Photoplethysmography Waveforms Research Articles

An algorithm to detect dicrotic notch in arterial blood pressure and photoplethysmography waveforms using the iterative envelope mean method

Background and objectiveDetection of the dicrotic notch (DN) within a cardiac cycle is essential for assessment of cardiac output, calculation of pulse wave velocity, estimation of left ventricular ejection time, and supporting feature-based machine learning models for noninvasive blood pressure estimation, and hypotension, or hypertension prediction. In this study, we present a new algorithm based on the iterative envelope mean (IEM) method to detect automatically the DN in arterial blood pressure (ABP) and photoplethysmography (PPG) waveforms. MethodsThe algorithm was evaluated on both ABP and PPG waveforms from a large perioperative dataset (MLORD dataset) comprising 17,327 patients. The analysis involved a total of 1,171,288 cardiac cycles for ABP waveforms and 3,424,975 cardiac cycles for PPG waveforms. To evaluate the algorithm's performance, the systolic phase duration (SPD) was employed, which represents the duration from the onset of the systolic phase to the DN in the cardiac cycle. Correlation plots and regression analysis were used to compare the algorithm against marked DN detection, while box plots and Bland-Altman plots were used to compare its performance with both marked DN detection and an established DN detection technique (second derivative). The marking of the DN temporal location was carried out by an experienced researcher using the help of the 'find_peaks' function from the scipy Python package, serving as a reference for the evaluation. The marking was visually validated by both an engineer and an anesthesiologist. The robustness of the algorithm was evaluated as the DN was made less visually distinct across signal-to-noise ratios (SNRs) ranging from -30 dB to -5 dB in both ABP and PPG waveforms. ResultsThe correlation between SPD estimated by the algorithm and that marked by the researcher is strong for both ABP (R2(87,343) =0.99, p<.001) and PPG (R2(86,764) =0.98, p<.001) waveforms. The algorithm had a lower mean error of DN detection (s): 0.0047 (0.0029) for ABP waveforms and 0.0046 (0.0029) for PPG waveforms, compared to 0.0693 (0.0770) for ABP and 0.0968 (0.0909) for PPG waveforms for the established 2nd derivative method. The algorithm has high rate of detectability of DN detection for SNR of >= -9 dB for ABP waveforms and >= -12 dB for PPG waveforms indicating robust performance in detecting the DN when it is less visibly distinct. ConclusionOur proposed IEM- based algorithm can detect DN in both ABP and PPG waveforms with low computational cost, even in cases where it is not distinctly defined within a cardiac cycle of the waveform (‘DN-less signals’). The algorithm can potentially serve as a valuable, fast, and reliable tool for extracting features from ABP and PPG waveforms. It can be especially beneficial in medical applications where DN-based features, such as SPD, diastolic phase duration, and DN amplitude, play a significant role.

Machine-learning guided differentiation between photoplethysmography waveforms of supraventricular and ventricular origin

Abstract Aims To test, whether a neural network-based classifier could aid in distinguishing photoplethysmographic (PPG) pulse waveforms of ventricular from those of supraventricular origin. Methods Thirty patients undergoing invasive electrophysiological studies for documented narrow complex tachycardias between May and December 2021 at a university medical centre were included in the trial. PPG waveforms were recorded using a PPG wristband (Empatica E4) in parallel to 12-lead surface electrograms (ECGs) and intracardiac electrograms from diagnostic catheters. Pulse waves were then annotated to either supraventricular (sinus rhythm, atrial pacing) or ventricular origin (ventricular pacing) based on invasive electrogram recordings, ECGs and stimulation protocols. 21147 data sets were split into test, training and validation data sets and used to develop, train and validate a convolutional neural network (CNN) for classifying PPG waveforms according to their origin into sinus rhythm, atrial or ventricular pacing. Results Datasets were complete for 27 patients. 74% were female and median age was 53 (range 18, 78) years, median BMI was 27±5, 19% had arterial hypertension, 11% had diabetes and 15% had coronary artery disease. During the electrophysiological study, typical AV-nodal re-entrant tachycardias could be induced in 63% of patients, 15% had inducible atrial tachycardias and the other patients had no inducible tachyarrhythmias. Our CNN-based classifier was able to correctly predict the origins of pulse waves in 95±1% for sinus rhythm, 96±1% for atrial paced rhythms and 95±1% for ventricular rhythms. Conclusion A neural network-based classifier trained on ground truth PPG data obtained by using invasive EP measurements for annotation could accurately predict atrial or ventricular origin from PPG waveforms alone.Graphical abstract

