Photoplethysmography and Deep Learning: Enhancing Hypertension Risk Stratification

A two-branch framework for blood pressure estimation using photoplethysmography signals with deep learning and clinical prior physiological knowledge

Objective.This paper presents a novel dual-branch framework for estimating blood pressure (BP) using photoplethysmography (PPG) signals. The method combines deep learning with clinical prior knowledge and models different time periods (morning, afternoon, and evening) to achieve precise, cuffless BP estimation.Approach.Preprocessed single-channel PPG signals are input into two feature extraction branches. The first branch converts PPG dimensions to 2D and uses pre-trained Mobile Vision Transformer-v2 (MobileViTv2) and Visual Geometry Group19 (Vgg19) backbones to extract deep PPG features based on the different mechanisms of systolic blood pressure (SBP) and diastolic blood pressure (DBP) formation. The second branch calculates multi-dimensional feature parameters based on the relationship between PPG waveforms and factors affecting BP. We fuse the features from both branches and consider diurnal BP variations, using AutoML strategy to construct specific SBP and DBP estimation models for the different periods. The algorithm was developed on the human resting state PPG and BP dataset (HRSD) and validated on the MIMIC-IV dataset for generalization performance.Main results.The mean absolute error (MAE) for BP estimation is 6.42 mmHg SBP and 4.96 mmHg DBP in the morning, 4.84 mmHg (SBP) and 3.73 mmHg (DBP) in the afternoon, and 2.65 mmHg (SBP) and 2.56 mmHg (DBP) in the evening. Performance on the MIMIC-IV database was 4.34 mmHg (SBP) and 3.11 mmHg (DBP). The method meets the standards of the Association for the Advancement of Medical Instrumentation and achieves Grade A of the British Hypertension Society (BHS) standards.Significance. This indicates that it is an accurate and reliable non-invasive BP monitoring technology, applicable for continuous health monitoring and cardiovascular disease prevention.

rU-Net, Multi-Scale Feature Fusion and Transfer Learning: Unlocking the Potential of Cuffless Blood Pressure Monitoring With PPG and ECG

This study introduces an innovative deep-learning model for cuffless blood pressure estimation using PPG and ECG signals, demonstrating state-of-the-art performance on the largest clean dataset, PulseDB. The rU-Net architecture, a fusion of U-Net and ResNet, enhances both generalization and feature extraction accuracy. Accurate multi-scale feature capture is facilitated by short-time Fourier transform (STFT) time-frequency distributions and multi-head attention mechanisms, allowing data-driven feature selection. The inclusion of demographic parameters as supervisory information further elevates performance. On the calibration-based dataset, our model excels, achieving outstanding accuracy (SBP MAE ± std: 4.49 ± 4.86 mmHg, DBP MAE ± std: 2.69 ± 3.10 mmHg), surpassing AAMI standards and earning a BHS Grade A rating. Addressing the challenge of calibration-free data, we propose a fine-tuning-based transfer learning approach. Remarkably, with only 10% data transfer, our model attains exceptional accuracy (SBP MAE ± std: 4.14 ± 5.01 mmHg, DBP MAE ± std: 2.48 ± 2.93 mmHg). This study sets the stage for the development of highly accurate and reliable wearable cuffless blood pressure monitoring devices.

Development of a novel light-sensitive PPG model using PPG scalograms and PPG-NET learning for non-invasive hypertension monitoring

Background and objective

Photoplethysmography (PPG) signals provide a non-invasive method for monitoring cardiovascular health, including blood pressure levels, which are critical for the early detection and management of hypertension. This study leverages wavelet transformation and special purpose deep learning model, enhanced by signal processing and normalization, to classify blood pressure stages from PPG signals. The primary objective is to advance non-invasive hypertension monitoring, improving the accuracy and efficiency of these assessments.

Methods

The study employed continuous wavelet transform (CWT) to prepare PPG signals for analysis using a special purpose PPG-NET designed by applying advanced deep-learning models. PPG-NET was verified by applying several pre-trained models, including Inception, MobileNetV2, InceptionResNetV2, and others to the PPG data. Rigorously five-fold cross-validated models were conducted to obtain the models' performance to ensure robustness and repeatability of results.

Results

The PPG-NET model demonstrated superior performance, achieving a perfect accuracy of 100 % in classifying the four stages of hypertension-normal, prehypertension, stage 1, and stage 2. The evaluation metrics reported include precision, sensitivity, and specificity, with the PPG-NET model achieving 100 % across all metrics. Other models showed varying levels of accuracy, with InceptionV3 also reaching 91.5 %, while some, like VGG-19, underperformed significantly.

Conclusions

Integrating CWT and PPG-NET offers a promising avenue for enhancing non-invasive blood pressure monitoring. The PPG-NET model, in particular, showed potential for clinical application due to its high accuracy and reliability. This study showed the effectiveness of combining advanced computational techniques with traditional PPG analysis, potentially leading to more personalized and accessible hypertension management strategies.

Detection of hypertension from pharyngeal images using deep learning algorithm in primary care settings in Japan

BACKGROUND

The early detection of hypertension using simple visual images in a way that does not require physical interaction or additional devices may improve quality of care in the era of telemedicine. Pharyngeal images include vascular morphological information and may therefore be useful for identifying hypertension.

OBJECTIVES

This study sought to develop a deep learning-based artificial intelligence algorithm for identifying hypertension from pharyngeal images.

METHODS

We conducted a secondary analysis of data from a clinical trial, in which demographic information, vital signs and pharyngeal images were obtained from patients with influenza-like symptoms in multiple primary care clinics in Japan. A deep learning-based algorithm that included a multi-instance convolutional neural network was trained to detect hypertension from pharyngeal images and demographic information. The classification performance was measured by area under the receiver operating characteristic curve. Importance heatmaps of the convolutional neural network were also examined to interpret the algorithm.

RESULTS

This study included 7710 patients from 64 clinics. The training dataset comprised 6171 patients from 51 clinics (460 positive cases), and the test dataset comprised 1539 patients from 13 clinics (130 positive cases). Our algorithm achieved an area under the receiver operating characteristic curve of 0.922 (95% CI, 0.904 to 0.940), significantly improving over the baseline prediction model incorporating only demographic information, which scored 0.887 (95% CI, 0.862 to 0.911). Our algorithm had consistent classification performance across all age and sex subgroups. Importance heatmaps revealed that the algorithm focused on the posterior pharyngeal wall area, where blood vessels are mainly located.

CONCLUSIONS

The results indicate that a deep learning-based algorithm can detect hypertension with high accuracy using pharyngeal images.

Data-knowledge co-driven feature based prediction model via photoplethysmography for evaluating blood pressure

BACKGROUND

Knowledge feature (KF) with clear physiological significance of photoplethysmography are widely used in predicting blood pressure. However, KF primarily focus on local information of photoplethysmography, which may struggle to capture the overall characteristics.

METHODS

Firstly, functional data analysis (FDA) was introduced to extract two types of data feature (DF). Furthermore, data-knowledge co-driven feature (DKCF) was proposed by combining FDA and constraints of KF. Finally, random forest, ada boost, gradient boosting, support vector machine and deep neural network were adopted, to compare the abilities of KF, DFs and DKCF in predicting blood pressure with two datasets (A published dataset and a self-collected dataset).

