CorNET: Deep Learning Framework for PPG-Based Heart Rate Estimation and Biometric Identification in Ambulant Environment

Medical intelligence using PPG signals and hybrid learning at the edge to detect fatigue in physical activities

The educational environment plays a vital role in the development of students who participate in athletic pursuits both in terms of their physical health and their ability to detect fatigue. As a result of recent advancements in deep learning and biosensors benefitting from edge computing resources, we are now able to monitor the physiological fatigue of students participating in sports in real time. These devices can then be used to analyze the data using contemporary technology. In this paper, we present an innovative deep learning framework for forecasting fatigue in athletic students following physical exercise. It addresses the issue of lack of precision computational models and extensive data analysis in current approaches to monitoring students' physical activity. In our study, we classified fatigue and non-fatigue based on photoplethysmography (PPG) signals. Several deep learning models are compared in the study. Using limited training data, determining the optimal parameters for PPG presents a significant challenge. For datasets containing many data points, several models were trained using PPG signals: a deep residual network convolutional neural network (ResNetCNN) ResNetCNN, an Xception architecture, a bidirectional long short-term memory (BILSTM), and a combination of these models. Training and testing datasets were assigned using a fivefold cross validation approach. Based on the testing dataset, the model demonstrated a proper classification accuracy of 91.8%.

Deep Learning-Based Estimation of Arterial Stiffness from PPG Spectrograms: A Novel Approach for Non-Invasive Cardiovascular Diagnostics

Cardiovascular diseases (CVDs), a leading cause of global mortality, are intricately linked to arterial stiffness, a key factor in cardiovascular health. Non-invasive assessment of arterial stiffness, particularly through Carotid-to-femoral Pulse Wave Velocity (cf-PWV) - the gold standard in this field - is vital for early detection and management of CVDs. This study introduces a novel approach, utilizing photoplethysmogram (PPG) signal spectrograms as inputs for deep learning models to estimate cf-PWV, a significant advancement over traditional methods. Employing a modified ResNet-18 architecture, we analyze PPG signals from digital, radial, and brachial arteries of a simulated dataset of 4374 healthy adults. Our methodology's innovation lies in its direct use of finely tuned spectrogram images, bypassing the complex feature extraction processes. This approach achieved R2 (correlation coefficient) values of up to 0.9902 for the digital artery, 0.9898 for the radial artery, and 0.9825 for the brachial artery, coupled with significantly lower Mean Absolute Percentage Errors (MAPE) of approximately 1.61% for the digital, 1.87% for the radial, and 2.08% for the brachial artery. These findings highlight the efficacy of PPG spectrograms, especially from the digital artery, in providing an accurate, user-friendly, and non-invasive method for cf-PWV estimation, thereby enhancing the capabilities of non-invasive cardiovascular diagnostics.

A Fast FPGA Hardware Accelerator for Remote Heart Rate Detection Based on RGB Vision

A fast hardware accelerator is created by this work via field programmable gate array (FPGA) to estimate heart rate (HR) through the video recorded by a RGB camera based on the technology of remote photoplethysmography (rPPG). The method of rPPG acquires physiological signals of a human body by analyzing the subtle color changes on the surface of the human skin. The hardware implementation of rPPG to estimate HR is proposed herein to aim for a much faster calculation speed than software for a number of applications, like heart failure pre-warning of an in-action athlete and drowsiness detection of a driver. In this accelerator, ICA (Independent Component Analysis) is used to recover the blood volume pulse from the raw signals of remote PPG, and then obtain the heart rate value. The architecture of the hardware circuit is described in Verilog HDL and verified by Quartus II, and also implemented in an Altera DE10-Standard FPGA board, which consists of image capture, heart rate algorithm and image display. A TRDB-D5M camera is utilized for image capture. Two experiments were conducted with image collecting duration of 16 seconds and 8 seconds respectively, and the commercial device Omron HEM-6111 was used as the golden value. The proposed system achieves an accuracy in (ME ± 1.96SD) of -0.76 ± 5.09 and -0.70 ± 8.71 bpm in the short periods of 16-second and 8-second versions, respectively, which outperforms all the reported prior works in combined computation time and accuracy.

Uncertainties in the Analysis of Heart Rate Variability: A Systematic Review

Heart rate variability (HRV) is an important metric with a variety of applications in clinical situations such as cardiovascular diseases, diabetes mellitus, and mental health. HRV data can be potentially obtained from electrocardiography and photoplethysmography signals, then computational techniques such as signal filtering and data segmentation are used to process the sampled data for calculating HRV measures. However, uncertainties arising from data acquisition, computational models, and physiological factors can lead to degraded signal quality and affect HRV analysis. Therefore, it is crucial to address these uncertainties and develop advanced models for HRV analysis. Although several reviews of HRV analysis exist, they primarily focus on clinical applications, trends in HRV methods, or specific aspects of uncertainties such as measurement noise. This paper provides a comprehensive review of uncertainties in HRV analysis, quantifies their impacts, and outlines potential solutions. To the best of our knowledge, this is the first study that presents a holistic review of uncertainties in HRV methods and quantifies their impacts on HRV measures from an engineer's perspective. This review is essential for developing robust and reliable models, and could serve as a valuable future reference in the field, particularly for dealing with uncertainties in HRV analysis.

A Novel PPG-Based Biometric Authentication System Using a Hybrid CVT-ConvMixer Architecture with Dense and Self-Attention Layers

Biometric authentication is a widely used method for verifying individuals' identities using photoplethysmography (PPG) cardiac signals. The PPG signal is a non-invasive optical technique that measures the heart rate, which can vary from person to person. However, these signals can also be changed due to factors like stress, physical activity, illness, or medication. Ensuring the system can accurately identify and authenticate the user despite these variations is a significant challenge. To address these issues, the PPG signals were preprocessed and transformed into a 2-D image that visually represents the time-varying frequency content of multiple PPG signals from the same human using the scalogram technique. Afterward, the features fusion approach is developed by combining features from the hybrid convolution vision transformer (CVT) and convolutional mixer (ConvMixer), known as the CVT-ConvMixer classifier, and employing attention mechanisms for the classification of human identity. This hybrid model has the potential to provide more accurate and reliable authentication results in real-world scenarios. The sensitivity (SE), specificity (SP), F1-score, and area under the receiver operating curve (AUC) metrics are utilized to assess the model's performance in accurately distinguishing genuine individuals. The results of extensive experiments on the three PPG datasets were calculated, and the proposed method achieved ACCs of 95%, SEs of 97%, SPs of 95%, and an AUC of 0.96, which indicate the effectiveness of the CVT-ConvMixer system. These results suggest that the proposed method performs well in accurately classifying or identifying patterns within the PPG signals to perform continuous human authentication.

Simulation of ambulatory electrodermal activity and the handling of low-quality segments

BACKGROUND AND OBJECTIVES

Monitoring electrodermal activity (EDA) in daily life requires effective handling of low-quality segments, which are common in ambulatory EDA data. Although several low-quality handling methods have been implemented, systematic comparison of these methods, which requires a large annotated dataset, is lacking.

