The use of multi-site photoplethysmography (PPG) as a screening tool for coronary arterial disease and atherosclerosis

Digital Transformation in the Diagnostics and Therapy of Cardiovascular Diseases: Comprehensive Literature Review

BACKGROUND

The digital transformation of our health care system has experienced a clear shift in the last few years due to political, medical, and technical innovations and reorganization. In particular, the cardiovascular field has undergone a significant change, with new broad perspectives in terms of optimized treatment strategies for patients nowadays.

OBJECTIVE

After a short historical introduction, this comprehensive literature review aimed to provide a detailed overview of the scientific evidence regarding digitalization in the diagnostics and therapy of cardiovascular diseases (CVDs).

METHODS

We performed an extensive literature search of the PubMed database and included all related articles that were published as of March 2022. Of the 3021 studies identified, 1639 (54.25%) studies were selected for a structured analysis and presentation (original articles: n=1273, 77.67%; reviews or comments: n=366, 22.33%). In addition to studies on CVDs in general, 829 studies could be assigned to a specific CVD with a diagnostic and therapeutic approach. For data presentation, all 829 publications were grouped into 6 categories of CVDs.

RESULTS

Evidence-based innovations in the cardiovascular field cover a wide medical spectrum, starting from the diagnosis of congenital heart diseases or arrhythmias and overoptimized workflows in the emergency care setting of acute myocardial infarction to telemedical care for patients having chronic diseases such as heart failure, coronary artery disease, or hypertension. The use of smartphones and wearables as well as the integration of artificial intelligence provides important tools for location-independent medical care and the prevention of adverse events.

CONCLUSIONS

Digital transformation has opened up multiple new perspectives in the cardiovascular field, with rapidly expanding scientific evidence. Beyond important improvements in terms of patient care, these innovations are also capable of reducing costs for our health care system. In the next few years, digital transformation will continue to revolutionize the field of cardiovascular medicine and broaden our medical and scientific horizons.

Pulse-wave velocity assessments derived from a simple photoplethysmography device: Agreement with a referent device

Objective

Pulse-wave velocity (PWV), a common measure of arterial stiffness, can be measured continuously and across multiple body sites using photoplethysmography (PPG). The objective was to determine whether a simple photoplethysmography PPG PWV method agrees with a referent device.

Approach

Photoplethysmography heart-finger PWV (hfPWV) and heart-toe PWV (htPWV) were compared to oscillometric carotid-wrist PWV (cwPWV) and carotid-ankle PWV (caPWV) referent measurements, respectively. In 30 adults (24.6 ± 4.8 years, body mass index 25.2 ± 5.9 kg/m², 18 female), three measurements were made: two supine baseline measurements (Base 1, Base 2) and one measurement (Tilt) 5 min after a modified head-up tilt test (mHUTT). Overall agreement and repeated measures agreement (change in PPG PWV from Base to Tilt vs. change in referent PWV from Base to Tilt) were calculated using linear mixed models. Agreement estimates were expressed as intra-class correlation coefficients (ICC).

Main results

For hfPWV there was strong overall agreement (ICC: 0.77, 95%CI: 0.67-0.85), but negligible and non-significant repeated measures agreement (ICC: 0.10, 95%CI: -0.18 to 0.36). For htPWV, there was moderate overall agreement (ICC:0.50, 95%CI: 0.31-0.65) and strong repeated measures agreement (ICC: 0.81, 95%CI: 0.69-0.89).

Significance

Photoplethysmography can continuously measure PWV at multiple arterial segments with moderate-strong overall agreement. While further work with upper-limb PPG PWV is needed, PPG can adequately capture acute changes in lower-limb PWV.

Advances in Cuffless Continuous Blood Pressure Monitoring Technology Based on PPG Signals

Objective. To review the progress of research on photoplethysmography- (PPG-) based cuffless continuous blood pressure monitoring technologies and prospect the challenges that need to be addressed in the future. Methods. Using Web of Science and PubMed as search engines, the literature on cuffless continuous blood pressure studies using PPG signals in the recent five years were searched. Results. Based on the retrieved literature, this paper describes the available open datasets, commonly used signal preprocessing methods, and model evaluation criteria. Early researches employed multisite PPG signals to calculate pulse wave velocity or time and predicted blood pressure by a simple linear equation. Later, extensive researches were dedicated to mine the features of PPG signals related to blood pressure and regressed blood pressure by machine learning models. Most recently, many researches have emerged to experiment with complex deep learning models for blood pressure prediction with the raw PPG signal as input. Conclusion. This paper summarized the methods in the retrieved literature, provided insight into the artificial intelligence algorithms employed in the literature, and concluded with a discussion of the challenges and opportunities for the development of cuffless continuous blood pressure monitoring technologies.