RESULTS

Under the premise of extracting only 9 features, the average mean absolute errors (MAE) of systolic blood pressure (SBP) and diastolic blood pressure (DBP) obtained by DKCF are both the smallest in dataset 1. In dataset 2, DKCF acquires the smallest MAE in predicting SBP and obtains the second smallest MAE in predicting DBP.

CONCLUSIONS

The results demonstrate that low-dimensional DKCF of photoplethysmography is closely correlated with blood pressure, which may serve as an important indicator for health assessment.

Raw Photoplethysmography as an Enhancement for Research-Grade Wearable Activity Monitors

Wearable monitors continue to play a critical role in scientific assessments of physical activity. Recently, research-grade monitors have begun providing raw data from photoplethysmography (PPG) alongside standard raw data from inertial sensors (accelerometers and gyroscopes). Raw PPG enables granular and transparent estimation of cardiovascular parameters such as heart rate, thus presenting a valuable alternative to standard PPG methodologies (most of which rely on consumer-grade monitors that provide only coarse output from proprietary algorithms). The implications for physical activity assessment are tremendous, since it is now feasible to monitor granular and concurrent trends in both movement and cardiovascular physiology using a single noninvasive device. However, new users must also be aware of challenges and limitations that accompany the use of raw PPG data. This viewpoint paper therefore orients new users to the opportunities and challenges of raw PPG data by presenting its mechanics, pitfalls, and availability, as well as its parallels and synergies with inertial sensors. This includes discussion of specific applications to the prediction of energy expenditure, activity type, and 24-hour movement behaviors, with an emphasis on areas in which raw PPG data may help resolve known issues with inertial sensing (eg, measurement during cycling activities). We also discuss how the impact of raw PPG data can be maximized through the use of open-source tools when developing and disseminating new methods, similar to current standards for raw accelerometer and gyroscope data. Collectively, our comments show the strong potential of raw PPG data to enhance the use of research-grade wearable activity monitors in science over the coming years.

SleepBP-Net: A Time-Distributed Convolutional Network for Nocturnal Blood Pressure Estimation from Photoplethysmogram

Nocturnal blood pressure (BP) monitoring offers valuable insights into various aspects of human wellbeing, particularly cardiovascular health. Despite recent advancements in medical technology, there remains a pressing need for a non-invasive, cuffless, and less burdensome method for overnight BP measurements. A range of machine learning models have been developed to estimate daytime BP using photoplethysmography (PPG), a readily available sensor embedded in modern wearable devices. However, investigations into nocturnal BP estimation, especially concerning long-term data patterns during sleep, are still lacking. This paper investigates the estimation of nocturnal BP from overnight PPG signals collected in a clinical-grade sleep laboratory setting. To address this, we propose SleepBP-Net, a lightweight time-distributed convolutional recurrent network. This novel model leverages long-term patterns within PPG waveforms to estimate systolic and diastolic BP (SBP and DBP), considering Portapres BP measurements as a reference. Our experiments, based on leave-one-subject-out validation on 1-minute sequences of PPG, resulted in a mean absolute error (MAE) of 15.7 mmHg (SBP) and 12.1 mmHg (DBP). Model personalization improved the results to 7.8 mmHg (SBP) and 5.9 mmHg (DBP). Further enhancements were observed when extending the sequence length to 30 minutes, resulting in MAE values of 7.2 mmHg (SBP) and 5.7 mmHg (DBP). These findings underscore the significance of learning long-term temporal patterns from sleep PPG data. Additionally, we demonstrate the superiority of hybrid convolutional recurrent networks over their convolutional network counterparts. Based on our results, SleepBP-Net holds promise for unobtrusive real-world nocturnal BP estimation, particularly in scenarios where computational efficiency is crucial.

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.

Automatic identification of hypertension and assessment of its secondary effects using artificial intelligence: A systematic review (2013-2023)

Artificial Intelligence (AI) techniques are increasingly used in computer-aided diagnostic tools in medicine. These techniques can also help to identify Hypertension (HTN) in its early stage, as it is a global health issue. Automated HTN detection uses socio-demographic, clinical data, and physiological signals. Additionally, signs of secondary HTN can also be identified using various imaging modalities. This systematic review examines related work on automated HTN detection. We identify datasets, techniques, and classifiers used to develop AI models from clinical data, physiological signals, and fused data (a combination of both). Image-based models for assessing secondary HTN are also reviewed. The majority of the studies have primarily utilized single-modality approaches, such as biological signals (e.g., electrocardiography, photoplethysmography), and medical imaging (e.g., magnetic resonance angiography, ultrasound). Surprisingly, only a small portion of the studies (22 out of 122) utilized a multi-modal fusion approach combining data from different sources. Even fewer investigated integrating clinical data, physiological signals, and medical imaging to understand the intricate relationships between these factors. Future research directions are discussed that could build better healthcare systems for early HTN detection through more integrated modeling of multi-modal data sources.

Estimation of phase distortions of the photoplethysmographic signal in digital IIR filtering

Pre-processing of the photoplethysmography (PPG) signal plays an important role in the analysis of the pulse wave signal. The task of pre-processing is to remove noise from the PPG signal, as well as to transmit the signal without any distortions for further analysis. The integrity of the pulse waveform is essential since many cardiovascular parameters are calculated from it using morphological analysis. Digital filters with infinite impulse response (IIR) are widely used in the processing of PPG signals. However, such filters tend to change the pulse waveform. The aim of this work is to quantify the PPG signal distortions that occur during IIR filtering in order to select a most suitable filter and its parameters. To do this, we collected raw finger PPG signals from 20 healthy volunteers and processed them by 5 main digital IIR filters (Butterworth, Bessel, Elliptic, Chebyshev type I and type II) with varying parameters. The upper cutoff frequency varied from 2 to 10 Hz and the filter order-from 2nd to 6th. To assess distortions of the pulse waveform, we used the following indices: skewness signal quality index (SSQI), reflection index (RI) and ejection time compensated (ETc). It was found that a decrease in the upper cutoff frequency leads to damping of the dicrotic notch and a phase shift of the pulse wave signal. The minimal distortions of a PPG signal are observed when using Butterworth, Bessel and Elliptic filters of the 2nd order. Therefore, we can recommend these filters for use in applications aimed at morphological analysis of finger PPG waveforms of healthy subjects.

Assessing heart rate and blood pressure estimation from image photoplethysmography using a digital blood pressure meter

This study presents a non-contact approach to measuring heart rate and blood pressure using an image photoplethysmography (iPPG) signal, and compares the results to those from an oscillometric blood pressure meter. Facial videos of 100 subjects were recorded via a webcam under ambient lighting conditions to extract iPPG signals. The results revealed a strong correlation between the heart rate derived from iPPG and that obtained from an oscillometric blood pressure meter. In addition, a continuous wavelet transform images with a 6-s duration were used as input for a custom convolutional neural network model, providing the most accurate blood pressure estimation. The proposed method received a grade A for diastolic and grade B for systolic blood pressure based on the British Hypertension Society's criteria. It also met the standards set by the Association for the Advancement of Medical Instrumentation. This non-contact framework shows promising potential for efficient screening purposes.