METHODS

Therefore, we proposed the simulation of realistic ambulatory EDA data starting from high-quality EDA signals, which were subsequently contaminated with varying concentrations of artifacts. Subsequently, three approaches for handling low-quality data were evaluated regarding the preservation of several EDA-derived features: removing all artifacts, interpolating over removed artifacts, and retaining all artifacts. Specifically, multiple EDA features were assessed, derived from response detection (evaluated using F1, precision, recall) as well as EDA, phasic, and tonic features (assessed using absolute error), by comparing the simulated EDA data with and without the inserted artifacts, using the latter as ground truth.

RESULTS

For response detection, retaining artifacts resulted in the highest F1-scores, while interpolating over removed artifacts achieved the highest F1-scores for the phasic signal. The approaches did significantly differ in the mean error for the phasic but not for the tonic component and raw EDA.

CONCLUSION

This work generated ambulatory EDA datasets of 200 h, containing 0.125 to 3 artifacts per minute, and showed that interpolation over removed artifacts was an effective approach to reconstruct phasic-derived features up to 2 artifacts per minute. The proposed simulation and evaluation methodology, which are easily customizable, offer opportunities for future research to develop and systematically compare signal quality indicators, decomposition methods, and response detectors for processing ambulatory EDA.

Fourier Synchrosqueezing Transform-ICA-EMD Framework Based EOG-Biometric Sustainable and Continuous Authentication via Voluntary Eye Blinking Activities

In recent years, limited works on EOG (electrooculography)-based biometric authentication systems have been carried out with eye movements or eye blinking activities in the current literature. EOGs have permanent and unique traits that can separate one individual from another. In this work, we have investigated FSST (Fourier Synchrosqueezing Transform)-ICA (Independent Component Analysis)-EMD (Empirical Mode Decomposition) robust framework-based EOG-biometric authentication (one-versus-others verification) performances using ensembled RNN (Recurrent Neural Network) deep models voluntary eye blinkings movements. FSST is implemented to provide accurate and dense temporal-spatial properties of EOGs on the state-of-the-art time-frequency matrix. ICA is a powerful statistical tool to decompose multiple recording electrodes. Finally, EMD is deployed to isolate EOG signals from the EEGs collected from the scalp. As our best knowledge, this is the first research attempt to explore the success of the FSST-ICA-EMD framework on EOG-biometric authentication generated via voluntary eye blinking activities in the limited EOG-related biometric literature. According to the promising results, improved and high recognition accuracies (ACC/Accuracy: ≥99.99% and AUC/Area under the Curve: 0.99) have been achieved in addition to the high TAR (true acceptance rate) scores (≥98%) and low FAR (false acceptance rate) scores (≤3.33%) in seven individuals. On the other hand, authentication and monitoring for online users/students are becoming essential and important tasks due to the increase of the digital world (e-learning, e-banking, or e-government systems) and the COVID-19 pandemic. Especially in order to ensure reliable access, a highly scalable and affordable approach for authenticating the examinee without cheating or monitoring high-data-size video streaming is required in e-learning platforms and online education strategies. Hence, this work may present an approach that offers a sustainable, continuous, and reliable EOG-biometric authentication of digital applications, including e-learning platforms for users/students.

Heart rate estimation from ballistocardiogram signals processing via low-cost telemedicine architectures: a comparative performance evaluation

Medical devices (MDs) have been designed for monitoring the parameters of patients in many sectors. Nonetheless, despite being high-performing and reliable, they often turn out to be expensive and intrusive. In addition, MDs are almost exclusively used in controlled, hospital-based environments. Paving a path of technological innovation in the clinical field, a very active line of research is currently dealing with the possibility to rely on non-medical-graded low-cost devices, to develop unattended telemedicine (TM) solutions aimed at non-invasively gathering data, signals, and images. In this article, a TM solution is proposed for monitoring the heart rate (HR) of patients during sleep. A remote patient monitoring system (RPMS) featuring a smart belt equipped with pressure sensors for ballistocardiogram (BCG) signals sampling was deployed. A field trial was then conducted over a 2-month period on 24 volunteers, who also agreed to wear a finger pulse oximeter capable of producing a photoplethysmography (PPG) signal as the gold standard, to examine the feasibility of the solution via the estimation of HR values from the collected BCG signals. For this purpose, two of the highest-performing approaches for HR estimation from BCG signals, one algorithmic and the other based on a convolutional neural network (CNN), were retrieved from the literature and updated for a TM-related use case. Finally, HR estimation performances were assessed in terms of patient-wise mean absolute error (MAE). Results retrieved from the literature (controlled environment) outperformed those achieved in the experimentation (TM environment) by 29% (MAE = 4.24 vs. 5.46, algorithmic approach) and 52% (MAE = 2.32 vs. 3.54, CNN-based approach), respectively. Nonetheless, a low packet loss ratio, restrained elaboration time of the collected biomedical big data, low-cost deployment, and positive feedback from the users, demonstrate the robustness, reliability, and applicability of the proposed TM solution. In light of this, further steps will be planned to fulfill new targets, such as evaluation of respiratory rate (RR), and pattern assessment of the movement of the participants overnight.

An ML-Based Approach to Reconstruct Heart Rate from PPG in Presence of Motion Artifacts

The heart rate (HR) is a widely used clinical variable that provides important information on a physical user's state. One of the most commonly used methods for ambulatory HR monitoring is photoplethysmography (PPG). The PPG signal retrieved from wearable devices positioned on the user's wrist can be corrupted when the user is performing tasks involving the motion of the arms, wrist, and fingers. In these cases, the obtained HR is altered as well. This problem increases when trying to monitor people with autism spectrum disorder (ASD), who are very reluctant to use foreign bodies, notably hindering the adequate attachment of the device to the user. This work presents a machine learning approach to reconstruct the user's HR signal using an own monitoring wristband especially developed for people with ASD. An experiment is carried out, with users performing different daily life activities in order to build a dataset with the measured signals from the monitoring wristband. From these data, an algorithm is applied to obtain a reliable HR value when these people are performing skill improvement activities where intensive wrist movement may corrupt the PPG.

Precision Heart Rate Estimation Using a PPG Sensor Patch Equipped with New Algorithms of Pre-Quality Checking and Hankel Decomposition

A new method for accurately estimating heart rates based on a single photoplethysmography (PPG) signal and accelerations is proposed in this study, considering motion artifacts due to subjects' hand motions and walking. The method comprises two sub-algorithms: pre-quality checking and motion artifact removal (MAR) via Hankel decomposition. PPGs and accelerations were collected using a wearable device equipped with a PPG sensor patch and a 3-axis accelerometer. The motion artifacts caused by hand movements and walking were effectively mitigated by the two aforementioned sub-algorithms. The first sub-algorithm utilized a new quality-assessment criterion to identify highly noise-contaminated PPG signals and exclude them from subsequent processing. The second sub-algorithm employed the Hankel matrix and singular value decomposition (SVD) to effectively identify, decompose, and remove motion artifacts. Experimental data collected during hand-moving and walking were considered for evaluation. The performance of the proposed algorithms was assessed using the datasets from the IEEE Signal Processing Cup 2015. The obtained results demonstrated an average error of merely 0.7345 ± 8.1129 beats per minute (bpm) and a mean absolute error of 1.86 bpm for walking, making it the second most accurate method to date that employs a single PPG and a 3-axis accelerometer. The proposed method also achieved the best accuracy of 3.78 bpm in mean absolute errors among all previously reported studies for hand-moving scenarios.