Estimation of carotid-femoral pulse wave velocity from finger photoplethysmography signal

Objective. This project compared a new method to estimate the carotid-femoral pulse wave velocity (cf-PWV) to the gold-standard cf-PWV technique.Approach. The cf-PWV was estimated from the pulse transit time (FPS-PTT) calculated by processing the finger photoplethysmographic signal of Finapres (FPS) and subject's height only (brief mode) as well as along with other variables (age, heart rate, arterial pressure, weight; complete mode). Doppler ultrasound cf-PWVs and FPS-PTTs were measured in 90 participants equally divided into 3 groups (18-30; 31-59; 60-79 years). Predictions were performed using multiple linear regressions (MLR) and with the best regression model identified by using MATLAB Regression Learner App. A validation set approach (60 training datasets, 30 testing datasets; VSA) and leave-one-out cross-validation (LOOCV) were used.Main results. With MLR, the discrepancies were: 0.01 ± 1.21 m s^-1(VSA) and 0.001 ± 1.11 m s^-1(LOOCV) in brief mode; -0.02 ± 0.83 m s^-1(VSA) and 0.001 ± 0.84 m s^-1(LOOCV) in complete mode. Using a linear support vector machine model (SVM) in brief mode, the discrepancies were: 0.01 ± 1.19 m s^-1(VSA) and -0.01 ± 1.06 m s^-1(LOOCV). Using an Exponential Gaussian process regression model (GPR) in complete mode, the discrepancies were: -0.03 ± 0.79 m s^-1(VSA) and 0.01 ± 0.75 m s^-1(LOOCV).Significance. The cf-PWV can be estimated by processing the FPS-PTT and subjects' height only, but the inclusion of other variables improves the prediction performance. Predictions through MLR qualify as acceptable in both brief and complete modes. Predictions via linear SVM in brief mode improve but still qualify as acceptable. Interestingly, predictions through Exponential GPR in complete mode improve and qualify as excellent.

Establishing best practices in photoplethysmography signal acquisition and processing

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

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.

Synthetic PPG signal Generation to Improve Coronary Artery Disease Classification: Study with Physical Model of Cardiovascular System

This paper presents a novel approach of generating synthetic Photoplethysmogram (PPG) data using a physical model of the cardiovascular system to improve classifier performance with a combination of synthetic and real data. The physical model is an in-silico cardiac computational model, consisting of a four-chambered heart with electrophysiology, hemodynamic, and blood pressure autoregulation functionality. Starting with a small number of measured PPG data, the cardiac model is used to synthesize healthy as well as PPG time-series pertaining to coronary artery disease (CAD) by varying pathophysiological parameters. A Variational Autoencoder (VAE) structure is proposed to derive a statistical feature space for CAD classification. Results are presented in two perspectives namely, (i) using artificially reduced real disease data and (ii) using all the real disease data. In both cases, by augmenting with the synthetic data for training, the performance (sensitivity, specificity) of the classifier changes from (i) (0.65, 1) to (1, 0.9) and (ii) (1, 0.95) to (1,1). The proposed hybrid approach of combining physical modelling and statistical feature space selection generates realistic PPG data with pathophysiological interpretation and can outperform a baseline Generative Adversarial Network (GAN) architecture with a relatively small amount of real data for training. This proposed method could aid as a substitution technique for handling the problem of bulk data required for training machine learning algorithms for cardiac health-care applications.

A supervised machine learning semantic segmentation approach for detecting artifacts in plethysmography signals from wearables

Objective. Wearable devices equipped with plethysmography (PPG) sensors provided a low-cost, long-term solution to early diagnosis and continuous screening of heart conditions. However PPG signals collected from such devices often suffer from corruption caused by artifacts. The objective of this study is to develop an effective supervised algorithm to locate the regions of artifacts within PPG signals.Approach. We treat artifact detection as a 1D segmentation problem. We solve it via a novel combination of an active-contour-based loss and an adapted U-Net architecture. The proposed algorithm was trained on the PPG DaLiA training set, and further evaluated on the PPG DaLiA testing set, WESAD dataset and TROIKA dataset.Main results. We evaluated with the DICE score, a well-established metric for segmentation accuracy evaluation in the field of computer vision. The proposed method outperforms baseline methods on all three datasets by a large margin (≈7 percentage points above the next best method). On the PPG DaLiA testing set, WESAD dataset and TROIKA dataset, the proposed method achieved 0.8734 ± 0.0018, 0.9114 ± 0.0033 and 0.8050 ± 0.0116 respectively. The next best method only achieved 0.8068 ± 0.0014, 0.8446 ± 0.0013 and 0.7247 ± 0.0050.Significance. The proposed method is able to pinpoint exact locations of artifacts with high precision; in the past, we had only a binary classification of whether a PPG signal has good or poor quality. This more nuanced information will be critical to further inform the design of algorithms to detect cardiac arrhythmia.