Effective IoT Based Analysis of Photoplethysmography Waveforms for Investigating Arterial Stiffness and Pulse Rate Variability

Effective IoT Based Analysis of Photoplethysmography Waveforms for Investigating Arterial Stiffness and Pulse Rate Variability

Physically Meaningful Surrogate Data for COPD.

The rapidly increasing prevalence of debilitating breathing disorders, such as chronic obstructive pulmonary disease (COPD), calls for a meaningful integration of artificial intelligence (AI) into respiratory healthcare. Deep learning techniques are "data hungry" whilst patient-based data is invariably expensive and time consuming to record. To this end, we introduce a novel COPD-simulator, a physical apparatus with an easy to replicate design which enables rapid and effective generation of a wide range of COPD-like data from healthy subjects, for enhanced training of deep learning frameworks. To ensure the faithfulness of our domain-aware COPD surrogates, the generated waveforms are examined through both flow waveforms and photoplethysmography (PPG) waveforms (as a proxy for intrathoracic pressure) in terms of duty cycle, sample entropy, FEV1/FVC ratios and flow-volume loops. The proposed simulator operates on healthy subjects and is able to generate FEV1/FVC obstruction ratios ranging from greater than 0.8 to less than 0.2, mirroring values that can observed in the full spectrum of real-world COPD. As a final stage of verification, a simple convolutional neural network is trained on surrogate data alone, and is used to accurately detect COPD in real-world patients. When training solely on surrogate data, and testing on real-world data, a comparison of true positive rate against false positive rate yields an area under the curve of 0.75, compared with 0.63 when training solely on real-world data.

Deep learning classification of systemic sclerosis from multi-site photoplethysmography signals.

Introduction: A pilot study assessing a novel approach to identify patients with Systemic Sclerosis (SSc) using deep learning analysis of multi-site photoplethysmography (PPG) waveforms ("DL-PPG"). Methods: PPG recordings having baseline, unilateral arm pressure cuff occlusion and reactive hyperaemia flush phases from 6 body sites were studied in 51 Controls and 20 SSc patients. RGB scalogram images were obtained from the PPG, using the continuous wavelet transform (CWT). 2 different pre-trained convolutional neural networks (CNNs, namely, GoogLeNet and EfficientNetB0) were trained to classify the SSc and Control groups, evaluating their performance using 10-fold stratified cross validation (CV). Their classification performance (i.e., accuracy, sensitivity, and specificity, with 95% confidence intervals) was also compared to traditional machine learning (ML), i.e., Linear Discriminant Analysis (LDA) and K-Nearest Neighbour (KNN). Results: On a participant basis DL-PPG accuracy, sensitivity and specificity for GoogLeNet were 83.1 (72.3-90.9), 75.0 (50.9-91.3) and 86.3 (73.7-94.3)% respectively, and for EfficientNetB0 were 87.3 (77.2-94.0), 80.0 (56.3-94.3) and 90.1 (78.6-96.7)%. The corresponding results for ML classification using LDA were 66.2 (53.9-77.0), 65.0 (40.8-84.6) and 66.7 (52.1-79.2)% respectively, and for KNN were 76.1 (64.5-85.4), 40.0 (19.1-63.9), and 90.2 (78.6-96.7)% respectively. Discussion: This study shows the potential of DL-PPG classification using CNNs to detect SSc. EfficientNetB0 gave an overall improved performance compared to GoogLeNet, with both CNNs performing better than the traditional ML methods tested. Our automatic AI approach, using transfer learning, could offer significant benefits for SSc diagnostics in a variety of clinical settings where low-cost portable and easy-to-use diagnostics can be beneficial.