Improved Hypertension Risk Assessment with Photoplethysmographic Recordings Combining Deep Learning and Calibration

Hypertension, a primary risk factor for various cardiovascular diseases, is a global health concern. Early identification and effective management of hypertensive individuals are vital for reducing associated health risks. This study explores the potential of deep learning (DL) techniques, specifically GoogLeNet, ResNet-18, and ResNet-50, for discriminating between normotensive (NTS) and hypertensive (HTS) individuals using photoplethysmographic (PPG) recordings. The research assesses the impact of calibration at different time intervals between measurements, considering intervals less than 1 h, 1-6 h, 6-24 h, and over 24 h. Results indicate that calibration is most effective when measurements are closely spaced, with an accuracy exceeding 90% in all the DL strategies tested. For calibration intervals below 1 h, ResNet-18 achieved the highest accuracy (93.32%), sensitivity (84.09%), specificity (97.30%), and F1-score (88.36%). As the time interval between calibration and test measurements increased, classification performance gradually declined. For intervals exceeding 6 h, accuracy dropped below 81% but with all models maintaining accuracy above 71% even for intervals above 24 h. This study provides valuable insights into the feasibility of using DL for hypertension risk assessment, particularly through PPG recordings. It demonstrates that closely spaced calibration measurements can lead to highly accurate classification, emphasizing the potential for real-time applications. These findings may pave the way for advanced, non-invasive, and continuous blood pressure monitoring methods that are both efficient and reliable.

Optimized deep neural network models for blood pressure classification using Fourier analysis-based time-frequency spectrogram of photoplethysmography signal

Appropriate blood pressure (BP) management through continuous monitoring and rapid diagnosis helps to take preventive care against cardiovascular diseases (CVD). As hypertension is one of the leading causes of CVDs, keeping hypertension under control by a timely screening of subjects becomes lifesaving. This work proposes estimating BP from motion artifact-affected photoplethysmography signals (PPG) by applying signal processing techniques in realtime. This paper proposes a deep neural network-based methodology to accurately classify PPG signals using a Fourier theory-based time-frequency (TF) spectrogram. This work uses the Fourier decomposition method (FDM) to transform a PPG signal into a TF spectrogram. In the proposed work, the last three layers of the pre-trained deep neural network, namely, GoogleNet, DenseNet, and AlexNet, are modified and then used to classify the PPG signal into normotension, pre-hypertension, and hypertension. The proposed framework is trained and tested using the MIMIC-III and PPG-BP databases using five-fold training and testing. Out of the three deep neural networks, the proposed framework with the DenseNet-201 network performs best, with a test accuracy of 96.5%. The proposed work uses FDM to compute the TF spectrogram to accurately separate the motion artifacts and noise components from a noise-corrupted PPG signal. Capturing more frequency components that contain more information from PPG signals makes the deep neural networks extract better and more meaningful features. Thus, training a deep neural network model with clean PPG signal features improves the generalized capability of a BP classification model when tested in realtime.

A novel CS-NET architecture based on the unification of CNN, SVM and super-resolution spectrogram to monitor and classify blood pressure using photoplethysmography

CONTEXT

Continuous blood pressure (BP) monitoring plays an important role while treating various cardiovascular diseases and hypertension. A high correlation between arterial blood pressure (ABP) and Photoplethysmogram (PPG) signal enables using a PPG signal to monitor and classify BP continuously. Control of BP in realtime is the basis for the prevention of hypertension.

PROPOSED APPROACH

This work proposes a CS-NET architecture by unifying CNN and SVM approaches to classify BP using PPG signals. The main objective of the CS-NET method is to establish an accurate and reliable algorithm for the ABP classification.

METHODOLOGY

ABP signals are labeled normal and abnormal using the hypertension criteria the American College of Cardiology (ACC)/American Heart Association (AHA) laid down. The proposed CS-NET model incorporates three critical steps in three successive stages. The first stage includes converting a preprocessed PPG signal into a time-frequency (TF) representation called a super-resolution spectrogram by superlet transform. The second stage uses a convolutional neural network (CNN) model with several hidden layers to extract morphological features from every PPG super-resolution spectrogram. The third stage uses a support vector machine (SVM) classifier to classify the PPG signal.

RESULTS

PPG signals are used to train and test the proposed model. The performance of the proposed CS-NET method is tested using MIMIC-II, MIMIC-III, and PPG-BP-figshare database in terms of accuracy and F1 score. Moreover, the CS-NET method achieves better results with high accuracy when compared with other benchmark approaches that require an electrocardiogram signal for reference.

CONCLUSIONS

The proposed model achieved an aggregate classification accuracy of 98.21% across a five-fold cross-validation technique, making it a reliable approach for BP classification in clinical settings and realtime monitoring.

Cuffless Hypertension Detection using Swarm Support Vector Machine Utilizing Photoplethysmogram and Electrocardiogram

Background

Hypertension is associated with severe complications, and its detection is important to provide early information about a hypertension event, which is essential to prevent further complications.

Objective

This study aimed to investigate a strategy for hypertension detection without a cuff using parameters of bioelectric signals, i.e., Electrocardiogram (ECG), Photoplethysmogram (PPG,) and an algorithm of Swarm-based Support Vector Machine (SSVM).

Material and Methods

This experimental study was conducted to develop a hypertension detection system. ECG and PPG bioelectrical records were collected from the Medical Information Mart for Intensive Care (MIMIC) from normal and hypertension participants and processed to find the parameters, used for the inputs of SSVM and comprised Pulse Arrival Time (PAT) and the characteristics of PPG signal derivatives. The SSVM was n Support Vector Machine (SVM) algorithm optimized using particle swarm optimization with Quantum Delta-potential-well (QDPSO). The SSVMs with different inputs were investigated to find the optimal detection performance.

Results

The proposed strategy was performed at 96% in terms of F1-score, accuracy, sensitivity, and specificity with better performance than the other methods tested and methods and also could develop a cuff-free hypertension monitoring system.

Conclusion

Hypertension using SSVM, ECG, and PPG parameters is acceptably performed. The hypertension detection had lower performance utilizing only PPG than both ECG and PPG.

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

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

Feasibility of Transfer Learning from Finger PPG to In-Ear PPG

The success of deep learning methods has enabled many modern wearable health applications, but has also highlighted the critical caveat of their extremely data hungry nature. While the widely explored wrist and finger photoplethysmography (PPG) sites are less affected, given the large available databases, this issue is prohibitive to exploring the full potential of novel recording locations such as in-ear wearables. To this end, we assess the feasibility of transfer learning from finger PPG to in-ear PPG in the context of deep learning for respiratory monitoring. This is achieved by introducing an encoder-decoder framework which is set up to extract respiratory waveforms from PPG, whereby simultaneously recorded gold standard respiratory waveforms (capnography, impedance pneumography and air flow) are used as a training reference. Next, the data augmentation and training pipeline is examined for both training on finger PPG and the subsequent fine tuning on in-ear PPG. The results indicate that, through training on two large finger PPG data sets (95 subjects) and then retraining on our own small in-ear PPG data set (6 subjects), the model achieves lower and more consistent test error for the prediction of the respiratory waveforms, compared to training on the small in-ear data set alone. This conclusively demonstrates the feasibility of transfer learning from finger PPG to in-ear PPG, leading to better generalisation across a wide range of respiratory rates.