Rapid Vital Sign Extraction for Real-Time Opto-Physiological Monitoring at Varying Physical Activity Intensity Levels

Robustness of physiological parameters obtained from photoplethysmographic (PPG) signals is highly dependent on a signal quality that is often affected by the motion artefacts (MAs) generated during physical activity. This study aims to suppress MAs and obtain reliable physiological readings using the part of the pulsatile signal, captured by a multi-wavelength illumination optoelectronic patch sensor (mOEPS), that minimizes the residual between the measured signal and the motion estimates obtained from an accelerometer. The minimum residual (MR) method requires the simultaneous collection of (1) multiple wavelength data from the mOEPS, and (2) motion reference signals from a triaxial accelerometer attached to the mOEPS. The MR method suppresses those frequencies associated with motion in a manner that is easily embedded on a microprocessor. The performance of the method in reducing both in-band and out-of-band frequencies of MAs is evaluated through two protocols with 34 subjects engaged in the study. The MA-suppressed PPG signal, obtained through MR, enables the calculation of the heart rate (HR) with an average absolute error of 1.47 beats/min for the IEEE-SPC datasets, and the calculation of HR and respiration rate (RR) to 1.44 beats/min and 2.85 breaths/min respectively for our in-house datasets. Oxygen saturation (SpO 2) levels calculated from the minimum residual wave forms were consistently [Formula: see text]. The comparison with the reference HR and RR show errors with an absolute accuracy of [Formula: see text] and the Pearson correlation ( R) for HR and RR are 0.9976 and 0.9118, respectively. These outcomes demonstrate that MR is capable of effective suppression of MAs for a range of physical activity intensities and to achieve real-time signal processing for wearable health monitoring.

Multi-Task Learning Radar Transformer (MLRT): A Personal Identification and Fall Detection Network Based on IR-UWB Radar

Radar-based personal identification and fall detection have received considerable attention in smart healthcare scenarios. Deep learning algorithms have been introduced to improve the performance of non-contact radar sensing applications. However, the original Transformer network is not suitable for multi-task radar-based applications to effectively extract temporal features from time-series radar signals. This article proposes the Multi-task Learning Radar Transformer (MLRT): a personal Identification and fall detection network based on IR-UWB radar. The proposed MLRT utilizes the attention mechanism of Transformer as its core to automatically extract features for personal identification and fall detection from radar time-series signals. Multi-task learning is applied to exploit the correlation between the personal identification task and the fall detection task, enhancing the performance of discrimination for both tasks. In order to suppress the impact of noise and interference, a signal processing approach is employed including DC removal and bandpass filtering, followed by clutter suppression using a RA method and Kalman filter-based trajectory estimation. An indoor radar signal dataset is generated with 11 persons under one IR-UWB radar, and the performance of MLRT is evaluated using this dataset. The measurement results show that the accuracy of MLRT improves by 8.5% and 3.6% for personal identification and fall detection, respectively, compared to state-of-the-art algorithms. The indoor radar signal dataset and the proposed MLRT source code are publicly available.

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.

Technology Roadmap for Flexible Sensors

Humans rely increasingly on sensors to address grand challenges and to improve quality of life in the era of digitalization and big data. For ubiquitous sensing, flexible sensors are developed to overcome the limitations of conventional rigid counterparts. Despite rapid advancement in bench-side research over the last decade, the market adoption of flexible sensors remains limited. To ease and to expedite their deployment, here, we identify bottlenecks hindering the maturation of flexible sensors and propose promising solutions. We first analyze challenges in achieving satisfactory sensing performance for real-world applications and then summarize issues in compatible sensor-biology interfaces, followed by brief discussions on powering and connecting sensor networks. Issues en route to commercialization and for sustainable growth of the sector are also analyzed, highlighting environmental concerns and emphasizing nontechnical issues such as business, regulatory, and ethical considerations. Additionally, we look at future intelligent flexible sensors. In proposing a comprehensive roadmap, we hope to steer research efforts towards common goals and to guide coordinated development strategies from disparate communities. Through such collaborative efforts, scientific breakthroughs can be made sooner and capitalized for the betterment of humanity.

Analysis of photoplethysmogram signal to estimate heart rate during physical activity using fractional fourier transform - A sampling frequency independent and reference signal-less method

BACKGROUND AND OBJECTIVE

Acquiring accurate and reliable health information using a PPG signal in wearable devices requires suppressing motion artifacts. This paper presents a method based on the Fractional Fourier transform (FrFT) to effectively suppress the motion artifacts in a Photoplethysmogram (PPG) signal for an accurate estimation of heart rate (HR).

METHODS

By analyzing various PPG signals recorded under various physiological conditions and sampling frequencies, the proposed work determines an optimal value of the fractional order of the proposed FrFT. The proposed FrFT-based algorithm separates the motion artifacts component from the acquired PPG signal. Finally, the HR estimation accuracy during the strong motion artifact-affected windows is improved using a post-processing technique. The efficacy of the proposed method is evaluated by computing the root mean square error (RMSE).

RESULTS

The performance of the proposed algorithm is compared with methods in recent studies using test and training datasets from the IEEE Signal Processing Cup (SPC). The proposed method provides the mean absolute error of 1.88 beats per minute (BPM) on all twenty-three recordings.

CONCLUSIONS

The proposed method uses the Fourier method in the fractional domain. A noisy signal is rotated into an intermediate plane between the time and frequency domains to separate the signal from the noise. The algorithm incorporates FrFT analysis to suppress motion artifacts from PPG signals to estimate HR accurately. Further, a post-processing step is used to track the HR for accurate and reliable HR estimation. The proposed FrFT-based algorithm doesn't require additional reference accelerometers or hardware to estimate HR in real-time. The noise and signal separation is optimum for a fractional order (a) value in the vicinity of 0.6. The optimized value of fractional order is constant irrespective of the physical activity and sampling frequency.

KPCA-WRF-prediction of heart rate using deep feature fusion and machine learning classification with tuned weighted hyper-parameter

The rapid advancement of technology such as stream processing technologies, deep-learning approaches, and artificial intelligence plays a prominent and vital role, to detect heart rate using a prediction model. However, the existing methods could not handle high -dimensional datasets, and deep feature learning to improvise the performance. Therefore, this work proposed a real-time heart rate prediction model, using K-nearest neighbour (KNN) adhered to the principle component analysis algorithm (PCA) and weighted random forest algorithm for feature fusion (KPCA-WRF) approach and deep CNN feature learning framework. The feature selection, from the fused features, was optimized by ant colony optimization (ACO) and particle swarm optimization (PSO) algorithm to enhance the selected fused features from deep CNN. The optimized features were reduced to low dimensions using the PCA algorithm. The significant straight heart rate features are plotted by capturing out nearest similar data point values using the algorithm. The fused features were then classified for aiding the training process. The weighted values are assigned to those tuned hyper parameters (feature matrix forms). The optimal path and continuity of the weighted feature representations are moved using the random forest algorithm, in K-fold validation iterations.

