Generalized Deep Neural Network Model for Cuffless Blood Pressure Estimation with Photoplethysmogram Signal Only

Examining the challenges of blood pressure estimation via photoplethysmogram

The use of observed wearable sensor data (e.g., photoplethysmograms [PPG]) to infer health measures (e.g., glucose level or blood pressure) is a very active area of research. Such technology can have a significant impact on health screening, chronic disease management and remote monitoring. A common approach is to collect sensor data and corresponding labels from a clinical grade device (e.g., blood pressure cuff) and train deep learning models to map one to the other. Although well intentioned, this approach often ignores a principled analysis of whether the input sensor data have enough information to predict the desired metric. We analyze the task of predicting blood pressure from PPG pulse wave analysis. Our review of the prior work reveals that many papers fall prey to data leakage and unrealistic constraints on the task and preprocessing steps. We propose a set of tools to help determine if the input signal in question (e.g., PPG) is indeed a good predictor of the desired label (e.g., blood pressure). Using our proposed tools, we found that blood pressure prediction using PPG has a high multi-valued mapping factor of 33.2% and low mutual information of 9.8%. In comparison, heart rate prediction using PPG, a well-established task, has a very low multi-valued mapping factor of 0.75% and high mutual information of 87.7%. We argue that these results provide a more realistic representation of the current progress toward the goal of wearable blood pressure measurement via PPG pulse wave analysis. For code, see our project page: https://github.com/lirus7/PPG-BP-Analysis.

Non-invasive prediction of massive transfusion during surgery using intraoperative hemodynamic monitoring data

OBJECTIVE

Failure to receive prompt blood transfusion leads to severe complications if massive bleeding occurs during surgery. For the timely preparation of blood products, predicting the possibility of massive transfusion (MT) is essential to decrease morbidity and mortality. This study aimed to develop a model for predicting MT 10 min in advance using non-invasive bio-signal waveforms that change in real-time.

METHODS

In this retrospective study, we developed a deep learning-based algorithm (DLA) to predict intraoperative MT within 10 min. MT was defined as the transfusion of 3 or more units of red blood cells within an hour. The datasets consisted of 18,135 patients who underwent surgery at Seoul National University Hospital (SNUH) for model development and internal validation and 621 patients who underwent surgery at the Boramae Medical Center (BMC) for external validation. We constructed the DLA by using features extracted from plethysmography (collected at 500 Hz) and hematocrit measured during surgery.

RESULTS

Among 18,135 patients in SNUH and 621 patients in BMC, 265 patients (1.46%) and 14 patients (2.25%) received MT during surgery, respectively. The area under the receiver operating characteristic curve (AUROC) of DLA predicting intraoperative MT before 10 min was 0.962 (95% confidence interval [CI], 0.948-0.974) in internal validation and 0.922 (95% CI, 0.882-0.959) in external validation, respectively.

CONCLUSION

The DLA can successfully predict intraoperative MT using non-invasive bio-signal waveforms.

A Systematic Approach Focused on Machine Learning Models for Exploring the Landscape of Physiological Measurement and Estimation Using Photoplethysmography (PPG)

A non-invasive optical technique known as photoplethysmography (PPG) can be used to provide various physiological measurements and estimations. PPG can be used to assess cardiovascular disease (CVD). Hypertension is a primary risk factor for CVD and a major health problem worldwide. PPG is popular because of its important applications in the evaluation of cardiac activity, variations in venous blood volume, blood oxygen saturation, blood pressure and heart rate variability, etc. In this study, we provide a comprehensive analysis of the extraction of various physiological parameters using PPG waveforms. In addition, we focused on the role of machine learning (ML) models used for the estimation of blood pressure and hypertension classification based on PPG waveforms to make future research and innovation recommendations. This study will be helpful for researchers, scientists, and medical practitioners working on PPG waveforms for monitoring, screening, and diagnosis, as a comparative study or reference.