PPGFeat: a novel MATLAB toolbox for extracting PPG fiducial points

Photoplethysmography is a non-invasive technique used for measuring several vital signs and for the identification of individuals with an increased disease risk. Its principle of work is based on detecting changes in blood volume in the microvasculature of the skin through the absorption of light. The extraction of relevant features from the photoplethysmography signal for estimating certain physiological parameters is a challenging task, where various feature extraction methods have been proposed in the literature. In this work, we present PPGFeat, a novel MATLAB toolbox supporting the analysis of raw photoplethysmography waveform data. PPGFeat allows for the application of various preprocessing techniques, such as filtering, smoothing, and removal of baseline drift; the calculation of photoplethysmography derivatives; and the implementation of algorithms for detecting and highlighting photoplethysmography fiducial points. PPGFeat includes a graphical user interface allowing users to perform various operations on photoplethysmography signals and to identify, and if required also adjust, the fiducial points. Evaluating the PPGFeat's performance in identifying the fiducial points present in the publicly available PPG-BP dataset, resulted in an overall accuracy of 99% and 3038/3066 fiducial points were correctly identified. PPGFeat significantly reduces the risk of errors in identifying inaccurate fiducial points. Thereby, it is providing a valuable new resource for researchers for the analysis of photoplethysmography signals.

Simultaneous photoplethysmography and blood flow measurements towards the estimation of blood pressure using speckle contrast optical spectroscopy

Non-invasive continuous blood pressure monitoring remains elusive. There has been extensive research using the photoplethysmographic (PPG) waveform for blood pressure estimation, but improvements in accuracy are still needed before clinical use. Here we explored the use of an emerging technique, speckle contrast optical spectroscopy (SCOS), for blood pressure estimation. SCOS provides measurements of both blood volume changes (PPG) and blood flow index (BFi) changes during the cardiac cycle, and thus provides a richer set of parameters compared to traditional PPG. SCOS measurements were taken on the finger and wrists of 13 subjects. We investigated the correlations between features extracted from both the PPG and BFi waveforms with blood pressure. Features from the BFi waveforms were more significantly correlated with blood pressure than PPG features ( R = - 0.55, p = 1.1 × 10^-4 for the top BFi feature versus R = - 0.53, p = 8.4 × 10^-4 for the top PPG feature). Importantly, we also found that features combining BFi and PPG data were highly correlated with changes in blood pressure ( R = - 0.59, p  = 1.7 × 10^-4 ). These results suggest that the incorporation of BFi measurements should be further explored as a means to improve blood pressure estimation using non-invasive optical techniques.

Photoplethysmography Driven Hypertension Identification: A Pilot Study

To prevent and diagnose hypertension early, there has been a growing demand to identify its states that align with patients. This pilot study aims to research how a non-invasive method using photoplethysmographic (PPG) signals works together with deep learning algorithms. A portable PPG acquisition device (Max30101 photonic sensor) was utilized to (1) capture PPG signals and (2) wirelessly transmit data sets. In contrast to traditional feature engineering machine learning classification schemes, this study preprocessed raw data and applied a deep learning algorithm (LSTM-Attention) directly to extract deeper correlations between these raw datasets. The Long Short-Term Memory (LSTM) model underlying a gate mechanism and memory unit enables it to handle long sequence data more effectively, avoiding gradient disappearance and possessing the ability to solve long-term dependencies. To enhance the correlation between distant sampling points, an attention mechanism was introduced to capture more data change features than a separate LSTM model. A protocol with 15 healthy volunteers and 15 hypertension patients was implemented to obtain these datasets. The processed result demonstrates that the proposed model could present satisfactory performance (accuracy: 0.991; precision: 0.989; recall: 0.993; F1-score: 0.991). The model we proposed also demonstrated superior performance compared to related studies. The outcome indicates the proposed method could effectively diagnose and identify hypertension; thus, a paradigm to cost-effectively screen hypertension could rapidly be established using wearable smart devices.

Upper-Arm Photoplethysmographic Sensor with One-Time Calibration for Long-Term Blood Pressure Monitoring

Wearable cuffless photoplethysmographic blood pressure monitors have garnered widespread attention in recent years; however, the long-term performance values of these devices are questionable. Most cuffless blood pressure monitors require initial baseline calibration and regular recalibrations with a cuffed blood pressure monitor to ensure accurate blood pressure estimation, and their estimation accuracy may vary over time if left uncalibrated. Therefore, this study assessed the accuracy and long-term performance of an upper-arm, cuffless photoplethysmographic blood pressure monitor according to the ISO 81060-2 standard. This device was based on a nonlinear machine-learning model architecture with a fine-tuning optimized method. The blood pressure measurement protocol followed a validation procedure according to the standard, with an additional four weekly blood pressure measurements over a 1-month period, to assess the long-term performance values of the upper-arm, cuffless photoplethysmographic blood pressure monitor. The results showed that the photoplethysmographic signals obtained from the upper arm had better qualities when compared with those measured from the wrist. When compared with the cuffed blood pressure monitor, the means ± standard deviations of the difference in BP at week 1 (baseline) were -1.36 ± 7.24 and -2.11 ± 5.71 mmHg for systolic and diastolic blood pressure, respectively, which met the first criterion of ≤5 ± ≤8.0 mmHg and met the second criterion of a systolic blood pressure ≤ 6.89 mmHg and a diastolic blood pressure ≤ 6.84 mmHg. The differences in the uncalibrated blood pressure values between the test and reference blood pressure monitors measured from week 2 to week 5 remained stable and met both criteria 1 and 2 of the ISO 81060-2 standard. The upper-arm, cuffless photoplethysmographic blood pressure monitor in this study generated high-quality photoplethysmographic signals with satisfactory accuracy at both initial calibration and 1-month follow-ups. This device could be a convenient and practical tool to continuously measure blood pressure over long periods of time.

Blood pressure stratification using photoplethysmography and light gradient boosting machine

Introduction: Globally, hypertension (HT) is a substantial risk factor for cardiovascular disease and mortality; hence, rapid identification and treatment of HT is crucial. In this study, we tested the light gradient boosting machine (LightGBM) machine learning method for blood pressure stratification based on photoplethysmography (PPG), which is used in most wearable devices. Methods: We used 121 records of PPG and arterial blood pressure (ABP) signals from the Medical Information Mart for Intensive Care III public database. PPG, velocity plethysmography, and acceleration plethysmography were used to estimate blood pressure; the ABP signals were used to determine the blood pressure stratification categories. Seven feature sets were established and used to train the Optuna-tuned LightGBM model. Three trials compared normotension (NT) vs. prehypertension (PHT), NT vs. HT, and NT + PHT vs. HT. Results: The F1 scores for these three classification trials were 90.18%, 97.51%, and 92.77%, respectively. The results showed that combining multiple features from PPG and its derivative led to a more accurate classification of HT classes than using features from only the PPG signal. Discussion: The proposed method showed high accuracy in stratifying HT risks, providing a noninvasive, rapid, and robust method for the early detection of HT, with promising applications in the field of wearable cuffless blood pressure measurement.