Heart Rate Estimation from Incomplete Electrocardiography Signals

As one of the most remarkable indicators of physiological health, heart rate (HR) has become an unfailing investigation for researchers. Unlike many existing methods, this article proposes an approach to implement short-time HR estimation from electrocardiography in time series missing patterns. Benefiting from the rapid development of deep learning, we adopted a bidirectional long short-term memory model (Bi-LSTM) and temporal convolution network (TCN) to recover complete heartbeat signals from those with durations are less than one cardiac cycle, and the estimated HR from recovered segment combining the input and the predicted output. We also compared the performance of Bi-LSTM and TCN in PhysioNet dataset. Validating the method over a resting heart rate range of 60−120 bpm in the database without significant arrhythmias and a corresponding range of 30−150 bpm in the database with arrhythmias, we found that networks provide an estimated approach for incomplete signals in a fixed format. These results are consistent with real heartbeats in the normal heartbeat dataset (γ > 0.7, RMSE < 10) and in the arrhythmia database (γ > 0.6, RMSE < 30), verifying that HR could be estimated by models in advance. We also discussed the short-time limits for the predictive model. It could be used for physiological purposes such as mobile sensing in time-constrained scenarios, and providing useful insights for better time series analyses in missing data patterns.

Intelligent Safety Ergonomics: A Cleaner Research Direction for Ergonomics in the Era of Big Data

Safety ergonomics is an important branch of safety science and environmental engineering. As humans enter the era of big data, the development of information technology has brought new opportunities and challenges to the innovation, transformation, and upgrading of safety ergonomics, as the traditional safety ergonomics theory has gradually failed to adapt to the need for safe and clean production. Intelligent safety ergonomics (ISE) is regarded as a new direction for the development of safety ergonomics in the era of big data. Unfortunately, since ISE is an emerging concept, there is no research to clarify its basic problems, which leads to a lack of theoretical guidance for the research and practice of ISE. In order to solve the shortcomings of traditional safety ergonomics theories and methods, first of all, this paper answers the basic questions of ISE, including the basic concepts, characteristics, attributes, contents, and research objects. Then, practical application functions of ISE are systematically clarified. Finally, following the life cycle of the design, implementation, operation, and maintenance of the system, it ends with a discussion of the challenges and application prospects of ISE. The conclusion shows that ISE is a cleaner research direction for ergonomics in the era of big data, that it can deepen the understanding of humans, machines, and environment systems, and it can provide a new method for further research on safety and cleaner production. Overall, this paper not only helps safety researchers and practitioners to correctly understand the concept of intelligent safety ergonomics, but it will certainly inject energy and vitality into the development of safety ergonomics and cleaner production.

DeepPulse: An Uncertainty-aware Deep Neural Network for Heart Rate Estimations from Wrist-worn Photoplethysmography

Wearable Photoplethysmography (PPG) has gained prominence as a low cost, unobtrusive and continuous method for physiological monitoring. The quality of the collected PPG signals is affected by several sources of interference, predominantly due to physical motion. Many methods for estimating heart rate (HR) from PPG signals have been proposed with Deep Neural Networks (DNNs) gaining popularity in recent years. However, the "black-box" and complex nature of DNNs has caused a lack of trust in the predicted values. This paper contributes DeepPulse, an uncertainty-aware DNN method for estimating HR from PPG and accelerometer signals, with aims of increasing trust of the predicted HR values. To the best of the authors' knowledge no PPG HR estimation method has considered aleatoric and epistemic uncertainty metrics. The results show DeepPulse is the most accurate method for DNNs with smaller network sizes. Finally, recommendations are given to reduce epistemic uncertainty, validate uncertainty estimates, improve the accuracy of DeepPulse as well as reduce the model size for resource-constrained edge devices.

A deep learning approach to estimate pulse rate by remote photoplethysmography

OBJECTIVE

This study proposes an U-net shaped Deep Neural Network (DNN) model to extract remote photoplethysmography (rPPG) signals from skin color signals to estimate Pulse Rate (PR).

APPROACH

Three input window sizes are used into the DNN: 256 samples (5.12 s), 512 samples (10.24 s), and 1024 (20.48 s). A data argumentation algorithm based on interpolation is also used here to artificially increase the number of training samples.

MAIN RESULTS

The proposed model outperformed a prior-knowledge rPPG method by using input signals with window of 256 and 512 samples. Also, it was found that the data augmentation procedure only increased the performance for window of 1024 samples. The trained model achieved a Mean Absolute Error (MAE) of 3.97 Beats per Minute (BPM) and Root Mean Squared Error (RMSE) of 6.47 BPM, for the 256 samples window, and MAE of 3.00 BPM and RMSE of 5.45 BPM for the window of 512 samples. On the other hand, the prior-knowledge rPPG method got a MAE of 8.04 BPM and RMSE of 16.63 BPM for the window of 256 samples, and MAE of 3.49 BPM and RMSE of 7.92 BPM for the window of 512. For the longest window (1024 samples), the concordance of the predicted PRs from the DNNs and the true PRs was higher when applying the data augmentation procedure.

SIGNIFICANCE

These results demonstrate a big potential of this technique for PR estimation, showing that the DNN proposed here may generate reliable rPPG signals even with short window lengths (5.12 s and 10.24 s), suggesting that it needs less data for a faster rPPG measurement and PR estimation.

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.

PPG biometric recognition with singular value decomposition and local mean decomposition

Although some methods of feature extraction for photoplethysmography (PPG) biometric recognition have been extensively studied, effectiveness of local features, time cost of feature extraction, and robust identification for small-scale data remain challenging. To address these issues, we proposed a feature-extraction method of PPG biometrics combining singular value decomposition with local mean decomposition and time-domain parameters. First, we used the singular-value-decomposition method to de-noise the original PPG data. Second, we extracted the local-mean-decomposition-based and time-domain features, which are fused into a concatenated feature. Finally, we combined the concatenated feature with four classifiers for classification and decision-making. Extensive experiments on the three datasets have shown that the waveform of the PPG signal de-noised by singular value decomposition was smoother and more regular, the concatenated feature had strong inter-subject distinguishability and intra-subject similarity, and the concatenated feature combined with a random-forest classifier was the best and could achieve 99.40%, 99.88%, and 99.56% recognition rates on the respective datasets. The method is competitive with several state-of-the-art methods.

Continuous Blood Pressure Estimation Based on Multi-Scale Feature Extraction by the Neural Network With Multi-Task Learning

In this article, a novel method for continuous blood pressure (BP) estimation based on multi-scale feature extraction by the neural network with multi-task learning (MST-net) has been proposed and evaluated. First, we preprocess the target (Electrocardiograph; Photoplethysmography) and label signals (arterial blood pressure), especially using peak-to-peak time limits of signals to eliminate the interference of the false peak. Then, we design a MST-net to extract multi-scale features related to BP, fully excavate and learn the relationship between multi-scale features and BP, and then estimate three BP values simultaneously. Finally, the performance of the developed neural network is verified by using a public multi-parameter intelligent monitoring waveform database. The results show that the mean absolute error ± standard deviation for systolic blood pressure (SBP), diastolic blood pressure (DBP), and mean arterial pressure (MAP) with the proposed method against reference are 4.04 ± 5.81, 2.29 ± 3.55, and 2.46 ± 3.58 mmHg, respectively; the correlation coefficients of SBP, DBP, and MAP are 0.96, 0.92, and 0.94, respectively, which meet the Association for the Advancement of Medical Instrumentation standard and reach A level of the British Hypertension Society standard. This study provides insights into the improvement of accuracy and efficiency of a continuous BP estimation method with a simple structure and without calibration. The proposed algorithm for BP estimation could potentially enable continuous BP monitoring by mobile health devices.

