Non-invasive continuous blood pressure monitoring systems: current and proposed technology issues and challenges

Physics-informed neural networks for parameter estimation in blood flow models

BACKGROUND

Physics-informed neural networks (PINNs) have emerged as a powerful tool for solving inverse problems, especially in cases where no complete information about the system is known and scatter measurements are available. This is especially useful in hemodynamics since the boundary information is often difficult to model, and high-quality blood flow measurements are generally hard to obtain.

METHODS

In this work, we use the PINNs methodology for estimating reduced-order model parameters and the full velocity field from scatter 2D noisy measurements in the aorta. Two different flow regimes, stationary and transient were studied.

RESULTS

We show robust and relatively accurate parameter estimations when using the method with simulated data, while the velocity reconstruction accuracy shows dependence on the measurement quality and the flow pattern complexity. Comparison with a Kalman filter approach shows similar results when the number of parameters to be estimated is low to medium. For a higher number of parameters, only PINNs were capable of achieving good results.

CONCLUSION

The method opens a door to deep-learning-driven methods in the simulations of complex coupled physical systems.

Hierarchical Piezoelectric Composites for Noninvasive Continuous Cardiovascular Monitoring

Continuous monitoring of blood pressure (BP) and multiparametric analysis of cardiac functions are crucial for the early diagnosis and therapy of cardiovascular diseases. However, existing monitoring approaches often suffer from bulky and intrusive apparatus, cumbersome testing procedures, and challenging data processing, hampering their applications in continuous monitoring. Here, a heterogeneously hierarchical piezoelectric composite is introduced for wearable continuous BP and cardiac function monitoring, overcoming the rigidity of ceramic and the insensitivity of polymer. By optimizing the hierarchical structure and components of the composite, the developed piezoelectric sensor delivers impressive performances, ensuring continuous and accurate monitoring of BP at Grade A level. Furthermore, the hemodynamic parameters are extracted from the detected signals, such as local pulse wave velocity, cardiac output, and stroke volume, all of which are in alignment with clinical results. Finally, the all-day tracking of cardiac function parameters validates the reliability and stability of the developed sensor, highlighting its potential for personalized healthcare systems, particularly in early diagnosis and timely intervention of cardiovascular disease.

Meta-Analysis of Pulse Transition Features in Non-Invasive Blood Pressure Estimation Systems: Bridging Physiology and Engineering Perspectives

The pulse transition features (PTFs), including pulse arrival time (PAT) and pulse transition time (PTT), hold significant importance in estimating non-invasive blood pressure (NIBP). However, the literature showcases considerable variations in terms of PTFs' correlation with blood pressure (BP), accuracy in NIBP estimation, and the comprehension of the relationship between PTFs and BP. This inconsistency is exemplified by the wide-ranging correlations reported across studies investigating the same feature. Furthermore, investigations comparing PAT and PTT have yielded conflicting outcomes. Additionally, PTFs have been derived from various bio-signals, capturing distinct characteristic points like the pulse's foot and peak. To address these inconsistencies, this study meticulously reviews a selection of such research endeavors while aligning them with the biological intricacies of blood pressure and the human cardiovascular system (CVS). Each study underwent evaluation, considering the specific signal acquisition locale and the corresponding recording procedure. Moreover, a comprehensive meta-analysis was conducted, yielding multiple conclusions that could significantly enhance the design and accuracy of NIBP systems. Grounded in these dual aspects, the study systematically examines PTFs in correlation with the specific study conditions and the underlying factors influencing the CVS. This approach serves as a valuable resource for researchers aiming to optimize the design of BP recording experiments, bio-signal acquisition systems, and the fine-tuning of feature engineering methodologies, ultimately advancing PTF-based NIBP estimation.