Pulse2AI: An Adaptive Framework to Standardize and Process Pulsatile Wearable Sensor Data for Clinical Applications

Goal: To establish Pulse2AI as a reproducible data preprocessing framework for pulsatile signals that generate high-quality machine-learning-ready datasets from raw wearable recordings. Methods: We proposed an end-to-end data preprocessing framework that adapts multiple pulsatile signal modalities and generates machine-learning-ready datasets agnostic to downstream medical tasks. Results: a dataset preprocessed by Pulse2AI improved systolic blood pressure estimation by 29.58%, from 11.41 to 8.03 mmHg in root-mean-square-error (RMSE) and its diastolic counterpart by 26.01%, from 7.93 to 5.87 mmHg in RMSE. For respiration rate (RR) estimation, Pulse2AI boosted performance by 19.69%, from 1.47 to 1.18 breaths per minute (BrPM) in mean-absolute-error (MAE). Conclusion: Pulse2AI turns pulsatile signals into machine learning (ML) ready datasets for arbitrary remote health monitoring tasks. We tested Pulse2AI on multiple pulsatile modalities and demonstrated its efficacy in two medical applications. This work bridges valuable assets in remote sensing and internet of medical things to ML-ready datasets for medical modeling.

Blood pressure estimation and classification using a reference signal-less photoplethysmography signal: a deep learning framework

The markers that help to predict th function of a cardiovascular system are hemodynamic parameters like blood pressure (BP), stroke volume, heart rate, and cardiac output. Continuous analysis of hemodynamic parameters such as BP can detect abnormalities earlier, preventing cardiovascular diseases (CVDs). However, sometimes due to motion artifacts, it becomes difficult to monitor the BP accurately and classify it. This work presents an optimized deep learning model having the capability to estimate the systolic blood pressure (SBP) and diastolic blood pressure (DBP) and classify the BP stages simultaneously from the same network using only a single channel photoplethysmography (PPG) signal. The proposed model is designed by exploiting the deep learning framework of a convolutional neural network (CNN), exhibiting the inherent ability to extract features automatically. Moreover, the proposed framework utilizes the superlet transform method to transform a 1-D PPG signal into a 2-D super-resolution time-frequency (TF) spectrogram. A superlet transform separates the peaks related to true PPG signal components and motion artifacts components. Thus, the superlet provides a robust realtime approach to accurately estimating and classifying BP using a single PPG sensor signal and does not require additional ECG and PPG sensor signals for reference. Using a super-resolution spectrogram and CNN model makes the method profitable in motion artifact removal, feature selection, and extraction. Hence the proposed framework becomes less complex for deployment on wearable devices having limited battery resources. The performance of the proposed framework is demonstrated on the publicly available larger dataset MIMIC-III. This work obtained a mean absolute error (MAE) of 2.71 mmHg and 2.42 mmHg for SBP and DBP, respectively. The classification accuracy for the SBP prediction is about 96.79%, whereas it is 98.94% for DBP. From a motion artifact-affected PPG signal, SBP and DBP are estimated. Then the estimated BP is classified into three categories: normotension, prehypertension, and hypertension, and is compared with the state of art methods to show the effectiveness of the proposed optimized framework.

Reconstruction of central arterial pressure waveform based on CBi-SAN network from radial pressure waveform

The central arterial pressure (CAP) is an important physiological indicator of the human cardiovascular system which represents one of the greatest threats to human health. Accurate non-invasive detection and reconstruction of CAP waveforms are crucial for the reliable treatment of cardiovascular system diseases. However, the traditional methods are reconstructed with relatively low accuracy, and some deep learning neural network models also have difficulty in extracting features, as a result, these methods have potential for further advancement. In this study, we proposed a novel model (CBi-SAN) to implement an end-to-end relationship from radial artery pressure (RAP) waveform to CAP waveform, which consisted of the convolutional neural network (CNN), the bidirectional long-short-time memory network (BiLSTM), and the self-attention mechanism to improve the performance of CAP reconstruction. The data on invasive measurements of CAP and RAP waveform were used in 62 patients before and after medication to develop and validate the performance of CBi-SAN model for reconstructing CAP waveform. We compared it with traditional methods and deep learning models in mean absolute error (MAE), root mean square error (RMSE), and Spearman correlation coefficient (SCC). Study results indicated the CBi-SAN model performed great performance on CAP waveform reconstruction (MAE: 2.23 ± 0.11 mmHg, RMSE: 2.21 ± 0.07 mmHg), concurrently, the best reconstruction effect was obtained in the central artery systolic pressure (CASP) and the central artery diastolic pressure(CADP) (RMSECASP: 2.94 ± 0.48 mmHg, RMSECADP: 1.96 ± 0.06 mmHg). These results implied the performance of the CAP reconstruction based on CBi-SAN model was superior to the existing methods, hopped to be effectively applied to clinical practice in the future.