The Current State of Optical Sensors in Medical Wearables

Optical sensors play an increasingly important role in the development of medical diagnostic devices. They can be very widely used to measure the physiology of the human body. Optical methods include PPG, radiation, biochemical, and optical fiber sensors. Optical sensors offer excellent metrological properties, immunity to electromagnetic interference, electrical safety, simple miniaturization, the ability to capture volumes of nanometers, and non-invasive examination. In addition, they are cheap and resistant to water and corrosion. The use of optical sensors can bring better methods of continuous diagnostics in the comfort of the home and the development of telemedicine in the 21st century. This article offers a large overview of optical wearable methods and their modern use with an insight into the future years of technology in this field.

Photoplethysmogram Analysis and Applications: An Integrative Review

Beyond its use in a clinical environment, photoplethysmogram (PPG) is increasingly used for measuring the physiological state of an individual in daily life. This review aims to examine existing research on photoplethysmogram concerning its generation mechanisms, measurement principles, clinical applications, noise definition, pre-processing techniques, feature detection techniques, and post-processing techniques for photoplethysmogram processing, especially from an engineering point of view. We performed an extensive search with the PubMed, Google Scholar, Institute of Electrical and Electronics Engineers (IEEE), ScienceDirect, and Web of Science databases. Exclusion conditions did not include the year of publication, but articles not published in English were excluded. Based on 118 articles, we identified four main topics of enabling PPG: (A) PPG waveform, (B) PPG features and clinical applications including basic features based on the original PPG waveform, combined features of PPG, and derivative features of PPG, (C) PPG noise including motion artifact baseline wandering and hypoperfusion, and (D) PPG signal processing including PPG preprocessing, PPG peak detection, and signal quality index. The application field of photoplethysmogram has been extending from the clinical to the mobile environment. Although there is no standardized pre-processing pipeline for PPG signal processing, as PPG data are acquired and accumulated in various ways, the recently proposed machine learning-based method is expected to offer a promising solution.

Normalization of photoplethysmography using deep neural networks for individual and group comparison

Photoplethysmography (PPG) is easy to measure and provides important parameters related to heart rate and arrhythmia. However, automated PPG methods have not been developed because of their susceptibility to motion artifacts and differences in waveform characteristics among individuals. With increasing use of telemedicine, there is growing interest in application of deep neural network (DNN) technology for efficient analysis of vast amounts of PPG data. This study is about an algorithm for measuring a patient's PPG and comparing it with their own data stored previously and with the average data of several groups. Six deep neural networks were used to normalize the PPG waveform according to the heart rate by removing uninformative regions from the PPG, distinguishing between heartbeat and reflection pulses, dividing the heartbeat waveform into 10 segments and averaging the values according to each segments. PPG data were measured using telemedicine in both groups. Group 1 consisted of healthy people aged 25 to 35 years, and Group 2 consisted of patients between 60 and 75 years of age taking antihypertensive medications. The proposed algorithm could accurately determine which group the subject belonged with the newly measured PPG data (AUC = 0.998). On the other hand, errors were frequently observed in identification of individuals (AUC = 0.819).

A Predictive Analysis of Heart Rates Using Machine Learning Techniques

Heart disease, caused by low heart rate, is one of the most significant causes of mortality in the world today. Therefore, it is critical to monitor heart health by identifying the deviation in the heart rate very early, which makes it easier to detect and manage the heart’s function irregularities at a very early stage. The fast-growing use of advanced technology such as the Internet of Things (IoT), wearable monitoring systems and artificial intelligence (AI) in the healthcare systems has continued to play a vital role in the analysis of huge amounts of health-based data for early and accurate disease detection and diagnosis for personalized treatment and prognosis evaluation. It is then important to analyze the effectiveness of using data analytics and machine learning to monitor and predict heart rates using wearable device (accelerometer)-generated data. Hence, in this study, we explored a number of powerful data-driven models including the autoregressive integrated moving average (ARIMA) model, linear regression, support vector regression (SVR), k-nearest neighbor (KNN) regressor, decision tree regressor, random forest regressor and long short-term memory (LSTM) recurrent neural network algorithm for the analysis of accelerometer data to make future HR predictions from the accelerometer’s univariant HR time-series data from healthy people. The performances of the models were evaluated under different durations. Evaluated on a very recently created data set, our experimental results demonstrate the effectiveness of using an ARIMA model with a walk-forward validation and linear regression for predicting heart rate under all durations and other models for durations longer than 1 min. The results of this study show that employing these data analytics techniques can be used to predict future HR more accurately using accelerometers.

Hybrid Convolutional Networks for End-to-End Event Detection in Concurrent PPG and PCG Signals Affected by Motion Artifacts

The accurate detection of physiologically-related events in photopletismographic (PPG) and phocardiographic (PCG) signals, recorded by wearable sensors, is mandatory to perform the estimation of relevant cardiovascular parameters like the heart rate and the blood pressure. However, the measurement performed in uncontrolled conditions without clinical supervision leaves the detection quality particularly susceptible to noise and motion artifacts. The performed work proposed a new fully-automatic computational framework, based on convolutional networks, to identify and localize fiducial points in time as the foot, maximum slope and peak in PPG signal and the S1 sound in the PCG signal, both acquired by a custom chest sensor, described recently in the literature by our group. The novelty entailing a custom neural architecture to process sequentially the PPG and PCG signals. Tests were performed analysing four different acquisition conditions (rest, cycling, rest recovery and walking). Cross-validation results for the three PPG fiducial points showed identification accuracy greater than 93 % and localization error (RMSE) less than 10 ms. As expected, cycling and walking conditions provided worse results than rest and recovery, however reaching an accuracy greater than 90 % and a localization error lower than 15 ms. Likewise, the identification and localization error for S1 sound were greater than 90 % and lower than 25 ms. Overall, this study showcased the ability of the proposed technique to detect events with high accuracy not only for steady acquisitions but also during subject movements. We also showed that the proposed network outperformed traditional Shannon-energy-envelope method in the detection of S1 sound. Therefore, we argue that coupling chest sensors and deep learning processing techniques may disclose wearable devices to unobtrusively acquire health information, being less affected by noise and motion artifacts.

A Real-Time PPG Peak Detection Method for Accurate Determination of Heart Rate during Sinus Rhythm and Cardiac Arrhythmia

OBJECTIVE

We have developed a peak detection algorithm for accurate determination of heart rate, using photoplethysmographic (PPG) signals from a smartwatch, even in the presence of various cardiac rhythms, including normal sinus rhythm (NSR), premature atrial contraction (PAC), premature ventricle contraction (PVC), and atrial fibrillation (AF). Given the clinical need for accurate heart rate estimation in patients with AF, we developed a novel approach that reduces heart rate estimation errors when compared to peak detection algorithms designed for NSR.

METHODS

Our peak detection method is composed of a sequential series of algorithms that are combined to discriminate the various arrhythmias described above. Moreover, a novel Poincaré plot scheme is used to discriminate between basal heart rate AF and rapid ventricular response (RVR) AF, and to differentiate PAC/PVC from NSR and AF. Training of the algorithm was performed only with Samsung Simband smartwatch data, whereas independent testing data which had more samples than did the training data were obtained from Samsung's Gear S3 and Galaxy Watch 3.