Flexible electronics for cardiovascular healthcare monitoring

Cardiovascular diseases (CVDs) are one of the most urgent threats to humans worldwide, which are responsible for almost one-third of global mortality. Over the last decade, research on flexible electronics for monitoring and treatment of CVDs has attracted tremendous attention. In contrast to conventional medical instruments in hospitals that are usually bulky, hard to move, monofunctional, and time-consuming, flexible electronics are capable of continuous, noninvasive, real-time, and portable monitoring. Notable progress has been made in this emerging field, and thus a number of significant achievements and concomitant research prospects deserve attention for practical implementation. Here, we comprehensively review the latest progress of flexible electronics for CVDs, focusing on new functions provided by flexible electronics. First, the characteristics of CVDs and flexible electronics and the foundation of their combination are briefly reviewed. Then, four representative applications of flexible electronics for CVDs are elaborated: blood pressure (BP) monitoring, electrocardiogram (ECG) monitoring, echocardiogram monitoring, and direct epicardium monitoring. Their operational principles, progress, merits and demerits, and future efforts are discussed. Finally, the remaining challenges and opportunities for flexible electronics for cardiovascular healthcare are outlined.

Thin, soft, wearable system for continuous wireless monitoring of artery blood pressure

Continuous monitoring of arterial blood pressure (BP) outside of a clinical setting is crucial for preventing and diagnosing hypertension related diseases. However, current continuous BP monitoring instruments suffer from either bulky systems or poor user-device interfacial performance, hampering their applications in continuous BP monitoring. Here, we report a thin, soft, miniaturized system (TSMS) that combines a conformal piezoelectric sensor array, an active pressure adaptation unit, a signal processing module, and an advanced machine learning method, to allow real wearable, continuous wireless monitoring of ambulatory artery BP. By optimizing the materials selection, control/sampling strategy, and system integration, the TSMS exhibits improved interfacial performance while maintaining Grade A level measurement accuracy. Initial trials on 87 volunteers and clinical tracking of two hypertension individuals prove the capability of the TSMS as a reliable BP measurement product, and its feasibility and practical usability in precise BP control and personalized diagnosis schemes development.

The identification of blood pressure variation with hypovolemia based on the volume compensation method

Objective: The purpose of this study is to identify the blood pressure variation, which is important in continuous blood pressure monitoring, especially in the case of low blood volume, which is critical for survival. Methods: A pilot study was conducted to identify blood pressure variation with hypovolemia using five Landrace pigs. New multi-dimensional morphological features of Photoplethysmography (PPG) were proposed based on experimental study of hemorrhagic shock in pigs, which were strongly correlated with blood pressure changes. Five machine learning methods were compared to develop the blood pressure variation identification model. Results: Compared with the traditional blood pressure variation identification model with single characteristic based on single period area of PPG, the identification accuracy of mean blood pressure variation based on the proposed multi-feature random forest model in this paper was up to 90%, which was 17% higher than that of the traditional blood pressure variation identification model. Conclusion: By the proposed multi-dimensional features and the identification method, it is more accurate to detect the rapid variation in blood pressure and to adopt corresponding measures. Significance: Rapid and accurate identification of blood pressure variation under low blood volume ultimately has the potential to effectively avoid complications caused by abnormal blood pressure in patients with clinical bleeding trauma.

Toward Real-Time Blood Pressure Monitoring via High-Fidelity Iontronic Tonometric Sensors with High Sensitivity and Large Dynamic Ranges