Deep learning identification of coronary artery disease from bilateral finger photoplethysmography sensing: A proof-of-concept study

BackgroundA proof-of-concept study assessing a novel approach to identify patients with coronary artery disease (CAD) using deep learning analysis of bilateral-site photoplethysmography (PPG) waveforms (“DL-PPG”). MethodologyDL-PPG was studied in 37 participants (with 21 having CAD). Scalogram ‘spectral’ images were derived from right and left index finger PPG measurements collected using a 3-phase protocol (baseline, unilateral arm pressure cuff occlusion, reactive hyperaemia flush). Artificial Intelligence (AI) analysis, namely deep learning, was employed for scalogram image classification using a Convolutional Neural Network (CNN, “GoogLeNet”), with classification performance obtained using 10-fold stratified cross validation (CV). A conventional machine learning (ML) classifier (K-nearest neighbour, K-NN, K = 9) was also evaluated for comparison with the CNN deep learning methodology. Blood samples were also collected giving 2 biochemical biomarkers of endothelial function. Test sensitivities, specificities, accuracies, and Kappa statistics were determined. ResultsDL-PPG sensitivity was 80.9 % (95% CI, 78.6–83.0), specificity 87.7% (85.5–89.7), accuracy 83.8 % (82.2–85.3), and Kappa 0.68 (0.65–0.71). Comparative K-NN ML performance was 69.4% (95% CI, 68.7–70.1), 37.5% (36.7–38.2), 53.9% (53.3–54.4), and 0.069 (0.058–0.079), respectively. No differences between patients and controls were found for the biochemical biomarkers of endothelial function. ConclusionSubstantial overall agreement was found between DL-PPG classification and CAD angiography, with DL-PPG performance clearly better than for a conventional ML technique. Our deep learning classification approach, using only basic pre-processing of the PPG pulse waveforms before classification, could offer significant benefits for the diagnosis of CAD in a variety of clinical settings needing low-cost portable and easy-to-use diagnostics.

Non-Contact Physiological Measurement System for Wearing Masks During the Epidemic

Physiological signals indicate a person’s physical and mental state at any given time. Accordingly, many studies extract physiological signals from the human body with non-contact methods, and most of them require facial feature points. However, under COVID-19, wearing a mask has become a must in many places, so how non-contact physiological information measurements can still be performed correctly even when a mask covers the facial information has become a focus of research. In this study, RGB and thermal infrared cameras were used to execute non-contact physiological information measurement systems for heart rate, blood pressure, respiratory rate, and forehead temperature for people wearing masks due to the pandemic. Using the green (G) minus red (R) signal in the RGB image, the region of interest (ROI) is established in the forehead and nose bridge regions. The photoplethysmography (PPG) waveforms of the two regions are obtained after the acquired PPG signal is subjected to the optical flow method, baseline drift calibration, normalization, and bandpass filtering. The relevant parameters in Deep Neural Networks (DNN) for the regression model can correctly predict the heartbeat and blood pressure. In addition, the temperature change in the ROI of the mask after thermal image processing and filtering can be used to correctly determine the number of breaths. Meanwhile, the thermal image can be used to read the temperature average of the ROI of the forehead, and the forehead temperature can be obtained smoothly. The experimental results show that the above-mentioned physiological signals of a subject can be obtained in 6-s images with the error for both heart rate and blood pressure within 2%~3% and the error of forehead temperature within ±0.5°C.

Characteristics of sarcopenia subjects in arterial pulse spectrum analysis.