Advancement in the Cuffless and Noninvasive Measurement of Blood Pressure: A Review of the Literature and Open Challenges

Hypertension is a chronic condition that is one of the prominent reasons behind cardiovascular disease, brain stroke, and organ failure. Left unnoticed and untreated, the deterioration in a health condition could even result in mortality. If it can be detected early, with proper treatment, undesirable outcomes can be avoided. Until now, the gold standard is the invasive way of measuring blood pressure (BP) using a catheter. Additionally, the cuff-based and noninvasive methods are too cumbersome or inconvenient for frequent measurement of BP. With the advancement of sensor technology, signal processing techniques, and machine learning algorithms, researchers are trying to find the perfect relationships between biomedical signals and changes in BP. This paper is a literature review of the studies conducted on the cuffless noninvasive measurement of BP using biomedical signals. Relevant articles were selected using specific criteria, then traditional techniques for BP measurement were discussed along with a motivation for cuffless measurement use of biomedical signals and machine learning algorithms. The review focused on the progression of different noninvasive cuffless techniques rather than comparing performance among different studies. The literature survey concluded that the use of deep learning proved to be the most accurate among all the cuffless measurement techniques. On the other side, this accuracy has several disadvantages, such as lack of interpretability, computationally extensive, standard validation protocol, and lack of collaboration with health professionals. Additionally, the continuing work by researchers is progressing with a potential solution for these challenges. Finally, future research directions have been provided to encounter the challenges.

Variation in blood pressure and heart rate of radiological technologists in worktime tracked by a wearable device: A preliminary study

The aim of this preliminary study was to measure the systolic BP (SBP) and diastolic BP (DBP) and heart rate (HR) of radiological technologists by WD, and evaluate variation among individuals by worktime, day of the week, job, and workplace. Measurements were obtained using a wristwatch-type WD with optical measurement technology that can measure SBP and DBP every 10 minutes and HR every 30 minutes. SBP, DBP, and HR data obtained at baseline and during work time were combined with the hours of work, day of the week, job, and workplace recorded by the participants in 8 consecutive weeks. We calculated the mean, the ratio to baseline and coefficient of variation [CV(%)] for SBP, DBP, and HR. SBP, DBP, and HR values were significantly higher during work hours than at baseline (p<0.03). The ratio to baseline values ranged from 1.02 to 1.26 for SBP and from 1.07 to 1.30 for DBP. The ratio to baseline for SBP and DBP showed CV(%) of approximately 10% according to the day of the week and over the study period. For HR, ratio to baseline ranged from 0.95 to 1.29. The ratio of mean BP to baseline was >1.2 at the time of starting work, middle and after lunch, and at 14:00. The ratio to baseline of SBP were 1.2 or more for irradiation, equipment accuracy control, registration of patient data, dose verification and conference time, and were also working in CT examination room, treatment planning room, linac room, and the office. CV(%) of BP and HR were generally stable for all workplaces. WD measurements of SBP, DBP, and HR were higher during working hours than at baseline and varied by the individuals, work time, job, and workplace. This method may enable evaluation of unconscious workload in individuals.

A Novel Pulse-Taking Device for Persian Medicine Based on Convolutional Neural Networks

Background

In Persian medicine (PM), measuring the wrist pulse is one of the main methods for determining a person's health status and temperament. One problem that can arise is the dependence of the diagnosis on the physician's interpretation of pulse wave features. Perhaps, this is one reason why this method has yet to be combined with modern medical methods. This paper addresses this concern and outlines a system for measuring pulse signals based on PM.

Methods

A system that uses data from a customized device that logs the pulse wave on the wrist was designed and clinically implemented based on PM. Seven convolutional neural networks (CNNs) have been used for classification.

Results

The pulse wave features of 34 participants were assessed by a specialist based on PM principles. Pulse taking was done on the wrist in the supine position (named Malmas in PM) under the supervision of the physician. Seven CNNs were implemented for each participant's pulse characteristic (pace, rate, vessel elasticity, strength, width, length, and height) assessment, and then, each participant was classified into three classes.

Conclusion

It appears that the design and construction of a customized device combined with the deep learning algorithm can measure the pulse wave features according to PM and it can increase the reliability and repeatability of the diagnostic results based on PM.

Classifying sepsis from photoplethysmography

Sepsis is a life-threatening organ dysfunction. It is caused by a dysregulated immune response to an infection and is one of the leading causes of death in the intensive care unit (ICU). Early detection and treatment of sepsis can increase the survival rate of patients. The use of devices such as the photoplethysmograph could allow the early evaluation in addition to continuous monitoring of septic patients. The aim of this study was to verify the possibility of detecting sepsis in patients from whom the photoplethysmographic signal was acquired via a pulse oximeter. In this work, we developed a deep learning-based model for sepsis identification. The model takes a single input, the photoplethysmographic signal acquired by pulse oximeter, and performs a binary classification between septic and nonseptic samples. To develop the method, we used MIMIC-III database, which contains data from ICU patients. Specifically, the selected dataset includes 85 septic subjects and 101 control subjects. The PPG signals acquired from these patients were segmented, processed and used as input for the developed model with the aim of identifying sepsis. The proposed method achieved an accuracy of 76.37% with a sensitivity of 70.95% and a specificity of 81.04% on the test set. As regards the ROC curve, the Area Under Curve reached a value of 0.842. The results of this study indicate how the plethysmographic signal can be used as a warning sign for the early detection of sepsis with the aim of reducing the time for diagnosis and therapeutic intervention. Furthermore, the proposed method is suitable for integration in continuous patient monitoring.

Blood Pressure Measurement: From Cuff-Based to Contactless Monitoring

Blood pressure (BP) determines whether a person has hypertension and offers implications as to whether he or she could be affected by cardiovascular disease. Cuff-based sphygmomanometers have traditionally provided both accuracy and reliability, but they require bulky equipment and relevant skills to obtain precise measurements. BP measurement from photoplethysmography (PPG) signals has become a promising alternative for convenient and unobtrusive BP monitoring. Moreover, the recent developments in remote photoplethysmography (rPPG) algorithms have enabled new innovations for contactless BP measurement. This paper illustrates the evolution of BP measurement techniques from the biophysical theory, through the development of contact-based BP measurement from PPG signals, and to the modern innovations of contactless BP measurement from rPPG signals. We consolidate knowledge from a diverse background of academic research to highlight the importance of multi-feature analysis for improving measurement accuracy. We conclude with the ongoing challenges, opportunities, and possible future directions in this emerging field of research.