Continuous, noninvasive blood pressure (CNIBP) monitoring provides valuable hemodynamic information that renders detection of the early onset of cardiovascular diseases. Wearable mechano-electric pressure sensors that mount on the skin are promising candidates for monitoring continuous blood pressure (BP) pulse waveforms due to their excellent conformability, simple sensing mechanisms, and convenient signal acquisition. However, it is challenging to acquire high-fidelity BP pulse waveforms since it requires highly sensitive sensors (sensitivity larger than 4 × 10-5 kPa-1 ) that respond linearly with pressure change over a large dynamic range, covering the typical BP range (5-25 kPa). Herein, this work introduces a high-fidelity, iontronic-based tonometric sensor (ITS) with high sensitivity (4.82 kPa-1 ), good linearity (R2 > 0.995), and a large dynamic range (up to 180% output change) over a broad working range (0 to 38 kPa). Additionally, the ITS demonstrates a low limit of detection at 40 Pa, a fast load response time (35 ms) and release time (35 ms), as well as a stable response over 5000 load per release cycles, paving ways for potential applications in human-interface interaction, electronic skins, and robotic haptics. This work further explores the application of the ITS in monitoring real-time, beat-to-beat BP by measuring the brachial and radial pulse waveforms. This work provides a rational design of a wearable pressure sensor with high sensitivity, good linearity, and a large dynamic range for real-time CNIBP monitoring.

A Flexible Pressure Sensor with a Mesh Structure Formed by Lost Hair for Human Epidermal Pulse Wave Monitoring

Flexible pressure sensors with the capability of monitoring human vital signs show broad application prospects in personalized healthcare. In this work, a hair-based flexible pressure sensor (HBPS) consisting of lost hair and polymer films was proposed for the continuous monitoring of the human epidermal arterial pulse waveform. A macroscale mesh structure formed by lost hair provides a simplified spacer that endows the triboelectric-based flexible pressure sensor with sufficient contact-separation space. Based on this mesh structure design, the hair-based flexible pressure sensor can respond to the slight pressure change caused by an object with 5 mg weight and hold a stable output voltage under 1-30 Hz external pressure excitation. Additionally, the hair-based flexible pressure sensor showed great sensitivity (0.9 V/kPa) and decent stability after 4500 cycles of operation. Given these compelling features, the HBPS can successfully measure the human epidermal arterial pulses with obvious details at different arteries. The proposed HBPS can also be used to monitor the pulse signals of different subjects. Furthermore, the three different pulse wave transmission time (PTT) values (PTT-foot, PTT-middle, and PTT-peak) can be obtained by simultaneously monitoring human pulse and electrocardiogram signals, which has enormous application potential for assessing cardiovascular system health.

Intra-beat biomarker for accurate continuous non-invasive blood pressure monitoring

Accurate continuous non-invasive blood pressure (CNIBP) monitoring is the holy grail of digital medicine but remains elusive largely due to significant drifts in signal and motion artifacts that necessitate frequent device recalibration. To address these challenges, we developed a unique approach by creating a novel intra-beat biomarker (Diastolic Transit Time, DTT) to achieve highly accurate blood pressure (BP) estimations. We demonstrated our approach’s superior performance, compared to other common signal processing techniques, in eliminating stochastic baseline wander, while maintaining signal integrity and measurement accuracy, even during significant hemodynamic changes. We applied this new algorithm to BP data collected using non-invasive sensors from a diverse cohort of high acuity patients and demonstrated that we could achieve close agreement with the gold standard invasive arterial line BP measurements, for up to 20 min without recalibration. We established our approach's generalizability by successfully applying it to pulse waveforms obtained from various sensors, including photoplethysmography and capacitive-based pressure sensors. Our algorithm also maintained signal integrity, enabling reliable assessments of BP variability. Moreover, our algorithm demonstrated tolerance to both low- and high-frequency motion artifacts during abrupt hand movements and prolonged periods of walking. Thus, our approach shows promise in constituting a necessary advance and can be applied to a wide range of wearable sensors for CNIBP monitoring in the ambulatory and inpatient settings.

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.