AimsSarcopenia is significantly associated with the number of cardiovascular and metabolic diseases, however, the underlying pathophysiological processes are largely unknown. This study performed harmonic index of finger photoplethysmography (PPG) waveforms with the aims of distinguishing different arterial pulse waveform signals between sarcopenia, presarcopenia, dynapenia, and healthy subjects.MethodsSixty-eight subjects were enrolled and obtained 1-min PPG signals, then were assigned to four age-matched groups: control, dynapenia, presarcopenia, and sarcopenia which definition according to Asian Working Group for Sarcopenia (AWGS): 2019 Consensus Update on Sarcopenia Diagnosis and Treatment. Harmonics 1–10 of the PPG waveform were obtained and calculated each of the amplitude proportions (Cn), standard deviations (SDn), coefficients of variations (CVn), and vascular elasticity index (VEI) for to evaluating the blood-pressure harmonic variability.ResultsThe prevalence of sarcopenia in women gender (8 out of 9, 88.9%, p = 0.046) and osteoporosis in dynapenia (7 out of 16, 43.8%, p = 0.005) were significant higher. Among the four groups, compared with control, dynapenia, and presarcopenia, sarcopenia had largest SDn-values for harmonics 1, 2, 3, and 5 (ratio 1, 2, 3, 5 = 0.354, 0.209, 0.137, 0.074); whereas sarcopenia had largest coefficients of variations (CVn) values for harmonics 1, 2, 3 and 10 (ratio 1, 2, 3, 10 = 0.263, 0.310, 0.402, 0.791). Besides, the Δ odds ratio of ratio 3, 4,and 6 tertile values were significantly increased in sarcopenia and possible sarcopenia group compared with control group. Subjects with sarcopenia had significantly higher VEI in mean, SD, and CV of the PPG waveform (mean = 2.332, SD = 1.479, CV = 0.634, p = 0.007) among the groups and the results of binary logistic regression analysis in the tertiles met statistical significance between the sarcopenia and non-sarcopenia groups whether adjusted or unadjusted (adjusted odds ratio 6.956, p = 0.030, unadjusted odds ratio 3.937, p = 0.039).ConclusionsThe elasticity of vessels among sarcopenia groups in lower-frequency components of harmonic ratio in which we defined as VEI showed a significantly highest VEI mean, SD, and CV in sarcopenia indicates the poorer elasticity of the arteries. The present findings showed finger PPG waveform measurements may be useful for early detection of vascular diseases with patients with sarcopenia in a non-invasive and easy-to-perform technique which may expand the clinical applicability in the future.

Leveraging Continuous Vital Sign Measurements for Real-Time Assessment of Autonomic Nervous System Dysfunction After Brain Injury: A Narrative Review of Current and Future Applications.

Subtle and profound changes in autonomic nervous system (ANS) function affecting sympathetic and parasympathetic homeostasis occur as a result of critical illness. Changes in ANS function are particularly salient in neurocritical illness, when direct structural and functional perturbations to autonomic network pathways occur and may herald impending clinical deterioration or intervenable evolving mechanisms of secondary injury. Sympathetic and parasympathetic balance can be measured quantitatively at the bedside using multiple methods, most readily by extracting data from electrocardiographic or photoplethysmography waveforms. Work from our group and others has demonstrated that data-analytic techniques can identify quantitative physiologic changes that precede clinical detection of meaningful events, and therefore may provide an important window for time-sensitive therapies. Here, we review data-analytic approaches to measuring ANS dysfunction from routine bedside physiologic data streams and integrating this data into multimodal machine learning-based model development to better understand phenotypical expression of pathophysiologic mechanisms and perhaps even serve as early detection signals. Attention will be given to examples from our work in acute traumatic brain injury on detection and monitoring of paroxysmal sympathetic hyperactivity and prediction of neurologic deterioration, and in large hemispheric infarction on prediction of malignant cerebral edema. We also discuss future clinical applications and data-analytic challenges and future directions.

Multiclass Arrhythmia Detection and Classification From Photoplethysmography Signals Using a Deep Convolutional Neural Network.