A Review of Noninvasive Methodologies to Estimate the Blood Pressure Waveform

Accurate estimation of blood pressure (BP) waveforms is critical for ensuring the safety and proper care of patients in intensive care units (ICUs) and for intraoperative hemodynamic monitoring. Normal cuff-based BP measurements can only provide systolic blood pressure (SBP) and diastolic blood pressure (DBP). Alternatively, the BP waveform can be used to estimate a variety of other physiological parameters and provides additional information about the patient's health. As a result, various techniques are being proposed for accurately estimating the BP waveforms. The purpose of this review is to summarize the current state of knowledge regarding the BP waveform, three methodologies (pressure-based, ultrasound-based, and deep-learning-based) used in noninvasive BP waveform estimation research and the feasibility of employing these strategies at home as well as in ICUs. Additionally, this article will discuss the physical concepts underlying both invasive and noninvasive BP waveform measurements. We will review historical BP waveform measurements, standard clinical procedures, and more recent innovations in noninvasive BP waveform monitoring. Although the technique has not been validated, it is expected that precise, noninvasive BP waveform estimation will be available in the near future due to its enormous potential.

Research on Digital Technology Use in Cardiology: Bibliometric Analysis

Background

Digital technology uses in cardiology have become a popular research focus in recent years. However, there has been no published bibliometric report that analyzed the corresponding academic literature in order to derive key publishing trends and characteristics of this scientific area.

Objective

We used a bibliometric approach to identify and analyze the academic literature on digital technology uses in cardiology, and to unveil popular research topics, key authors, institutions, countries, and journals. We further captured the cardiovascular conditions and diagnostic tools most commonly investigated within this field.

Methods

The Web of Science electronic database was queried to identify relevant papers on digital technology uses in cardiology. Publication and citation data were acquired directly from the database. Complete bibliographic data were exported to VOSviewer, a dedicated bibliometric software package, and related to the semantic content of titles, abstracts, and keywords. A term map was constructed for findings visualization.

Results

The analysis was based on data from 12,529 papers. Of the top 5 most productive institutions, 4 were based in the United States. The United States was the most productive country (4224/12,529, 33.7%), followed by United Kingdom (1136/12,529, 9.1%), Germany (1067/12,529, 8.5%), China (682/12,529, 5.4%), and Italy (622/12,529, 5.0%). Cardiovascular diseases that had been frequently investigated included hypertension (152/12,529, 1.2%), atrial fibrillation (122/12,529, 1.0%), atherosclerosis (116/12,529, 0.9%), heart failure (106/12,529, 0.8%), and arterial stiffness (80/12,529, 0.6%). Recurring modalities were electrocardiography (170/12,529, 1.4%), angiography (127/12,529, 1.0%), echocardiography (127/12,529, 1.0%), digital subtraction angiography (111/12,529, 0.9%), and photoplethysmography (80/12,529, 0.6%). For a literature subset on smartphone apps and wearable devices, the Journal of Medical Internet Research (20/632, 3.2%) and other JMIR portfolio journals (51/632, 8.0%) were the major publishing venues.

Conclusions

Digital technology uses in cardiology target physicians, patients, and the general public. Their functions range from assisting diagnosis, recording cardiovascular parameters, and patient education, to teaching laypersons about cardiopulmonary resuscitation. This field already has had a great impact in health care, and we anticipate continued growth.

The Relevance of Calibration in Machine Learning-Based Hypertension Risk Assessment Combining Photoplethysmography and Electrocardiography

The detection of hypertension (HT) is of great importance for the early diagnosis of cardiovascular diseases (CVDs), as subjects with high blood pressure (BP) are asymptomatic until advanced stages of the disease. The present study proposes a classification model to discriminate between normotensive (NTS) and hypertensive (HTS) subjects employing electrocardiographic (ECG) and photoplethysmographic (PPG) recordings as an alternative to traditional cuff-based methods. A total of 913 ECG, PPG and BP recordings from 69 subjects were analyzed. Then, signal preprocessing, fiducial points extraction and feature selection were performed, providing 17 discriminatory features, such as pulse arrival and transit times, that fed machine-learning-based classifiers. The main innovation proposed in this research uncovers the relevance of previous calibration to obtain accurate HT risk assessment. This aspect has been assessed using both close and distant time test measurements with respect to calibration. The k-nearest neighbors-classifier provided the best outcomes with an accuracy for new subjects before calibration of 51.48%. The inclusion of just one calibration measurement into the model improved classification accuracy by 30%, reaching gradually more than 96% with more than six calibration measurements. Accuracy decreased with distance to calibration, but remained outstanding even days after calibration. Thus, the use of PPG and ECG recordings combined with previous subject calibration can significantly improve discrimination between NTS and HTS individuals. This strategy could be implemented in wearable devices for HT risk assessment as well as to prevent CVDs.

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.

Application of photoplethysmography signals for healthcare systems: An in-depth review

BACKGROUND AND OBJECTIVES

Photoplethysmography (PPG) is a device that measures the amount of light absorbed by the blood vessel, blood, and tissues, which can, in turn, translate into various measurements such as the variation in blood flow volume, heart rate variability, blood pressure, etc. Hence, PPG signals can produce a wide variety of biological information that can be useful for the detection and diagnosis of various health problems. In this review, we are interested in the possible health disorders that can be detected using PPG signals.

METHODS

We applied PRISMA guidelines to systematically search various journal databases and identified 43 PPG studies that fit the criteria of this review.

RESULTS

Twenty-five health issues were identified from these studies that were classified into six categories: cardiac, blood pressure, sleep health, mental health, diabetes, and miscellaneous. Various routes were employed in these PPG studies to perform the diagnosis: machine learning, deep learning, and statistical routes. The studies were reviewed and summarized.

CONCLUSIONS

We identified limitations such as poor standardization of sampling frequencies and lack of publicly available PPG databases. We urge that future work should consider creating more publicly available databases so that a wide spectrum of health problems can be covered. We also want to promote the use of PPG signals as a potential precision medicine tool in both ambulatory and hospital settings.

Assessing hemodynamics from the photoplethysmogram to gain insights into vascular age: a review from VascAgeNet

The photoplethysmogram (PPG) signal is widely measured by clinical and consumer devices, and it is emerging as a potential tool for assessing vascular age. The shape and timing of the PPG pulse wave are both influenced by normal vascular aging, changes in arterial stiffness and blood pressure, and atherosclerosis. This review summarizes research into assessing vascular age from the PPG. Three categories of approaches are described: 1) those which use a single PPG signal (based on pulse wave analysis), 2) those which use multiple PPG signals (such as pulse transit time measurement), and 3) those which use PPG and other signals (such as pulse arrival time measurement). Evidence is then presented on the performance, repeatability and reproducibility, and clinical utility of PPG-derived parameters of vascular age. Finally, the review outlines key directions for future research to realize the full potential of photoplethysmography for assessing vascular age.