Blood pressure measurement using only a smartphone

Hypertension is an immense challenge in public health. As one of the most prevalent medical conditions worldwide, it is a major cause of premature death. At present, the detection, diagnosis and monitoring of hypertension are subject to several limitations. In this review, we conducted a literature search on blood pressure measurement using only a smartphone, which has the potential to overcome current limitations and thus pave the way for long-term ambulatory blood pressure monitoring on a large scale. Among the 333 articles identified, we included 25 relevant articles over the past decade (November 2011–November 2021) and analyzed the described approaches to the types of underlying data recorded with smartphone sensors, the signal processing techniques applied to construct the desired signals, the features extracted from the constructed signals, and the algorithms used to estimate blood pressure. In addition, we analyzed the validation of the proposed methods against reference blood pressure measurements. We further examined and compared the effectiveness of the proposed approaches. Among the 25 articles, 23 propose an approach that requires direct contact between the sensor and the subject and two articles propose a contactless approach based on facial videos. The sample sizes in the identified articles range from three to 3000 subjects, where 8 articles used sample sizes of 85 or more subjects. Furthermore, 10 articles include hypertensive subjects in their participant pools. The methodologies applied for the evaluation of blood pressure measurement accuracy vary considerably among the analyzed articles. There is no consistency regarding the methods for blood pressure data collection and the reference blood pressure measurement and validation. Moreover, no established protocol is currently available for the validation of blood pressure measuring technologies using only a smartphone. We conclude the review with a discussion of the results and with recommendations for future research on the topic.

Piezoelectric Dynamics of Arterial Pulse for Wearable Continuous Blood Pressure Monitoring

Piezoelectric arterial pulse wave dynamics are traditionally considered to be similar to those of typical blood pressure waves. However, achieving accurate continuous blood pressure wave monitoring based on arterial pulse waves remains challenging, because the correlation between piezoelectric pulse waves and their related blood pressure waves is unclear. To address this, the correlation between piezoelectric pulse waves and blood pressure waves is first elucidated via theoretical, simulation, and experimental analysis of these dynamics. Based on this correlation, the authors develop a wireless wearable continuous blood pressure monitoring system, with better portability than conventional systems that are based on the pulse wave velocity between multiple sensors. They explore the feasibility of achieving wearable continuous blood pressure monitoring without motion artifacts, using a single piezoelectric sensor. These findings eliminate the controversy over the arterial pulse wave piezoelectric response, and can potentially be used to develop a portable wearable continuous blood pressure monitoring device for the early prevention and daily control of hypertension.

High-frame-rate A-mode ultrasound for calibration-free cuffless carotid pressure: feasibility study using lower body negative pressure intervention

PURPOSE

Existing technologies to measure central blood pressure (CBP) intrinsically depend on peripheral pressure or calibration models derived from it. Pharmacological or physiological interventions yielding different central and peripheral responses compromise the accuracy of such methods. We present a high-frame-rate ultrasound technology for cuffless and calibration-free evaluation of BP from the carotid artery. The system uses a pair of single-element ultrasound transducers to capture the arterial diameter and local pulse wave velocity (PWV) for the evaluation of beat-by-beat BP employing a novel biomechanical model.

MATERIALS AND METHODS

System's functionality assessment was conducted on eight male subjects (26 ± 4 years, normotensive and no history of cardiovascular risks) by perturbing pressure via short-term moderate lower body negative pressure (LBNP) intervention (-40 mmHg for 1 min). The ability of the system to capture dynamic responses of carotid pressure to LBNP was investigated and compared against the responses of peripheral pressure measured using a continuous BP monitor.

RESULTS

While the carotid pressure manifested trends similar to finger measurements during LBNP, the system also captured the differential carotid-to-peripheral pressure response, which corroborates the literature. The carotid diastolic and mean pressures agreed with the finger pressures (limits-of-agreement within ±7 mmHg) and exhibited acceptable uncertainty (mean absolute errors were 2.4 ± 3.5 and 2.6 ± 4.0 mmHg, respectively). Concurrent to the literature, the carotid systolic and pulse pressures (PPs) were significantly lower than those of the finger pressures by 11.1 ± 9.4 and 11.3 ± 8.2 mmHg, respectively (p < .0001).