BackgroundStudies have reported the use of photoplethysmography signals to detect atrial fibrillation; however, the use of photoplethysmography signals in classifying multiclass arrhythmias has rarely been reported. Our study investigated the feasibility of using photoplethysmography signals and a deep convolutional neural network to classify multiclass arrhythmia types.Methods and ResultsECG and photoplethysmography signals were collected simultaneously from a group of patients who underwent radiofrequency ablation for arrhythmias. A deep convolutional neural network was developed to classify multiple rhythms based on 10‐second photoplethysmography waveforms. Classification performance was evaluated by calculating the area under the microaverage receiver operating characteristic curve, overall accuracy, sensitivity, specificity, and positive and negative predictive values against annotations on the rhythm of arrhythmias provided by 2 cardiologists consulting the ECG results. A total of 228 patients were included; 118 217 pairs of 10‐second photoplethysmography and ECG waveforms were used. When validated against an independent test data set (23 384 photoplethysmography waveforms from 45 patients), the DCNN achieved an overall accuracy of 85.0% for 6 rhythm types (sinus rhythm, premature ventricular contraction, premature atrial contraction, ventricular tachycardia, supraventricular tachycardia, and atrial fibrillation); the microaverage area under the microaverage receiver operating characteristic curve was 0.978; the average sensitivity, specificity, and positive and negative predictive values were 75.8%, 96.9%, 75.2%, and 97.0%, respectively.ConclusionsThis study demonstrated the feasibility of classifying multiclass arrhythmias from photoplethysmography signals using deep learning techniques. The approach is attractive for population‐based screening and may hold promise for the long‐term surveillance and management of arrhythmia.RegistrationURL: www.chictr.org.cn. Identifier: ChiCTR2000031170.

Achieving High Accuracy in Predicting Blood Flow Volume at the Arteriovenous Fistulas of Hemodialysis Patients by Intelligent Quality Assessment on PPGs

A machine learning (ML) algorithm is successfully developed to assess the signal quality of measured photoplethysmography (PPG) waveforms for effective real-time prediction of blood flow volume (BFV). This algorithm is essential to achieve high prediction accuracy of BFV for hemodialysis patients to monitor the quality of their arteriovenous fistulas (AVFs) at home by themselves using a new hand-held device. The algorithm is built on an ML classifier of 1-dimensional convolutional neural network (1D-CNN), calibrated by two groups of PPG waveforms that are pre-identified by experienced experts to 300 qualified and 202 un-qualified PPG waveforms in 6-second windows, as the ones to render accurate BFV predictions and the others to inaccurates, respectively. The qualified ones satisfy at least the deterministic criteria such as adequate signal-to-noise-ratio (SNR), stable/small fluctuations of AC/DC components, and also the presence of secondary bio-features, to ensure minimized influence from mis-positioning, hand movement, pressurization and varied ambient lighting. With the classifier algorithm in hand, together with another fully-connected neural network for estimating BFV, a combined real-time computation algorithm is built, being able to achieve much better accuracy for real-time measurement ubiquitously. In results, using the newly-developed quality-assessment algorithm, the error of predicted BFVs is improved from ±175.577 ml/min to ±122.8259 ml/min significantly.

Abnormal photoplethysmography waveforms are associated with chemotherapy induced neuropathy.

Background: Chemotherapy-induced peripheral neuropathy (CIPN) is a dose-dependent adverse event of many chemotherapy agents that affects autonomic, motor, and sensory nerve fibers. The purpose of this study is to describe abnormal photoplethysmography waveforms (PPGs) in the setting of CIPN in cancer patients screened for peripheral arterial disease (PAD), which to our knowledge has not been previously described. Patients and methods: 147 patients who underwent vascular physiologic testing in evaluation for PAD with an ankle brachial index (ABI) or toe brachial index (TBI), segmental pressures, pulse volume recordings, and toe PPGs, in a tertiary cancer center's vascular lab between January 1, 2019 and January 31, 2021 were included in the study. Results: Odds ratio analysis demonstrates 3 times increased odds of abnormal PPGs in patients with PAD (OR 3.2256 95% CI 1.523-6.832, p=0.002), 7 times increased odds of abnormal PPGs in patients with CIPN (OR 7.802 95% CI 3.606-16.880, p<0.001), 9 times increased odds of abnormal PPGs in patients with both CIPN and PAD (9.895 95% CI 2.643-37.043, p=0.001), and 7 times increased odds of abnormal PPGs in patients with chemotherapy agent known to cause CIPN (7.821 95% CI 3.619-16.902, p<0.001). Logistic regression demonstrated that PAD (coefficient 1.171 std. error 0.383 wald 9.354 p=0.002), CIPN (coefficient 2.054 std. error 0.394 wald 27.227 p<0.001), and chemo agent known to cause CIPN (coefficient 2.057 std. error 0.393 wald 27.370 p<0.001) were all predictors of abnormal PPGs. Conclusions: CIPN had greater odds for abnormal PPGs than PAD. Additional larger studies are needed to assess if PPG analysis could be utilized to assess for early diagnosis of CIPN.