RESULTS

The new PPG peak detection algorithm provides significantly lower average heart rate and interbeat interval beat-to-beat estimation errors-30% and 66% lower-and mean heart rate and mean interbeat interval estimation errors-60% and 77% lower-when compared to the best of the seven other traditional peak detection algorithms that are known to be accurate for NSR. Our new PPG peak detection algorithm was the overall best performers for other arrhythmias.

CONCLUSION

The proposed method for PPG peak detection automatically detects and discriminates between various arrhythmias among different waveforms of PPG data, delivers significantly lower heart rate estimation errors for participants with AF, and reduces the number of false negative peaks.

SIGNIFICANCE

By enabling accurate determination of heart rate despite the presence of AF with rapid ventricular response or PAC/PVCs, we enable clinicians to make more accurate recommendations for heart rate control from PPG data.

Q-PPG: Energy-Efficient PPG-Based Heart Rate Monitoring on Wearable Devices

Hearth Rate (HR) monitoring is increasingly performed in wrist-worn devices using low-cost photoplethysmography (PPG) sensors. However, Motion Artifacts (MAs) caused by movements of the subject's arm affect the performance of PPG-based HR tracking. This is typically addressed coupling the PPG signal with acceleration measurements from an inertial sensor. Unfortunately, most standard approaches of this kind rely on hand-tuned parameters, which impair their generalization capabilities and their applicability to real data in the field. In contrast, methods based on deep learning, despite their better generalization, are considered to be too complex to deploy on wearable devices. In this work, we tackle these limitations, proposing a design space exploration methodology to automatically generate a rich family of deep Temporal Convolutional Networks (TCNs) for HR monitoring, all derived from a single "seed" model. Our flow involves a cascade of two Neural Architecture Search (NAS) tools and a hardware-friendly quantizer, whose combination yields both highly accurate and extremely lightweight models. When tested on the PPG-Dalia dataset, our most accurate model sets a new state-of-the-art in Mean Absolute Error. Furthermore, we deploy our TCNs on an embedded platform featuring a STM32WB55 microcontroller, demonstrating their suitability for real-time execution. Our most accurate quantized network achieves 4.41 Beats Per Minute (BPM) of Mean Absolute Error (MAE), with an energy consumption of 47.65 mJ and a memory footprint of 412 kB. At the same time, the smallest network that obtains a MAE 8 BPM, among those generated by our flow, has a memory footprint of 1.9 kB and consumes just 1.79 mJ per inference.

Multimodal Multiresolution Data Fusion Using Convolutional Neural Networks for IoT Wearable Sensing

With advances in circuit design and sensing technology, the acquisition of data from a large number of Internet of Things (IoT) sensors simultaneously to enable more accurate inferences has become mainstream. In this work, we propose a novel convolutional neural network (CNN) model for the fusion of multimodal and multiresolution data obtained from several sensors. The proposed model enables the fusion of multiresolution sensor data, without having to resort to padding/ resampling to correct for frequency resolution differences even when carrying out temporal inferences like high-resolution event detection. The performance of the proposed model is evaluated for sleep apnea event detection, by fusing three different sensor signals obtained from UCD St. Vincent University Hospital's sleep apnea database. The proposed model is generalizable and this is demonstrated by incremental performance improvements, proportional to the number of sensors used for fusion. A selective dropout technique is used to prevent overfitting of the model to any specific high-resolution input, and increase the robustness of fusion to signal corruption from any sensor source. A fusion model with electrocardiogram (ECG), Peripheral oxygen saturation signal (SpO2), and abdominal movement signal achieved an accuracy of 99.72% and a sensitivity of 98.98%. Energy per classification of the proposed fusion model was estimated to be approximately 5.61 μJ for on-chip implementation. The feasibility of pruning to reduce the complexity of the fusion models was also studied.

LSTM-only Model for Low-complexity HR Estimation from Wrist PPG

Continuous and non-invasive cardiovascular monitoring has gained attention due to the miniaturization of wearable devices. Particularly, wrist-worn photoplethysmography (PPG) sensors present an alternative to electrocardiogram recording for heart rate (HR) monitoring as it is cheaper and non-intrusive for daily activities. Yet, the accuracy of PPG measurements is heavily affected by motion artifacts which are inherent to ambulatory environments. In this paper, we propose a low-complexity LSTM-only neural network for HR estimation from a single PPG channel during intense physical activity. This work explored the trade-off between model complexity and accuracy by exploring different model dataflows, number of layers, and number of training epochs to capture the intrinsic time-dependency between PPG samples. The best model achieves a mean absolute error of 4.47 ± 3.68 bpm when evaluated on 12 IEEE SPC subjects.Clinical relevance- This work aims to improve the quality of HR inference from PPG signals using neural network, enabling continuous vital signal monitoring with little interference in daily activities from embedded monitoring devices.

A Deep Learning Approach to Predict Blood Pressure from PPG Signals

Blood Pressure (BP) is one of the four primary vital signs indicating the status of the body's vital (life-sustaining) functions. BP is difficult to continuously monitor using a sphygmomanometer (i.e. a blood pressure cuff), especially in everyday-setting. However, other health signals which can be easily and continuously acquired, such as photoplethysmography (PPG), show some similarities with the Aortic Pressure waveform. Based on these similarities, in recent years several methods were proposed to predict BP from the PPG signal. Building on these results, we propose an advanced personalized data-driven approach that uses a three-layer deep neural network to estimate BP based on PPG signals. Different from previous work, the proposed model analyzes the PPG signal in time-domain and automatically extracts the most critical features for this specific application, then uses a variation of recurrent neural networks (RNN) called Long-Short-Term-Memory (LSTM) to map the extracted features to the BP value associated with that time window. Experimental results on two separate standard hospital datasets, yielded absolute errors mean and absolute error standard deviation for systolic and diastolic BP values outperforming prior works.

Signal Quality Assessment of PPG Signals using STFT Time-Frequency Spectra and Deep Learning Approaches

Photoplethysmography (PPG) is an important signal which contains much physiological information like heart rate and cardiovascular health etc. However, PPG signals are easily corrupted by motion artifacts and body movements during their recordings, which may lead to poor quality. In order to accurately extract cardiovascular information, it is necessary to ensure high PPG quality in these applications. Although there are several existed methods to get the PPG signal quality, those algorithms are complex and the accuracies are not very high. Thus, this work proposes a deep learning network for the signal quality assessment using the STFT time-frequency spectra. A total of 5804 10s signals are preprocessed and transformed into 2D STFT spectra with 250 × 334 pixels. The STFT figures are as the input of the CNN networks, and the model gives the result as good or bad quality. The model accuracy is 98.3% with 98.9% sensitivity, 96.7% specificity, and 98.8% F1-score. And the heart rate error is much reduced after classification with the reference of ECG signals. Thus, the proposed deep learning approaches can be useful in the classification of good and bad PPG signals. As far as we know, this is the first article using deep learning methods combined with STFT time-frequency spectra to get the signal quality assessment of PPG signals.