CONCLUSIONS

The study demonstrated the method's potential for providing cuffless and calibration-free pressure measurements while reliably capturing the physiological aspects, such as PP amplification and dynamic pressure responses to intervention.

Effects of vessel wall mechanics on non-invasive evaluation of cardiovascular intrinsic frequencies

Intrinsic Frequency (IF) is a systems-based approach that provides valuable information for hemodynamic monitoring of the left ventricle (LV), the arterial system, and their coupling. Recent clinical studies have demonstrated the clinical significance of this method for prognosis and diagnosis of cardiovascular diseases. In IF analysis, two dominant instantaneous frequencies (ω₁ and ω₂) are extracted from arterial pressure waveforms. The value of ω₁ is related to the dynamics of the LV and the value of ω₂ is related to the dynamics of vascular function. This work investigates the effects of vessel wall mechanics on the accuracy and applicability of IFs extracted from vessel wall displacement waveforms compared to IFs extracted from pressure waveforms. In this study, we used a computational approach employing a fluid-structure interaction finite element method for various wall mechanics governed by linearly elastic, hyperelastic, and viscoelastic models. Results show that for vessels with elastic wall behavior, the error between displacement-based and pressure-based IFs is negligible. In the presence of stenosis or aneurysm in elastic arteries, the maximum errors associated with displacement-based IFs is less than 2%. For non-linear elastic and viscoelastic arteries, errors are more pronounced (where the former reaches up to 11% and the latter up to 27%). Our results ultimately suggest that displacement-based computations of ω₁ and ω₂ are accurate in vessels that exhibit elastic behavior (such as carotid arteries) and are suitable surrogates for pressure-based IFs. This is clinically significant because displacement-based IFs can be measured non-invasively.

An Alignment-Free Sensing Module for Noninvasive Radial Artery Blood Pressure Measurement

Sensor–artery alignment has always been a significant problem in arterial tonometry devices and prevents their application to wearable continuous blood pressure (BP) monitoring. Traditional solutions are to use a complex servo system to search for the best measurement position or to use an inefficient pressure sensor array. In this study, a novel solid–liquid mixture pressure sensing module is proposed. A flexible film with unique liquid-filled structures greatly reduces the pulse measurement error caused by sensor misplacement. The ideal measuring location was defined as −2.5 to 2.5 mm from the center of the module and the pressure variation was within 5.4%, which is available in the real application. Even at a distance of ±4 mm from the module center, the pressure decays by 23.7%, and its dynamic waveform is maintained. In addition, the sensing module is also endowed with the capability of measuring the pulse wave transmit time as a complementary method for BP measuring. The capability of the developed alignment-free sensing module in BP measurement was been validated. Twenty subjects were selected for the BP measurement experiment, which followed IEEE standards. The experimental results showed that the mean error of SBP is −4.26 mmHg with a standard deviation of 7.0 mmHg, and the mean error of DBP is 2.98 mmHg with a standard deviation of 5.07 mmHg. The device is expected to provide a new solution for wearable continuous BP monitoring.

Blood pressure assessment with in-ear photoplethysmography

Objective. In this study, we aimed to estimate blood pressure (BP) from in-ear photoplethysmography (PPG). This novel implementation provided an unobtrusive and steady way of recording PPG, whereas previous PPG measurements were mostly performed at the wrist, finger, or earlobe.Methods. The time between forward and reflected PPG waves was very short at the ear site. To minimize errors introduced by feature extraction, a multi-Gaussian decomposition of in-ear PPG was performed. Both hand-crafted and whole-based features were extracted and the best combination of features was selected using a backward-search wrapper method and evaluated by the Akaike information criteria. Hemodynamic parameters such as compliance and inertance were estimated from a four-element Windkessel (WK4) model, which was used to pre-classify PPG signals and generate different BP estimation algorithms. Calibration was done by using previous measurements from the same class. To validate this novel approach, 53 subjects were recruited for a one-month follow-up study, and 17 subjects were recruited for a two-month follow-up study. Calibrated systolic BP estimation accuracy was significantly improved with inertance-based pre-classification, while diastolic BP showed less improvement.Results. With proper feature selection, pre-classification and calibration, we have achieved a mean absolute error of 5.35 mmHg for SBP estimation, compared to 6.16 mmHg if no pre-classification was carried out. The performance did not deteriorate in two months, showing a decent BP trend-tracking ability.Conclusion. The study demonstrated the feasibility of in-ear PPG to reliably measure BP, which represents an important technological advancement in terms of unobtrusiveness and steadiness.