Applied machine learning for blood pressure estimation using a small, real-world electrocardiogram and photoplethysmogram dataset.

Applying machine learning techniques to electrocardiography and photoplethysmography signals and their multivariate-derived waveforms is an ongoing effort to estimate non-occlusive blood pressure. Unfortunately, real ambulatory electrocardiography and photoplethysmography waveforms are inevitably affected by motion and noise artifacts, so established machine learning architectures perform poorly when trained on data of the Multiparameter Intelligent Monitoring in Intensive Care II type, a publicly available ICU database. Our study addresses this problem by applying four well-established machine learning methods, i.e., random forest regression, support vector regression, Adaboost regression and artificial neural networks, to a small, self-sampled electrocardiography-photoplethysmography dataset (n = 54) to improve the robustness of machine learning to real-world BP estimates. We evaluated the performance using a selection of optimal feature morphologies of waveforms by using pulse arrival time, morphological and frequency photoplethysmography parameters and heart rate variability as characterization data. On the basis of the root mean square error and mean absolute error, our study showed that support vector regression gave the best performance for blood pressure estimation from noisy data, achieving an mean absolute error of 6.97 mmHg, which meets the level C criteria set by the British Hypertension Society. We demonstrate that ambulatory electrocardiography- photoplethysmography signals acquired by mobile discrete devices can be used to estimate blood pressure.

External factors that affect the photoplethysmography waveforms

Photoplethysmography (PPG) is a simple and inexpensive technology used in many smart devices to monitor cardiovascular health. The PPG sensors use LED lights to penetrate into the bloodstream to detect the different blood volume changes in the tissue through skin contact by sensing the amount of light that hits the sensor. Typically, the data are displayed on a graph and it forms the pulse waveform. The information from the produced pulse waveform can be useful in calculating measurements that help monitor cardiovascular health, such as blood pressure. With many more people beginning to monitor their health status on their smart devices, it is extremely important that the PPG signal is accurate. Designing a simple experiment with standard laboratory equipment and commercial sensors, we wanted to find how external factors influence the results. In this study, it was found that external factors, touch force and temperature, can have a large impact on the resulting waveform, so the effects of those factors need to be considered in order for the information to become more reliable.

PPGSynth: An Innovative Toolbox for Synthesizing Regular and Irregular Photoplethysmography Waveforms.

Photoplethysmography (PPG) is increasingly used in digital health, exceptionally in smartwatches. The PPG signal contains valuable information about heart activity, and there is lots of research interest in its means and analysis for cardiovascular diseases. Unfortunately, to our knowledge, there is no arrhythmic PPG dataset publicly available—this paper attempt to provide a toolbox that can generate synthesized arrhythmic PPG signals. The model of a single PPG pulse in this toolbox utilizes two combined Gaussian functions. This toolbox supports synthesizing PPG waveform with regular heartbeats and three irregular heartbeats: compensation, interpolation, and reset. The user can generate a large amount of PPG data with a certain irregularity, with different sampling frequency, time length, and a range of noise types (Gaussian noise and multi-frequency noise) can be added to the synthesized PPG which can all be modified from the interface, and different types of arrhythmic PPGs (as calculated by the model) generated. The generation for large PPG datasets that simulate PPG collected from real humans could be used for testing the robustness of developed algorithms that are targeting arrhythmic PPG signals. Our PPG synthesis tool is publicly available.