Quantitative analysis of pulse arrival time and PPG morphological features based cuffless blood pressure estimation: a comparative study between diabetic and non-diabetic groups

Pulse arrival time (PAT) and PPG morphological features have attracted much interest in cuffless blood pressure (BP) estimation, but their effects are not clearly understood when vascular characteristics are affected by diseases such as diabetes. This work quantitatively analyzes the effect of diabetic disease on the PAT and PPG morphological features-based BP estimation. We selected 112 diabetic patients and 308 non-diabetic subjects from VitalDB, and extracted 16 features including PAT, PPG morphological features, and heart rate. BP estimation performance was statistically compared between groups using linear regression models with several feature sets, and the relative importance of each feature in the optimal feature set was extracted. As a result, the standard deviation of the error and mean absolute error of PAT-based BP estimation were significantly higher in the diabetic group than in the non-diabetic group (p < 0.01). A feature set containing PAT and PPG morphological features achieved the best performance in both groups. However, the relative importance of each feature for BP estimation differed notably between groups. The results indicate that different features are important depending on the vascular characteristics, which could help to construct different models to accommodate specific diseases.

Non-Invasive Blood Pressure Sensing via Machine Learning

In this paper, a machine learning (ML) approach to estimate blood pressure (BP) using photoplethysmography (PPG) is presented. The final aim of this paper was to develop ML methods for estimating blood pressure (BP) in a non-invasive way that is suitable in a telemedicine health-care monitoring context. The training of regression models useful for estimating systolic blood pressure (SBP) and diastolic blood pressure (DBP) was conducted using new extracted features from PPG signals processed using the Maximal Overlap Discrete Wavelet Transform (MODWT). As a matter of fact, the interest was on the use of the most significant features obtained by the Minimum Redundancy Maximum Relevance (MRMR) selection algorithm to train eXtreme Gradient Boost (XGBoost) and Neural Network (NN) models. This aim was satisfactorily achieved by also comparing it with works in the literature; in fact, it was found that XGBoost models are more accurate than NN models in both systolic and diastolic blood pressure measurements, obtaining a Root Mean Square Error (RMSE) for SBP and DBP, respectively, of 5.67 mmHg and 3.95 mmHg. For SBP measurement, this result is an improvement compared to that reported in the literature. Furthermore, the trained XGBoost regression model fulfills the requirements of the Association for the Advancement of Medical Instrumentation (AAMI) as well as grade A of the British Hypertension Society (BHS) standard.

Nocturnal Blood Pressure Estimation from Sleep Plethysmography Using Machine Learning

BACKGROUND

Elevated nocturnal blood pressure (BP) is a risk factor for cardiovascular disease (CVD) and mortality. Cuffless BP assessment aided by machine learning could be a desirable alternative to traditional cuff-based methods for monitoring BP during sleep. We describe a machine-learning-based algorithm for predicting nocturnal BP using single-channel fingertip plethysmography (PPG) in healthy adults.

METHODS

Sixty-eight healthy adults with no apparent sleep or CVD (53% male), with a median (IQR) age of 29 (23-46 years), underwent overnight polysomnography (PSG) with fingertip PPG and ambulatory blood pressure monitoring (ABPM). Features based on pulse morphology were extracted from the PPG waveforms. Random forest models were used to predict night-time systolic blood pressure (SBP) and diastolic blood pressure (DBP).