Scalability of cardiovascular intrinsic frequencies: Validations in preclinical models and non-invasive clinical studies

AIMS

Cardiovascular intrinsic frequencies (IFs) are associated with cardiovascular health and disease, separately capturing the systolic and diastolic information contained in a single (uncalibrated) arterial waveform. Previous clinical investigations related to IF have been restricted to studying chronic conditions, and hence its applicability for acute cardiovascular diseases has not been explored. Studies of cardiovascular complications such as acute myocardial infarction are difficult to perform in humans due to the high-risk and invasive nature of such procedures. Although they can be performed in preclinical (animal) models, the corresponding interpretation of IF measures and how they ultimately translate to humans is unknown. Hence, we studied the scalability of IF across species and sensor platforms.

MATERIALS AND METHODS

Scaled values of the two intrinsic frequencies ω1 and ω2 (corresponding to systolic and diastolic dynamics, respectively) were extracted from carotid waveforms acquired either non-invasively (via tonometry, Vivio or iPhone) in humans or invasively in rabbits and rats.

KEY FINDINGS

The scaled IF parameters for all species were found to fall within the same physiological ranges carrying similar statistical characteristics, even though body sizes and corresponding heart rates of the species were substantially different. Additionally, results demonstrated that all non-invasive sensor platforms were significantly correlated with each other for scaled IFs, suggesting that such analysis is device-agnostic and can be applied to upcoming wearable technologies.

SIGNIFICANCE

Ultimately, our results found that IFs are scalable across species, which is particularly valuable for the training of IF-based artificial intelligence systems using both preclinical and clinical data.

Flexible Wearable Sensors for Cardiovascular Health Monitoring

Cardiovascular diseases account for the highest mortality globally, but recent advances in wearable technologies may potentially change how these illnesses are diagnosed and managed. In particular, continuous monitoring of cardiovascular vital signs for early intervention is highly desired. To this end, flexible wearable sensors that can be comfortably worn over long durations are gaining significant attention. In this review, advanced flexible wearable sensors for monitoring cardiovascular vital signals are outlined and discussed. Specifically, the functional materials, configurations, mechanisms, and recent advances of these flexible sensors for heart rate, blood pressure, blood oxygen saturation, and blood glucose monitoring are highlighted. Different mechanisms in bioelectric, mechano-electric, optoelectric, and ultrasonic wearable sensors are presented to monitor cardiovascular vital signs from different body locations. Present challenges, possible strategies, and future directions of these wearable sensors are also discussed. With rapid development, these flexible wearable sensors will potentially be applicable for both medical diagnosis and daily healthcare use in tackling cardiovascular diseases.