Quantifying changes in size of arrhythmic photoplethysmography waveforms during a Valsalva maneuver for assessing cardiac filling pressure

Objective: We previously showed that the change in amplitude of a finger photoplethysmography waveform during the Valsalva maneuver reflects cardiac filling pressure. However, the automated determination of peaks and valleys to calculate amplitude is limited in significant arrhythmias such as atrial fibrillation and premature ventricular complexes, which are common in heart failure. The purpose of this study was to assess the change in size of the waveform by calculating the change in root mean square (RMS) of the signal, thereby utilizing the entire cardiac cycle, and to compare it to change in size of peak-to-valley amplitude. Approach: We compared the two approaches in signals obtained from participants of a prior study who were tested prior to a clinically indicated cardiac catheterization. Correlation between the two methods was assessed in cases without, and then with, significant arrhythmias including atrial fibrillation or premature ventricular complexes. Main results: Calculations from the two methods of peak-valley amplitude and RMS were highly correlated with each other in signals without (0.99, p < 0.0001, n = 252) and with significant arrhythmias (0.90, p < 0.0001, n = 34). Significance: RMS analysis of photoplethysmography signal size during the Valsalva maneuver is highly correlated with the method of analyzing changes in peak-valley amplitude, but does not rely on identifying peaks and valleys. The RMS method may be a more robust automated method of assessing cardiac filling pressure in patients with significant arrhythmias.

Heart-rate tuned comb filters for processing photoplethysmogram (PPG) signals in pulse oximetry

Calculation of peripheral capillary oxygen saturation {text{(SpO}}_{{text{2}}} {text{)}} levels in humans is often made with a pulse oximeter, using photoplethysmography (PPG) waveforms. However, measurements of PPG waveforms are susceptible to motion noise due to subject and sensor movements. In this study, we compare two {text{SpO}}_{{text{2}}}-level calculation techniques, and measure the effect of pre-filtering by a heart-rate tuned comb peak filter on their performance. These techniques are: (1) “Red over Infrared,” calculating the ratios of AC and DC components of the red and infrared PPG signals,frac{(AC/DC)_{red}}{(AC/DC)_{infrared}}, followed by the use of a calibration curve to determine the {text{SpO}}_{{text{2}}} level Webster (in: Design of pulse oximeters, CRC Press, Boca Raton, 1997); and (2) a motion-resistant algorithm which uses the Discrete Saturation Transform (DST) (Goldman in J Clin Monit Comput 16:475–83, 2000). The DST algorithm isolates individual “saturation components” in the optical pathway, which allows separation of components corresponding to the {text{SpO}}_{{text{2}}} level from components corresponding to noise and interference, including motion artifacts. The comparison we provide here (employing the two techniques with and without pre-filtering) addresses two aspects: (1) accuracy of the {text{SpO}}_{{text{2}}} calculations; and (2) computational complexity. We used both synthetic data and experimental data collected from human subjects. The human subjects were tested at rest and while exercising; while exercising, their measurements were subject to the impacts of motion. Our main conclusion is that if an uninterrupted high-quality heart rate measurement is available, then the “Red over Infrared” approach preceded by a heart-rate tuned comb filter provides the preferred trade-off between {text{SpO}}_{{text{2}}}-level accuracy and computational complexity. A modest improvement in {text{SpO}}_{{text{2}}} estimate accuracy at very low SNR environments may be achieved by switching to the pre-filtered DST-based algorithm (up to 6% improvement in {text{SpO}}_{{text{2}}} level accuracy at −10 dB over unfiltered DST algorithm and the filtered “Red over Infrared” approach). However, this improvement comes at a significant computational cost.

Review of mobile applications for the detection and management of atrial fibrillation.

BackgroundFree mobile applications (apps) that use photoplethysmography (PPG) waveforms may extend atrial fibrillation (AF) detection to underserved populations, but they have not been rigorously evaluated.ObjectiveThe purpose of this study was to systematically review and evaluate the quality, functionality, and adherence to self-management behaviors of existing mobile apps for AF.MethodsWe systematically searched 3 app stores for apps that were free, available in English, and intended for use by patients to detect and manage AF. A minimum of 2 reviewers evaluated (1) app quality, using the Mobile Application Rating Scale (MARS); (2) functionality using published criteria; and (3) features that support 4 self-management behaviors (including PPG waveform monitoring) identified using evidence-based guidelines. Interrater reliability between the reviewers was calculated.ResultsOf 12 included apps, 5 (42%) scored above average for quality (MARS score ≥3.0). App quality was highest for their ease of use, navigation, layout, and visual appeal (eg, functionality and aesthetics) and lowest for their behavioral change support and subjective impressions of quality. The most common app functionalities were capturing and graphically displaying user-entered data (n = 9 [75%]). Nearly all apps (n = 11 [92%]) supported PPG waveform monitoring, but only 2 (17%) supported all 4 self-management behaviors. Interrater reliability was high (0.75–0.83).ConclusionThe reviewed apps had wide variability in quality, functionality, and adherence to self-management behaviors. Given the accessibility of these apps to underserved populations and the tremendous potential they hold for improving AF detection and management, high priority should be given to improving app quality and functionality.