RESULTS

Our model achieved the highest out-of-sample performance with a window length of 7 s across window lengths explored (60 s, 30 s, 15 s, 7 s, and 3 s). The mean absolute error (MAE ± STD) was 5.72 ± 4.51 mmHg for SBP and 4.52 ± 3.60 mmHg for DBP. Similarly, the root mean square error (RMSE ± STD) was 6.47 ± 1.88 mmHg for SBP and 4.62 ± 1.17 mmHg for DBP. The mean correlation coefficient between measured and predicted values was 0.87 for SBP and 0.86 for DBP. Based on Shapley additive explanation (SHAP) values, the most important PPG waveform feature was the stiffness index, a marker that reflects the change in arterial stiffness.

CONCLUSION

Our results highlight the potential of machine learning-based nocturnal BP prediction using single-channel fingertip PPG in healthy adults. The accuracy of the predictions demonstrated that our cuffless method was able to capture the dynamic and complex relationship between PPG waveform characteristics and BP during sleep, which may provide a scalable, convenient, economical, and non-invasive means to continuously monitor blood pressure.

A Multi-Parameter Fusion Method for Cuffless Continuous Blood Pressure Estimation Based on Electrocardiogram and Photoplethysmogram

Blood pressure (BP) is an essential physiological indicator to identify and determine health status. Compared with the isolated BP measurement conducted by traditional cuff approaches, cuffless BP monitoring can reflect the dynamic changes in BP values and is more helpful to evaluate the effectiveness of BP control. In this paper, we designed a wearable device for continuous physiological signal acquisition. Based on the collected electrocardiogram (ECG) and photoplethysmogram (PPG), we proposed a multi-parameter fusion method for noninvasive BP estimation. An amount of 25 features were extracted from processed waveforms and Gaussian copula mutual information (MI) was introduced to reduce feature redundancy. After feature selection, random forest (RF) was trained to realize systolic BP (SBP) and diastolic BP (DBP) estimation. Moreover, we used the records in public MIMIC-III as the training set and private data as the testing set to avoid data leakage. The mean absolute error (MAE) and standard deviation (STD) for SBP and DBP were reduced from 9.12 ± 9.83 mmHg and 8.31 ± 9.23 mmHg to 7.93 ± 9.12 mmHg and 7.63 ± 8.61 mmHg by feature selection. After calibration, the MAE was further reduced to 5.21 mmHg and 4.15 mmHg. The result showed that MI has great potential in feature selection during BP prediction and the proposed multi-parameter fusion method can be used for long-term BP monitoring.

A Novel Cuffless Blood Pressure Prediction: Uncovering New Features and New Hybrid ML Models

This paper investigates new feature extraction and regression methods for predicting cuffless blood pressure from PPG signals. Cuffless blood pressure is a technology that measures blood pressure without needing a cuff. This technology can be used in various medical applications, including home health monitoring, clinical uses, and portable devices. The new feature extraction method involves extracting meaningful features (time and chaotic features) from the PPG signals in the prediction of systolic blood pressure (SBP) and diastolic blood pressure (DBP) values. These extracted features are then used as inputs to regression models, which are used to predict cuffless blood pressure. The regression model performances were evaluated using root mean squared error (RMSE), R2, mean square error (MSE), and the mean absolute error (MAE). The obtained RMSE was 4.277 for systolic blood pressure (SBP) values using the Matérn 5/2 Gaussian process regression model. The obtained RMSE was 2.303 for diastolic blood pressure (DBP) values using the rational quadratic Gaussian process regression model. The results of this study have shown that the proposed feature extraction and regression models can predict cuffless blood pressure with reasonable accuracy. This study provides a novel approach for predicting cuffless blood pressure and can be used to develop more accurate models in the future.