Future trends in measuring physiology in free-living animals

Thus far, ecophysiology research has predominantly been conducted within controlled laboratory-based environments, owing to a mismatch between the recording technologies available for physiological monitoring in wild animals and the suite of behaviours and environments they need to withstand, without unduly affecting subjects. While it is possible to record some physiological variables for free-living animals using animal-attached logging devices, including inertial-measurement, heart-rate and temperature loggers, the field is still in its infancy. In this opinion piece, we review the most important future research directions for advancing the field of 'physiologging' in wild animals, including the technological development that we anticipate will be required, and the fiscal and ethical challenges that must be overcome. Non-invasive, multi-sensor miniature devices are ubiquitous in the world of human health and fitness monitoring, creating invaluable opportunities for animal and human physiologging to drive synergistic advances. We argue that by capitalizing on the research efforts and advancements made in the development of human wearables, it will be possible to design the non-invasive loggers needed by ecophysiologists to collect accurate physiological data from free-ranging animals ethically and with an absolute minimum of impact. In turn, findings have the capacity to foster transformative advances in human health monitoring. Thus, we invite biomedical engineers and researchers to collaborate with the animal-tagging community to drive forward the advancements necessary to realize the full potential of both fields. This article is part of the theme issue 'Measuring physiology in free-living animals (Part II)'.

A Continuous Cuffless Blood Pressure Estimation Using Tree-Based Pipeline Optimization Tool

High blood pressure (BP) may lead to further health complications if not monitored and controlled, especially for critically ill patients. Particularly, there are two types of blood pressure monitoring, invasive measurement, whereby a central line is inserted into the patient’s body, which is associated with infection risks. The second measurement is cuff-based that monitors BP by detecting the blood volume change at the skin surface using a pulse oximeter or wearable devices such as a smartwatch. This paper aims to estimate the blood pressure using machine learning from photoplethysmogram (PPG) signals, which is obtained from cuff-based monitoring. To avoid the issues associated with machine learning such as improperly choosing the classifiers and/or not selecting the best features, this paper utilized the tree-based pipeline optimization tool (TPOT) to automate the machine learning pipeline to select the best regression models for estimating both systolic BP (SBP) and diastolic BP (DBP) separately. As a pre-processing stage, notch filter, band-pass filter, and zero phase filtering were applied by TPOT to eliminate any potential noise inherent in the signal. Then, the automated feature selection was performed to select the best features to estimate the BP, including SBP and DBP features, which are extracted using random forest (RF) and k-nearest neighbors (KNN), respectively. To train and test the model, the PhysioNet global dataset was used, which contains 32.061 million samples for 1000 subjects. Finally, the proposed approach was evaluated and validated using the mean absolute error (MAE). The results obtained were 6.52 mmHg for SBS and 4.19 mmHg for DBP, which show the superiority of the proposed model over the related works.

Wearable Biosensors: An Alternative and Practical Approach in Healthcare and Disease Monitoring

With the increasing prevalence of growing population, aging and chronic diseases continuously rising healthcare costs, the healthcare system is undergoing a vital transformation from the traditional hospital-centered system to an individual-centered system. Since the 20th century, wearable sensors are becoming widespread in healthcare and biomedical monitoring systems, empowering continuous measurement of critical biomarkers for monitoring of the diseased condition and health, medical diagnostics and evaluation in biological fluids like saliva, blood, and sweat. Over the past few decades, the developments have been focused on electrochemical and optical biosensors, along with advances with the non-invasive monitoring of biomarkers, bacteria and hormones, etc. Wearable devices have evolved gradually with a mix of multiplexed biosensing, microfluidic sampling and transport systems integrated with flexible materials and body attachments for improved wearability and simplicity. These wearables hold promise and are capable of a higher understanding of the correlations between analyte concentrations within the blood or non-invasive biofluids and feedback to the patient, which is significantly important in timely diagnosis, treatment, and control of medical conditions. However, cohort validation studies and performance evaluation of wearable biosensors are needed to underpin their clinical acceptance. In the present review, we discuss the importance, features, types of wearables, challenges and applications of wearable devices for biological fluids for the prevention of diseased conditions and real-time monitoring of human health. Herein, we summarize the various wearable devices that are developed for healthcare monitoring and their future potential has been discussed in detail.

