coronary artery disease (p < .001) when compared with healthy persons. On the contrary, LF% is reduced by 1.5-5 times in all patients when compared with healthy people (p < .001). We identified cut-off points for each photoplethysmographic index to distinguish patients with coronary artery disease or hypertension from healthy subjects. Multiple logistic regression models based on photoplethysmographic waveform variability indices had sufficient sensitivity and specificity for patients with hypertension or coronary artery disease.Conclusion: Frequency-domain indices of photoplethysmographic waveform variability (in particular, HF%, LF%, and LF/HF) are sufficiently sensitive and specific markers of hypertension and coronary artery disease in adult males.

Blood Pressure Estimation from Photoplethysmogram Using a Spectro-Temporal Deep Neural Network

Blood pressure (BP) is a direct indicator of hypertension, a dangerous and potentially deadly condition. Regular monitoring of BP is thus important, but many people have aversion towards cuff-based devices, and their limitation is that they can only be used at rest. Using just a photoplethysmogram (PPG) to estimate BP is a potential solution investigated in our study. We analyzed the MIMIC III database for high-quality PPG and arterial BP waveforms, resulting in over 700 h of signals after preprocessing, belonging to 510 subjects. We then used the PPG alongside its first and second derivative as inputs into a novel spectro-temporal deep neural network with residual connections. We have shown in a leave-one-subject-out experiment that the network is able to model the dependency between PPG and BP, achieving mean absolute errors of 9.43 for systolic and 6.88 for diastolic BP. Additionally we have shown that personalization of models is important and substantially improves the results, while deriving a good general predictive model is difficult. We have made crucial parts of our study, especially the list of used subjects and our neural network code, publicly available, in an effort to provide a solid baseline and simplify potential comparison between future studies on an explicit MIMIC III subset.