Photoplethysmography Signal Wavelet Enhancement and Novel Features Selection for Non-Invasive Cuff-Less Blood Pressure Monitoring

In this paper, new features relevant to blood pressure (BP) estimation using photoplethysmography (PPG) are presented. A total of 195 features, including the proposed ones and those already known in the literature, have been calculated on a set composed of 50,000 pulses from 1080 different patients. Three feature selection methods, namely Correlation-based Feature Selection (CFS), RReliefF and Minimum Redundancy Maximum Relevance (MRMR), have then been applied to identify the most significant features for BP estimation. Some of these features have been extracted through a novel PPG signal enhancement method based on the use of the Maximal Overlap Discrete Wavelet Transform (MODWT). As a matter of fact, the enhanced signal leads to a reliable identification of the characteristic points of the PPG signal (e.g., systolic, diastolic and dicrotic notch points) by simple means, obtaining results comparable with those from purposely defined algorithms. For systolic points, mean and std of errors computed as the difference between the locations obtained using a purposely defined already known algorithm and those using the MODWT enhancement are, respectively, 0.0097 s and 0.0202 s; for diastolic points they are, respectively, 0.0441 s and 0.0486 s; for dicrotic notch points they are 0.0458 s and 0.0896 s. Hence, this study leads to the selection of several new features from the MODWT enhanced signal on every single pulse extracted from PPG signals, in addition to features already known in the literature. These features can be employed to train machine learning (ML) models useful for estimating systolic blood pressure (SBP) and diastolic blood pressure (DBP) in a non-invasive way, which is suitable for telemedicine health-care monitoring.

PPG2ABP: Translating Photoplethysmogram (PPG) Signals to Arterial Blood Pressure (ABP) Waveforms

Cardiovascular diseases are one of the most severe causes of mortality, annually taking a heavy toll on lives worldwide. Continuous monitoring of blood pressure seems to be the most viable option, but this demands an invasive process, introducing several layers of complexities and reliability concerns due to non-invasive techniques not being accurate. This motivates us to develop a method to estimate the continuous arterial blood pressure (ABP) waveform through a non-invasive approach using Photoplethysmogram (PPG) signals. We explore the advantage of deep learning, as it would free us from sticking to ideally shaped PPG signals only by making handcrafted feature computation irrelevant, which is a shortcoming of the existing approaches. Thus, we present PPG2ABP, a two-stage cascaded deep learning-based method that manages to estimate the continuous ABP waveform from the input PPG signal with a mean absolute error of 4.604 mmHg, preserving the shape, magnitude, and phase in unison. However, the more astounding success of PPG2ABP turns out to be that the computed values of Diastolic Blood Pressure (DBP), Mean Arterial Pressure (MAP), and Systolic Blood Pressure (SBP) from the estimated ABP waveform outperform the existing works under several metrics (mean absolute error of 3.449 ± 6.147 mmHg, 2.310 ± 4.437 mmHg, and 5.727 ± 9.162 mmHg, respectively), despite that PPG2ABP is not explicitly trained to do so. Notably, both for DBP and MAP, we achieve Grade A in the BHS (British Hypertension Society) Standard and satisfy the AAMI (Association for the Advancement of Medical Instrumentation) standard.

Beat-to-Beat Blood Pressure Estimation by Photoplethysmography and Its Interpretation

Blood pressure (BP) is among the most important vital signals. Estimation of absolute BP solely using photoplethysmography (PPG) has gained immense attention over the last years. Available works differ in terms of used features as well as classifiers and bear large differences in their results. This work aims to provide a machine learning method for absolute BP estimation, its interpretation using computational methods and its critical appraisal in face of the current literature. We used data from three different sources including 273 subjects and 259,986 single beats. We extracted multiple features from PPG signals and its derivatives. BP was estimated by xgboost regression. For interpretation we used Shapley additive values (SHAP). Absolute systolic BP estimation using a strict separation of subjects yielded a mean absolute error of 9.456mmHg and correlation of 0.730. The results markedly improve if data separation is changed (MAE: 6.366mmHg, r: 0.874). Interpretation by means of SHAP revealed four features from PPG, its derivation and its decomposition to be most relevant. The presented approach depicts a general way to interpret multivariate prediction algorithms and reveals certain features to be valuable for absolute BP estimation. Our work underlines the considerable impact of data selection and of training/testing separation, which must be considered in detail when algorithms are to be compared. In order to make our work traceable, we have made all methods available to the public.