Trends and Challenges of Wearable Multimodal Technologies for Stroke Risk Prediction

We review in this paper the wearable-based technologies intended for real-time monitoring of stroke-related physiological parameters. These measurements are undertaken to prevent death and disability due to stroke. We compare the various characteristics, such as weight, accessibility, frequency of use, data continuity, and response time of these wearables. It was found that the most user-friendly wearables can have limitations in reporting high-precision prediction outcomes. Therefore, we report also the trend of integrating these wearables into the internet of things (IoT) and combining electronic health records (EHRs) and machine learning (ML) algorithms to establish a stroke risk prediction system. Due to different characteristics, such as accessibility, time, and spatial resolution of various wearable-based technologies, strategies of applying different types of wearables to maximize the efficacy of stroke risk prediction are also reported. In addition, based on the various applications of multimodal electroencephalography-functional near-infrared spectroscopy (EEG-fNIRS) on stroke patients, the perspective of using this technique to improve the prediction performance is elaborated. Expected prediction has to be dynamically delivered with high-precision outcomes. There is a need for stroke risk stratification and management to reduce the resulting social and economic burden.

Treatment of the hypertensive patient in 2030

Sarah Bingham, a 45 year old carer for her grandmother who suffered a stroke 4 months ago, feels a buzz on her wrist. It's time for them both to take their medications. Sarah makes dinner and leaves for her evening run. Her smartwatch detects her exit and turns off her TV as advertisements for incentivised private health insurance commence.

Management of Hypertension in the Digital Era: Small Wearable Monitoring Devices for Remote Blood Pressure Monitoring

Out-of-office blood pressure measurement is an essential part of diagnosing and managing hypertension. In the era of advanced digital health information technology, the approach to achieving this is shifting from traditional methods (ambulatory and home blood pressure monitoring) to wearable devices and technology. Wearable blood pressure monitors allow frequent blood pressure measurements (ideally continuous beat-by-beat monitoring of blood pressure) with minimal stress on the patient. It is expected that wearable devices will dramatically change the quality of detection and management of hypertension by increasing the number of measurements in different situations, allowing accurate detection of phenotypes that have a negative impact on cardiovascular prognosis, such as masked hypertension and abnormal blood pressure variability. Frequent blood pressure measurements and the addition of new features such as monitoring of environmental conditions allows interpretation of blood pressure data in the context of daily stressors and different situations. This new digital approach to hypertension contributes to anticipation medicine, which refers to strategies designed to identify increasing risk and predict the onset of cardiovascular events based on a series of data collected over time, allowing proactive interventions to reduce risk. To achieve this, further research and validation is required to develop wearable blood pressure monitoring devices that provide the same accuracy as current approaches and can effectively contribute to personalized medicine.

Non-Invasive Risk Stratification of Hypertension: A Systematic Comparison of Machine Learning Algorithms

One of the most important physiological parameters of the cardiovascular circulatory system is Blood Pressure. Several diseases are related to long-term abnormal blood pressure, i.e., hypertension; therefore, the early detection and assessment of this condition are crucial. The identification of hypertension, and, even more the evaluation of its risk stratification, by using wearable monitoring devices are now more realistic thanks to the advancements in Internet of Things, the improvements of digital sensors that are becoming more and more miniaturized, and the development of new signal processing and machine learning algorithms. In this scenario, a suitable biomedical signal is represented by the PhotoPlethysmoGraphy (PPG) signal. It can be acquired by using a simple, cheap, and wearable device, and can be used to evaluate several aspects of the cardiovascular system, e.g., the detection of abnormal heart rate, respiration rate, blood pressure, oxygen saturation, and so on. In this paper, we take into account the Cuff-Less Blood Pressure Estimation Data Set that contains, among others, PPG signals coming from a set of subjects, as well as the Blood Pressure values of the latter that is the hypertension level. Our aim is to investigate whether or not machine learning methods applied to these PPG signals can provide better results for the non-invasive classification and evaluation of subjects’ hypertension levels. To this aim, we have availed ourselves of a wide set of machine learning algorithms, based on different learning mechanisms, and have compared their results in terms of the effectiveness of the classification obtained.