Automated Detection of Hypertension Using Continuous Wavelet Transform and a Deep Neural Network with Ballistocardiography Signals

Managing hypertension (HPT) remains a significant challenge for humanity. Despite advancements in blood pressure (BP)-measuring systems and the accessibility of effective and safe anti-hypertensive medicines, HPT is a major public health concern. Headaches, dizziness and fainting are common symptoms of HPT. In HPT patients, normalcy may be observed at one instant and abnormality may prevail during a long duration of 24 h ambulatory BP. This may cause difficulty in identifying patients with HPT, and hence there is a possibility that individuals may be untreated or administered insufficiently. Most importantly, uncontrolled HPT can lead to severe complications (stroke, heart attack, kidney disease, and heart failure), mainly ignoring the signs in nascent stages. HPT in the beginning stages may not present distinct symptoms and may be difficult to diagnose from standard physiological signals. Hence, ballistocardiography (BCG) signal was used in this study to detect HPT automatically. The processed signals from BCG were converted into scalogram images using a continuous wavelet transform (CWT) and were then fed into a 2-D convolutional neural network model (2D-CNN). The model was trained to learn and recognize BCG patterns of healthy controls (HC) and HPT classes. Our proposed model obtained a high classification accuracy of 86.14% with a ten-fold cross-validation (CV) strategy. Hence, this is the first use of a 2D-CNN model (deep-learning algorithm) to detect HPT employing BCG signals.

A Data-Driven Model with Feedback Calibration Embedded Blood Pressure Estimator Using Reflective Photoplethysmography

Ambulatory blood pressure (BP) monitoring (ABPM) is vital for screening cardiovascular activity. The American College of Cardiology/American Heart Association guideline for the prevention, detection, evaluation, and management of BP in adults recommends measuring BP outside the office setting using daytime ABPM. The recommendation to use night-day BP measurements to confirm hypertension is consistent with the recommendation of several other guidelines. In recent studies, ABPM was used to measure BP at regular intervals, and it reduces the effect of the environment on BP. Out-of-office measurements are highly recommended by almost all hypertension organizations. However, traditional ABPM devices based on the oscillometric technique usually interrupt sleep. For all-day ABPM purposes, a photoplethysmography (PPG)-based wrist-type device has been developed as a convenient tool. This optical, noninvasive device estimates BP using morphological characteristics from PPG waveforms. As measurement can be affected by multiple variables, calibration is necessary to ensure that the calculated BP values are accurate. However, few studies focused on adaptive calibration. A novel adaptive calibration model, which is data-driven and embedded in a wearable device, was proposed. The features from a 15 s PPG waveform and personal information were input for estimation of BP values and our data-driven calibration model. The model had a feedback calibration process using the exponential Gaussian process regression method to calibrate BP values and avoid inter- and intra-subject variability, ensuring accuracy in long-term ABPM. The estimation error of BP (ΔBP = actual BP-estimated BP) of systolic BP was -0.1776 ± 4.7361 mmHg; ≤15 mmHg, 99.225%, and of diastolic BP was -0.3846 ± 6.3688 mmHg; ≤15 mmHg, 98.191%. The success rate was improved, and the results corresponded to the Association for the Advancement of Medical Instrumentation standard and British Hypertension Society Grading criteria for medical regulation. Using machine learning with a feedback calibration model could be used to assess ABPM for clinical purposes.

Cuffless Blood Pressure Measurement Using Linear and Nonlinear Optimized Feature Selection

The cuffless blood pressure (BP) measurement allows for frequent measurement without discomfort to the patient compared to the cuff inflation measurement. With the availability of a large dataset containing physiological waveforms, now it is possible to use them through different learning algorithms to produce a relationship with changes in BP. In this paper, a novel cuffless noninvasive blood pressure measurement technique has been proposed using optimized features from electrocardiogram and photoplethysmography based on multivariate symmetric uncertainty (MSU). The technique is an improvement over other contemporary methods due to the inclusion of feature optimization depending on both linear and nonlinear relationships with the change of blood pressure. MSU has been used as a selection criterion with algorithms such as the fast correlation and ReliefF algorithms followed by the penalty-based regression technique to make sure the features have maximum relevance as well as minimum redundancy. The result from the technique was compared with the performance of similar techniques using the MIMIC-II dataset. After training and testing, the root mean square error (RMSE) comes as 5.28 mmHg for systolic BP and 5.98 mmHg for diastolic BP. In addition, in terms of mean absolute error, the result improved to 4.27 mmHg for SBP and 5.01 for DBP compared to recent cuffless BP measurement techniques which have used substantially large datasets and feature optimization. According to the British Hypertension Society Standard (BHS), our proposed technique achieved at least grade B in all cumulative criteria for cuffless BP measurement.

Artificial Intelligence for Detection of Cardiovascular-Related Diseases from Wearable Devices: A Systematic Review and Meta-Analysis

PURPOSE

Several artificial intelligence (AI) models for the detection and prediction of cardiovascular-related diseases, including arrhythmias, diabetes, and sleep apnea, have been reported. This systematic review and meta-analysis aimed to identify AI models developed for or applicable to wearable and mobile devices for diverse cardiovascular-related diseases.

MATERIALS AND METHODS

The searched databases included Medline, Embase, and Cochrane Library. For AI models for atrial fibrillation (AF) detection, a meta-analysis of diagnostic accuracy was performed to summarize sensitivity and specificity.

RESULTS

A total of 102 studies were included in the qualitative review. There were AI models for the detection of arrythmia (n=62), followed by sleep apnea (n=11), peripheral vascular diseases (n=6), diabetes mellitus (n=5), hyper/hypotension (n=5), valvular heart disease (n=4), heart failure (n=3), myocardial infarction and cardiac arrest (n=2), and others (n=4). For quantitative analysis of 26 studies reporting AI models for AF detection, meta-analyzed sensitivity was 94.80% and specificity was 96.96%. Deep neural networks showed superior performance [meta-analyzed area under receiver operating characteristics curve (AUROC) of 0.981] compared to conventional machine learning algorithms (meta-analyzed AUROC of 0.961). However, AI models tested with proprietary dataset (meta-analyzed AUROC of 0.972) or data acquired from wearable devices (meta-analyzed AUROC of 0.977) showed inferior performance than those with public dataset (meta-analyzed AUROC of 0.986) or data from in-hospital devices (meta-analyzed AUROC of 0.983).

CONCLUSION

This review found that AI models for diverse cardiovascular-related diseases are being developed, and that they are gradually developing into a form that is suitable for wearable and mobile devices.

Hypertension-Related Drug Activity Identification Based on Novel Ensemble Method

Hypertension is a chronic disease and major risk factor for cardiovascular and cerebrovascular diseases that often leads to damage to target organs. The prevention and treatment of hypertension is crucially important for human health. In this paper, a novel ensemble method based on a flexible neural tree (FNT) is proposed to identify hypertension-related active compounds. In the ensemble method, the base classifiers are Multi-Grained Cascade Forest (gcForest), support vector machines (SVM), random forest (RF), AdaBoost, decision tree (DT), Gradient Boosting Decision Tree (GBDT), KNN, logical regression, and naïve Bayes (NB). The classification results of nine classifiers are utilized as the input vector of FNT, which is utilized as a nonlinear ensemble method to identify hypertension-related drug compounds. The experiment data are extracted from hypertension-unrelated and hypertension-related compounds collected from the up-to-date literature. The results reveal that our proposed ensemble method performs better than other single classifiers in terms of ROC curve, AUC, TPR, FRP, Precision, Specificity, and F1. Our proposed method is also compared with the averaged and voting ensemble methods. The results reveal that our method could identify hypertension-related compounds more accurately than two classical ensemble methods.