Subject-Based Model for Reconstructing Arterial Blood Pressure from Photoplethysmogram

The continuous prediction of arterial blood pressure (ABP) waveforms via non-invasive methods is of great significance for the prevention and treatment of cardiovascular disease. Photoplethysmography (PPG) can be used to reconstruct ABP signals due to having the same excitation source and high signal similarity. The existing methods of reconstructing ABP signals from PPG only focus on the similarities between systolic, diastolic, and mean arterial pressures without evaluating their global similarity. This paper proposes a deep learning model with a W-Net architecture to reconstruct ABP signals from PPG. The W-Net consists of two concatenated U-Net architectures, the first acting as an encoder and the second as a decoder to reconstruct ABP from PPG. Five hundred records of different lengths were used for training and testing. The experimental results yielded high values for the similarity measures between the reconstructed ABP signals and their reference ABP signals: the Pearson correlation, root mean square error, and normalized dynamic time warping distance were 0.995, 2.236 mmHg, and 0.612 mmHg on average, respectively. The mean absolute errors of the SBP and DBP were 2.602 mmHg and 1.450 mmHg on average, respectively. Therefore, the model can reconstruct ABP signals that are highly similar to the reference ABP signals.

A Hardware-based Lightweight ANN for Real-time Wearable Blood Pressure Estimation

Blood pressure (BP) is an important indicator of the state of cardiovascular health. BP estimation is an essential method to prevent the occurrence of hypertension. Currently, there is a strong focus on low power design for a wearable BP estimation device. This paper proposes a lightweight artificial neural network (ANN) for BP estimation and implements it on an ultra-low-power application-specific integrated circuit (ASIC). On the test set, the mean absolute error (MAE) and standard deviation (SD) of the estimated systolic BP and diastolic BP are 2.47 ± 3.48 mmHg and 1.45 ± 1.88 mmHg. Besides, in the case of 8-bit quantization, the MAE ± SD of the estimated systolic BP and diastolic BP are 12.41 ± 5.32 mmHg and 6.29 ± 3.03 mmHg respectively. The regression result R² of overall SBP and DBP is 0.9702. This ASIC whose power is 19.72 µW is validated via the 0.18 µm CMOS process, occupying an area of 730 µmx 730 µm.

Methods for Continuous Blood Pressure Estimation Using Temporal Convolutional Neural Networks and Ensemble Empirical Mode Decomposition

Arterial blood pressure is not only an important index that must be measured in routine physical examination but also a key monitoring parameter of the cardiovascular system in cardiac surgery, drug testing, and intensive care. To improve the measurement accuracy of continuous blood pressure, this paper uses photoplethysmography (PPG) signals to estimate diastolic blood pressure and systolic blood pressure based on ensemble empirical mode decomposition (EEMD) and temporal convolutional network (TCN). In this method, the clean PPG signal is decomposed by EEMD to obtain n-order intrinsic mode functions (IMF), and then the IMF and the original PPG are input into the constructed TCN neural network model, and the results are output. The results show that TCN has better performance than CNN, CNN-LSTM, and CNN-GRU. Using the data added with IMF, the results of the above neural network model are better than those of the model with only PPG as input, in which the systolic blood pressure (SBP) and diastolic blood pressure (DBP) results of EEMD-TCN are −1.55 ± 9.92 mmHg and 0.41 ± 4.86 mmHg. According to the estimation results, DBP meets the requirements of the AAMI standard, BHS evaluates it as Grade A, SD of SBP is close to the standard AAMI, and BHS evaluates it as Grade B.