PPG-based Biometric Identification: Discovering and Identifying a New User

The convenience of Photoplethysmography (PPG) signal acquisition from wearable devices makes it becomes a hot topic in biometric identification. A majority of studies focus on PPG biometric technology in a verification application rather than an identification application. Yet, in the identification application, it is an inevitable problem in discovering and identifying a new user. However, so far few works have investigated this problem. Existing approaches can only identify trained old users. Their identification model needs to be retrained when a new user joins, which reduces the identification accuracy. This work investigates the approach and performance of identifying both old users and new users on a deep neural network trained only by old users. We used a deep neural network as a feature extractor, and the distance of the feature vector to discover and identify a new user, which avoids retraining the identification model. On the BIDMC data set, we achieved an accuracy of more than 99% for old users, an accuracy of more than 90% for discovering a new user, and an average accuracy of about 90% for identifying a new user. Our proposed approach can accurately identify old users and has feasibility in discovering and identifying a new user without retraining in the identification application.

Learning based Quality Indicator Aiding Heart Rate Estimation in Wrist-Worn PPG

Technological advancements and miniaturization of wearable sensors have enabled long-term pervasive physiological monitoring. Wrist-worn photoplethysmography (PPG) sensors, although quite popular owing to their form factor, suffer from poor signal quality in ambulatory settings due to motion artifacts. This affects the reliable estimation of vital cardiac parameters, especially during motion/activities of daily living. Hence, in this paper, we have developed a learningbased quality indicator engine (QIE), evaluating on 23 PPG records of the TROIKA database. The engine comprises the fundamental steps of frequency-domain feature extraction, feature selection and classification by an ensemble of decision trees, achieving an accuracy of 83% in the testing set. To the best of our knowledge, the proposed quality engine is the first to be evaluated on wrist-PPG data acquired during various physical activities and with respect to improvement in heart rate (HR) estimation. The QIE demonstrated an average improvement of 43% in HR estimation, when used in conjunction with state-ofthe-art WFPV algorithm.Clinical Relevance- The proposed quality indicator engine helps to increase the efficacy of vital parameter estimation (e.g. heart rate) from pervasive, wrist-worn PPG sensors on the backdrop of motion artifacts when used in ambulatory settings (e.g. activities of daily living).

Fixed-Point NLMS and IPNLMS VLSI Architectures for Accurate FECG and FHR Processing

Capturing signals without noise and interference while monitoring the maternal abdomen's fetal electrocardiogram (FECG) is a challenging task. This method can provide fetal monitoring for long hours, not harming the pregnant woman or the fetus. Such non-invasive FECG raw signal suffers from various interference sources as the bio-electric maternal potentials include her ECG component. Therefore, a critical step in the non-invasive FECG is to design the filtering of components derived from the maternal ECG. There is an increasing demand for portable devices to extract a pure FECG signal and to detect fetal heart rate (FHR) with precision. Dedicated CMOS architectures enable higher energy efficiency in portable devices. This paper proposes VLSI architectures dedicated to FECG extraction and FHR processing. Fixed-point architectures for the FECG detection exploring the NLMS (normalized least mean square), IPNLMS (improved proportional NLMS), and three different division VLSI CMOS architectures are designed herein. An architecture based on the Pan-Tompkins algorithm that processes the FECG for extracting the FHR, extending the functionally of the system, is also proposed. The results show that the NLMS and IPNLMS based architectures effectively detect the R-peaks of FECG with a detection accuracy of 92.86% and 93.75%, respectively. The synthesis results shows that our NLMS architecture proposal saves 13.3 % energy, due to a reduction of 279 clock cycles, compared to the state of the art. On the other hand, the IPNLMS algorithm results in +0.89% detection accuracy at the price of 42% additional energy consumption w.r.t NLMS.

Prediction of arterial blood pressure waveforms from photoplethysmogram signals via fully convolutional neural networks

Cardiovascular disease (CVD) is one of the most serious diseases threatening human health. Arterial blood pressure (ABP) waveforms, containing vivid cardiovascular information, are of great significance for the diagnosis and the prevention of CVD. This paper proposes a deep learning model, named ABP-Net, to transform photoplethysmogram (PPG) signals into ABP waveforms that contain vital physiological information related to cardiovascular systems. In order to guarantee the quality of the predicted ABP waveforms, the structure of the network, the input signals and the loss functions are carefully designed. Specifically, a Wave-U-Net, one kind of fully convolutional neural networks (CNN), is taken as the core architecture of the ABP-Net. Besides the original PPG signals, its first derivative and second derivative signals are all utilized as the inputs of the ABP-Net. Additionally, the maximal absolute loss, accompany with the mean squared error loss is employed to ensure the match of the predicted ABP waveform with the reference one. The performance of the proposed ABP network is tested on the public MIMIC II database both in subject-dependent and subject-independent manners. Both results verify the superior performance of the proposed model over those existing methods accordingly. The mean absolute error (MAE) and the root-mean-square error (RMSE) between the predicted waveforms via the ABP-Net and the reference ones are 3.20 mmHg and 4.38 mmHg during the subject-dependent experiments while those are 5.57 mmHg and 7.15 mmHg during the subject-independent experiments. Benefiting from the predicted high-quality ABP waveforms, more ABP related physiological parameters can be better obtained, which effectively expands the application scope of PPG devices.

Transcending Conventional Biometry Frontiers: Diffusive Dynamics PPG Biometry

This paper presents the first photoplethysmographic (PPG) signal dynamic-based biometric authentication system with a Siamese convolutional neural network (CNN). Our method extracts the PPG signal's biometric characteristics from its diffusive dynamics, characterized by geometric patterns in the (p,q)-planes specific to the 0-1 test. PPG signal diffusive dynamics are strongly dependent on the vascular bed's biostructure, unique to each individual. The dynamic characteristics of the PPG signal are more stable over time than its morphological features, particularly in the presence of psychosomatic conditions. Besides its robustness, our biometric method is anti-spoofing, given the complex nature of the blood network. Our proposal trains using a national research study database with 40 real-world PPG signals measured with commercial equipment. Biometric system results for input data, raw and preprocessed, are studied and compared with eight primary biometric methods related to PPG, achieving the best equal error rate (ERR) and processing times with a single attempt, among all of them.

Psychological health analysis based on fuzzy assisted neural network model for sports person

Stress is indeed a life aspect that influences everyone, even though athletes seem to suffer from it one step ahead of others because of the extent they are expected to balance between coursework, workouts, and competitions, along with everyday life and family stress. Therefore, an efficient psychological health analysis for sportspersons is crucial in sports training. This paper introduces a Fuzzy-assisted Neural Network model for Psychological Health Analysis (FNN-PHA) to assess mental stress by monitoring the Electro Cardio Gram signal (ECG), Electroencephalogram (EEG), and Pulse rate. This paper integrates the fuzzy assisted Petri nets, fuzzy assisted k-complex detector, and fuzzy assisted transient time analyzer to ensure the psychological health analysis neural network model’s adaptive performance. The strength of the proposed fuzzy model demonstrates interpretability against the accuracy of different criteria. The simulation analysis shows that the FNN-PHA model enhances the prediction ratio of 98.7%, emotional stability of 96.7%, personal growth of 95.7%, physical fitness level of 97.8%, and depression ratio of 12.5% when compared to other existing models.