Multi-model fusion of classifiers for blood pressure estimation

Prehypertension is a new risky disease defined in the seventh report issued by the Joint National Commission. Hence, detecting prehypertension in time plays a very important role in protecting human lives. This study proposes a method for categorising blood pressure values into two classes, namely the class of healthy blood pressure values and the class of prehypertension blood pressure values, as well as estimating the blood pressure values continuously only by employing photoplethysmograms. First, the denoising of photoplethysmograms is performed via a discrete cosine transform approach. Then, the features of the photoplethysmograms in both the time domain and the frequency domain are extracted. Next, the feature vectors are categorised into the two classes of blood pressure values by a multi‐model fusion of the classifiers. Here, the support vector machine, the random forest and the K‐nearest neighbour classifier are employed for performing the fusion. There are two types of blood pressure values. They are the systolic blood pressure values and the diastolic blood pressure values. For each class and each type of blood pressure values, support vector regression is used to estimate the blood pressure values. Since different classes and different types of blood pressure values are considered separately, the proposed method achieves an accurate estimation. The computed numerical simulation results show that the proposed method based on the multi‐model fusion of the classifiers achieves both higher classification accuracy and higher regression accuracy than the individual classification methods.

Classification of blood pressure in critically ill patients using photoplethysmography and machine learning

OBJECTIVE

The aim of this study was to evaluate the capability of features extracted from photoplethysmography (PPG) based Pulse Rate Variability (PRV) to classify hypertensive, normotensive and hypotensive events, and to estimate mean arterial, systolic and diastolic blood pressure in critically ill patients.

METHODS

Time-domain, frequency-domain and non-linear indices from PRV were extracted from 5-min and 1-min segments obtained from PPG signals. These features were filtered using machine learning algorithms in order to obtain the optimal combination for the classification of hypertensive, hypotensive and normotensive events, and for the estimation of blood pressure.

RESULTS

5-min segments allowed for an improved performance in both classification and estimation tasks. Classification of blood pressure states showed around 70% accuracy and around 75% specificity. The sensitivity, precision and F1 scores were around 50%. In estimating mean arterial, systolic, and diastolic blood pressure, mean absolute errors as low as 2.55 ± 0.78 mmHg, 4.74 ± 2.33 mmHg, and 1.78 ± 0.14 mmHg were obtained, respectively. Bland-Altman analysis and Wilcoxon rank sum tests showed good agreement between real and estimated values, especially for mean and diastolic arterial blood pressures.

CONCLUSION

PRV-based features could be used for the classification of blood pressure states and the estimation of blood pressure values, although including additional features from the PPG waveform could improve the results.

SIGNIFICANCE

PRV contains information related to blood pressure, which may aid in the continuous, noninvasive, non-intrusive estimation of blood pressure and detection of hypertensive and hypotensive events in critically ill subjects.

Improving the Accuracy in Classification of Blood Pressure from Photoplethysmography Using Continuous Wavelet Transform and Deep Learning

BACKGROUND

Continuous wavelet transform (CWT) based scalogram can be used for photoplethysmography (PPG) signal transformation to classify blood pressure (BP) with deep learning. We aimed to investigate the determinants that can improve the accuracy of BP classification based on PPG and deep learning and establish a better algorithm for the prediction.

METHODS

The dataset from PhysioNet was accessed to extract raw PPG signals for testing and its corresponding BPs as category labels. The BP category of normal or abnormal followed the criteria of the 2017 American College of Cardiology/American Heart Association (ACC/AHA) Hypertension Guidelines. The PPG signals were transformed into 224 ∗ 224 ∗ 3-pixel scalogram via different CWTs and segment units. All of them are fed into different convolutional neural networks (CNN) for training and validation. The receiver-operating characteristic and loss and accuracy curves were used to evaluate and compare the performance of different methods.

RESULTS

Both wavelet type and segment length could affect the accuracy, and Cgau1 wavelet and segment-300 revealed the best performance (accuracy 90%) without obvious overfitting. This method performed better than previously reported MATLAB Morse wavelet transformed scalogram on both of our proposed CNN and CNN-GoogLeNet.

CONCLUSIONS

We have established a new algorithm with high accuracy to predict BP classification from PPG via matching of CWT type and segment length, which is a promising solution for rapid prediction of BP classification from real-time processing of PPG signal on a wearable device.

Deep learning-based photoplethysmography classification for peripheral arterial disease detection: a proof-of-concept study

Objective.A proof-of-concept study to assess the potential of a deep learning (DL) based photoplethysmography PPG ('DLPPG') classification method to detect peripheral arterial disease (PAD) using toe PPG signals.Approach.PPG spectrogram images derived from our previously published multi-site PPG datasets (214 participants; 31.3% legs with PAD by ankle brachial pressure index (ABPI)) were input into a pretrained 8-layer (five convolutional layers + three fully connected layers) AlexNet as tailored to the 2-class problem with transfer learning to fine tune the convolutional neural network (CNN).k-fold random cross validation (CV) was performed (fork = 5 andk = 10), with each evaluated over k training/validation runs. Overall test sensitivity, specificity, accuracy, and Cohen's Kappa statistic with 95% confidence interval ranges were calculated and compared, as well as sensitivities in detecting mild-moderate (0.5 ≤ ABPI < 0.9) and major (ABPI < 0.5) levels of PAD.Main results.CV with eitherk = 5 or 10 folds gave similar diagnostic performances. The overall test sensitivity was 86.6%, specificity 90.2% and accuracy 88.9% (Kappa: 0.76 [0.70-0.82]) (atk= 5). The sensitivity to mild-moderate disease was 83.0% (75.5%-88.9%) and to major disease was 100.0% (90.5%-100.0%).Significance.Substantial agreements have been demonstrated between the DL-based PPG classification technique and the ABPI PAD diagnostic reference. This novel automatic approach, requiring minimal pre-processing of the pulse waveforms before PPG trace classification, could offer significant benefits for the diagnosis of PAD in a variety of clinical settings where low-cost, portable and easy-to-use diagnostics are desirable.

Personalized Blood Pressure Estimation Using Photoplethysmography: A Transfer Learning Approach

In this paper, we present a personalized deep learning approach to estimate blood pressure (BP) using the photoplethysmogram (PPG) signal. We propose a hybrid neural network architecture consisting of convolutional, recurrent, and fully connected layers that operates directly on the raw PPG time series and provides BP estimation every 5 seconds. To address the problem of limited personal PPG and BP data for individuals, we propose a transfer learning technique that personalizes specific layers of a network pre-trained with abundant data from other patients. We use the MIMIC III database which contains PPG and continuous BP data measured invasively via an arterial catheter to develop and analyze our approach. Our transfer learning technique, namely BP-CRNN-Transfer, achieves a mean absolute error (MAE) of 3.52 and 2.20 mmHg for SBP and DBP estimation, respectively, outperforming existing methods. Our approach satisfies both the BHS and AAMI blood pressure measurement standards for SBP and DBP. Moreover, our results demonstrate that as little as 50 data samples per person are required to train accurate personalized models. We carry out Bland-Altman and correlation analysis to compare our method to the invasive arterial catheter, which is the gold-standard BP measurement method.