A Shallow U-Net Architecture for Reliably Predicting Blood Pressure (BP) from Photoplethysmogram (PPG) and Electrocardiogram (ECG) Signals

Cardiovascular diseases are the most common causes of death around the world. To detect and treat heart-related diseases, continuous blood pressure (BP) monitoring along with many other parameters are required. Several invasive and non-invasive methods have been developed for this purpose. Most existing methods used in hospitals for continuous monitoring of BP are invasive. On the contrary, cuff-based BP monitoring methods, which can predict systolic blood pressure (SBP) and diastolic blood pressure (DBP), cannot be used for continuous monitoring. Several studies attempted to predict BP from non-invasively collectible signals such as photoplethysmograms (PPG) and electrocardiograms (ECG), which can be used for continuous monitoring. In this study, we explored the applicability of autoencoders in predicting BP from PPG and ECG signals. The investigation was carried out on 12,000 instances of 942 patients of the MIMIC-II dataset, and it was found that a very shallow, one-dimensional autoencoder can extract the relevant features to predict the SBP and DBP with state-of-the-art performance on a very large dataset. An independent test set from a portion of the MIMIC-II dataset provided a mean absolute error (MAE) of 2.333 and 0.713 for SBP and DBP, respectively. On an external dataset of 40 subjects, the model trained on the MIMIC-II dataset provided an MAE of 2.728 and 1.166 for SBP and DBP, respectively. For both the cases, the results met British Hypertension Society (BHS) Grade A and surpassed the studies from the current literature.

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

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

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.

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

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

Notable Papers and New Directions in Sensors, Signals, and Imaging Informatics

OBJECTIVE

To identify and highlight research papers representing noteworthy developments in signals, sensors, and imaging informatics in 2020.

METHOD

A broad literature search was conducted on PubMed and Scopus databases. We combined Medical Subject Heading (MeSH) terms and keywords to construct particular queries for sensors, signals, and image informatics. We only considered papers that have been published in journals providing at least three articles in the query response. Section editors then independently reviewed the titles and abstracts of preselected papers assessed on a three-point Likert scale. Papers were rated from 1 (do not include) to 3 (should be included) for each topical area (sensors, signals, and imaging informatics) and those with an average score of 2 or above were subsequently read and assessed again by two of the three co-editors. Finally, the top 14 papers with the highest combined scores were considered based on consensus.

RESULTS

The search for papers was executed in January 2021. After removing duplicates and conference proceedings, the query returned a set of 101, 193, and 529 papers for sensors, signals, and imaging informatics, respectively. We filtered out journals that had less than three papers in the query results, reducing the number of papers to 41, 117, and 333, respectively. From these, the co-editors identified 22 candidate papers with more than 2 Likert points on average, from which 14 candidate best papers were nominated after intensive discussion. At least five external reviewers then rated the remaining papers. The four finalist papers were found using the composite rating of all external reviewers. These best papers were approved by consensus of the International Medical Informatics Association (IMIA) Yearbook editorial board.

CONCLUSIONS

Sensors, signals, and imaging informatics is a dynamic field of intense research. The four best papers represent advanced approaches for combining, processing, modeling, and analyzing heterogeneous sensor and imaging data. The selected papers demonstrate the combination and fusion of multiple sensors and sensor networks using electrocardiogram (ECG), electroencephalogram (EEG), or photoplethysmogram (PPG) with advanced data processing, deep and machine learning techniques, and present image processing modalities beyond state-of-the-art that significantly support and further improve medical decision making.

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

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

Beat-to-Beat Continuous Blood Pressure Estimation Using Bidirectional Long Short-Term Memory Network

Continuous blood pressure (BP) monitoring is important for patients with hypertension. However, BP measurement with a cuff may be cumbersome for the patient. To overcome this limitation, various studies have suggested cuffless BP estimation models using deep learning algorithms. A generalized model should be considered to decrease the training time, and the model reproducibility should be taken into account in multi-day scenarios. In this study, a BP estimation model with a bidirectional long short-term memory network is proposed. The features are extracted from the electrocardiogram, photoplethysmogram, and ballistocardiogram. The leave-one-subject-out (LOSO) method is incorporated to generalize the model and fine-tuning is applied. The model was evaluated using one-day and multi-day tests. The proposed model achieved a mean absolute error (MAE) of 2.56 and 2.05 mmHg for the systolic and diastolic BP (SBP and DBP), respectively, in the one-day test. Moreover, the results demonstrated that the LOSO method with fine-tuning was more compatible in the multi-day test. The MAE values of the model were 5.82 and 5.24 mmHg for the SBP and DBP, respectively.