A Review of Deep Learning-Based Contactless Heart Rate Measurement Methods

The interest in contactless or remote heart rate measurement has been steadily growing in healthcare and sports applications. Contactless methods involve the utilization of a video camera and image processing algorithms. Recently, deep learning methods have been used to improve the performance of conventional contactless methods for heart rate measurement. After providing a review of the related literature, a comparison of the deep learning methods whose codes are publicly available is conducted in this paper. The public domain UBFC dataset is used to compare the performance of these deep learning methods for heart rate measurement. The results obtained show that the deep learning method PhysNet generates the best heart rate measurement outcome among these methods, with a mean absolute error value of 2.57 beats per minute and a mean square error value of 7.56 beats per minute.

Interface sensors with skin piezo-thermic transduction enable motion artifact removal for wearable physiological monitoring

Accurate measurements on physiological parameters using wearable monitoring devices during physical exercises are essential for personal healthcare and rehabilitation training, but still challenging owing to various motion artifacts (MA) caused by the interfacial dynamic change between wearable sensors and human skin. Here, we propose an interface sensor to detect noncontact proximity and contact pressure between wearable sensors and human skin. The interface sensor employs natural piezo-thermic transduction of human skin and enables direct interfacial proximity/pressure detection by using simple thin-film thermistors to detect the interfacial thermal field change. We develop a wearable watch-type heart rate (HR) monitor utilizing interface sensors to remove MA for a photoplethysmography (PPG) sensor through adaptive filtering. To validate the method, we conduct experiments for multiple subjects, who carry out HR monitoring using the wearable device while doing various physical exercises. The PPG-based HR estimations are corrected through MA removal using interface sensors and compared with that using conventional accelerometer-based MA removal. The experimental results verify that the interface sensors capture the interfacial dynamic change between the PPG sensor and skin better, and obtain more accurate HR estimations during irregular and muscle strength exercises. Utilizing natural transduction of human skin and simple thermometry, the interface sensor provides an advantageous way to overcome MA for wearable monitoring devices during physical activities and thus broadens wearable monitoring applications.

Adaptive Extreme Edge Computing for Wearable Devices

Wearable devices are a fast-growing technology with impact on personal healthcare for both society and economy. Due to the widespread of sensors in pervasive and distributed networks, power consumption, processing speed, and system adaptation are vital in future smart wearable devices. The visioning and forecasting of how to bring computation to the edge in smart sensors have already begun, with an aspiration to provide adaptive extreme edge computing. Here, we provide a holistic view of hardware and theoretical solutions toward smart wearable devices that can provide guidance to research in this pervasive computing era. We propose various solutions for biologically plausible models for continual learning in neuromorphic computing technologies for wearable sensors. To envision this concept, we provide a systematic outline in which prospective low power and low latency scenarios of wearable sensors in neuromorphic platforms are expected. We successively describe vital potential landscapes of neuromorphic processors exploiting complementary metal-oxide semiconductors (CMOS) and emerging memory technologies (e.g., memristive devices). Furthermore, we evaluate the requirements for edge computing within wearable devices in terms of footprint, power consumption, latency, and data size. We additionally investigate the challenges beyond neuromorphic computing hardware, algorithms and devices that could impede enhancement of adaptive edge computing in smart wearable devices.

Machine Learning-Based Pulse Wave Analysis for Early Detection of Abdominal Aortic Aneurysms Using In Silico Pulse Waves

An abdominal aortic aneurysm (AAA) is usually asymptomatic until rupture, which is associated with extremely high mortality. Consequently, the early detection of AAAs is of paramount importance in reducing mortality; however, most AAAs are detected by medical imaging only incidentally. The aim of this study was to investigate the feasibility of machine learning-based pulse wave (PW) analysis for the early detection of AAAs using a database of in silico PWs. PWs in the large systemic arteries were simulated using one-dimensional blood flow modelling. A database of in silico PWs representative of subjects (aged 55, 65 and 75 years) with different AAA sizes was created by varying the AAA-related parameters with major impacts on PWs—identified by parameter sensitivity analysis—in an existing database of in silico PWs representative of subjects without AAAs. Then, a machine learning architecture for AAA detection was trained and tested using the new in silico PW database. The parameter sensitivity analysis revealed that the AAA maximum diameter and stiffness of the large systemic arteries were the dominant AAA-related biophysical properties considerably influencing the PWs. However, AAA detection by PW indexes was compromised by other non-AAA related cardiovascular parameters. The proposed machine learning model produced a sensitivity of 86.8 % and a specificity of 86.3 % in early detection of AAA from the photoplethysmogram PW signal measured in the digital artery with added random noise. The number of false positive and negative results increased with increasing age and decreasing AAA size, respectively. These findings suggest that machine learning-based PW analysis is a promising approach for AAA screening using PW signals acquired by wearable devices.

SPARE: A Spectral Peak Recovery Algorithm for PPG Signals Pulsewave Reconstruction in Multimodal Wearable Devices

The photoplethysmographic (PPG) signal is an unobtrusive blood pulsewave measure that has recently gained popularity in the context of the Internet of Things. Even though it is commonly used for heart rate detection, it has been lately employed on multimodal health and wellness monitoring applications. Unfortunately, this signal is prone to motion artifacts, making it almost useless in all situations where a person is not entirely at rest. To overcome this issue, we propose SPARE, a spectral peak recovery algorithm for PPG signals pulsewave reconstruction. Our solution exploits the local semiperiodicity of the pulsewave signal, together with the information about the cardiac rhythm provided by an available simultaneous ECG, to reconstruct its full waveform, even when affected by strong artifacts. The developed algorithm builds on state-of-the-art signal decomposition methods, and integrates novel techniques for signal reconstruction. Experimental results are reported both in the case of PPG signals acquired during physical activity and at rest, but corrupted in a systematic way by synthetic noise. The full PPG waveform reconstruction enables the identification of several health-related features from the signal, showing an improvement of up to 65% in the detection of different biomarkers from PPG signals affected by noise.

Explainability Metrics of Deep Convolutional Networks for Photoplethysmography Quality Assessment

Photoplethysmography (PPG) is a noninvasive way to monitor various aspects of the circulatory system, and is becoming more and more widespread in biomedical processing. Recently, deep learning methods for analyzing PPG have also become prevalent, achieving state of the art results on heart rate estimation, atrial fibrillation detection, and motion artifact identification. Consequently, a need for interpretable deep learning has arisen within the field of biomedical signal processing. In this paper, we pioneer novel explanatory metrics which leverage domain-expert knowledge to validate a deep learning model. We visualize model attention over a whole testset using saliency methods and compare it to human expert annotations. Congruence, our first metric, measures the proportion of model attention within expert-annotated regions. Our second metric, Annotation Classification, measures how much of the expert annotations our deep learning model pays attention to. Finally, we apply our metrics to compare between a signal based model and an image based model for PPG signal quality classification. Both models are deep convolutional networks based on the ResNet architectures. We show that our signal-based one dimensional model acts in a more explainable manner than our image based model; on average 50.78% of the one dimensional model's attention are within expert annotations, whereas 36.03% of the two dimensional model's attention are within expert annotations. Similarly, when thresholding the one dimensional model attention, one can more accurately predict if each pixel of the PPG is annotated as artifactual by an expert. Through this testcase, we demonstrate how our metrics can provide a quantitative and dataset-wide analysis of how explainable the model is.