Estimation and Tracking of Blood Pressure Using Routinely Acquired Photoplethysmographic Signals and Deep Neural Networks.

Retrospective cohort study with split sampling for model training and testing. A single urban academic hospital. Three-hundred twenty-nine adult patients admitted to a critical care unit. None. One hundred thirty-six thousand four-hundred fifty-nine photoplethysmography waveforms of length 30 seconds were used for training (60%), validation (20%), and testing (20%) of the blood pressure estimation network. Each sample had an associated systolic, mean, and diastolic blood pressures extracted from concurrently recorded invasive arterial line waveforms. Blood pressure estimation using photoplethysmography waveforms is achieved using advanced machine learning methods (convolutional neural networks and a Siamese architectural configuration) calibrated for each patient on a single, first available photoplethysmography sample and associated blood pressure reading. The average estimation bias error was 0.52, 0.1, and -0.76 mm Hg for diastolic, mean, and systolic blood pressure, respectively, with associated mean absolute errors of 4.11, 5.51, and 7.98 mm Hg. If used to identify clinically important changes in blood pressure from the initial baseline, with a threshold of a 10 mm Hg increase or decrease in blood pressure, our algorithm shows an accuracy of 85%, 78%, and 74% for diastolic, mean, and systolic blood pressure, respectively. We also report the network's performance in detecting systolic and diastolic hypo- or hypertension with accuracies ranging from 86% to 97%. Using advanced machine learning tools, we show that blood pressure estimation can be achieved using a common noninvasively recorded signal, the photoplethysmography. Such tools can allow for better monitoring of patients that do not have invasively recorded blood pressure, both in the critical care setting and on inpatient wards.

Atrial fibrillation detection from raw photoplethysmography waveforms: A deep learning application.

BackgroundAtrial fibrillation (AF), a common cause of stroke, often is asymptomatic. Smartphones and smartwatches can detect AF using heart rate patterns inferred using photoplethysmography (PPG); however, enhanced accuracy is required to reduce false positives in screening populations.ObjectiveThe purpose of this study was to test the hypothesis that a deep learning algorithm given raw, smartwatch-derived PPG waveforms would discriminate AF from normal sinus rhythm better than algorithms using heart rate alone.MethodsPatients presenting for cardioversion of AF (n = 51) were given wrist-worn fitness trackers containing PPG sensors (Jawbone Health). Standard 12-lead electrocardiograms over-read by board-certified cardiac electrophysiologists were used as the reference standard. The accuracy of PPG signals to discriminate AF from sinus rhythm was evaluated by conventional measures of heart rate variability, a long short-term memory (LSTM) neural network given heart rate data only, and a deep convolutional-recurrent neural net (DNN) given the raw PPG data.ResultsFrom among 51 patients with persistent AF (age 63.6 ± 11.3 years; 78% male; 88% white), we randomly assigned 40 to train and 11 to test the algorithms. Whereas logistic regression analysis of heart rate variability yielded an area under the receiver operating characteristic curve (AUC) of 0.717 (sensitivity 0.741; specificity 0.584), the LSTM model given heart rate data exhibited AUC of 0.954 (sensitivity 0.810; specificity 0.921), and the DNN model given raw PPG data yielded the highest AUC of 0.983 (sensitivity 0.985; specificity 0.880).ConclusionA deep learning model given the raw PPG-based signal resulted in AF detection with high accuracy, performing better than conventional analyses relying on heart rate series data alone.