ABSTRACT
The continuous monitoring of an individual's breathing can be an instrument for the assessment and enhancement of human wellness. Specific respiratory features are unique markers of the deterioration of a health condition, the onset of a disease, fatigue and stressful circumstances. The early and reliable prediction of high-risk situations can result in the implementation of appropriate intervention strategies that might be lifesaving. Hence, smart wearables for the monitoring of continuous breathing have recently been attracting the interest of many researchers and companies. However, most of the existing approaches do not provide comprehensive respiratory information. For this reason, a meta-learning algorithm based on LSTM neural networks for inferring the respiratory flow from a wearable system embedding FBG sensors and inertial units is herein proposed. Different conventional machine learning approaches were implemented as well to ultimately compare the results. The meta-learning algorithm turned out to be the most accurate in predicting respiratory flow when new subjects are considered. Furthermore, the LSTM model memory capability has been proven to be advantageous for capturing relevant aspects of the breathing pattern. The algorithms were tested under different conditions, both static and dynamic, and with more unobtrusive device configurations. The meta-learning results demonstrated that a short one-time calibration may provide subject-specific models which predict the respiratory flow with high accuracy, even when the number of sensors is reduced. Flow RMS errors on the test set ranged from 22.03 L/min, when the minimum number of sensors was considered, to 9.97 L/min for the complete setting (target flow range: 69.231 ± 21.477 L/min). The correlation coefficient r between the target and the predicted flow changed accordingly, being higher (r = 0.9) for the most comprehensive and heterogeneous wearable device configuration. Similar results were achieved even with simpler settings which included the thoracic sensors (r ranging from 0.84 to 0.88; test flow RMSE = 10.99 L/min, when exclusively using the thoracic FBGs). The further estimation of respiratory parameters, i.e., rate and volume, with low errors across different breathing behaviors and postures proved the potential of such approach. These findings lay the foundation for the implementation of reliable custom solutions and more sophisticated artificial intelligence-based algorithms for daily life health-related applications.
Subject(s)
Artificial Intelligence , Wearable Electronic Devices , Algorithms , Humans , Machine Learning , RespirationABSTRACT
The interest in photoplethysmography (PPG) for sleep monitoring is increasing because PPG may allow assessing heart rate variability (HRV), which is particularly important in breathing disorders. Thus, we aimed to evaluate how PPG wearable systems measure HRV during sleep at high altitudes, where hypobaric hypoxia induces respiratory disturbances. We considered PPG and electrocardiographic recordings in 21 volunteers sleeping at 4554 m a.s.l. (as a model of sleep breathing disorder), and five alpine guides sleeping at sea level, 6000 m and 6800 m a.s.l. Power spectra, multiscale entropy, and self-similarity were calculated for PPG tachograms and electrocardiography R-R intervals (RRI). Results demonstrated that wearable PPG devices provide HRV measures even at extremely high altitudes. However, the comparison between PPG tachograms and RRI showed discrepancies in the faster spectral components and at the shorter scales of self-similarity and entropy. Furthermore, the changes in sleep HRV from sea level to extremely high altitudes quantified by RRI and PPG tachograms in the five alpine guides tended to be different at the faster frequencies and shorter scales. Discrepancies may be explained by modulations of pulse wave velocity and should be considered to interpret correctly autonomic alterations during sleep from HRV analysis.
Subject(s)
Photoplethysmography , Wearable Electronic Devices , Altitude , Electrocardiography/methods , Heart Rate/physiology , Humans , Photoplethysmography/methods , Polysomnography , Pulse Wave Analysis , SleepABSTRACT
The indexes of cardiac mechanics can be derived from the cardiac time intervals, CTIs, i.e., the timings among the opening and closure of the aortic and mitral valves and the Q wave in the ECG. Traditionally, CTIs are estimated by ultrasound (US) techniques, but they may also be more easily assessed by the identification of specific fiducial points (FPs) inside the waveform of the seismocardiogram (SCG), i.e., the measure of the thorax micro-accelerations produced by the heart motion. While the correspondence of the FPs with the valve movements has been verified in healthy subjects, less information is available on whether this methodology may be routinely employed in the clinical practice for the monitoring of cardiac patients, in which an SCG waveform distortion is expected because of the heart dysfunction. In this study we checked the SCG shape in 90 patients with myocardial infarction (MI), heart failure (HF), or transplanted heart (TX), referred to our hospital for rehabilitation after an acute event or after surgery. The SCG shapes were classified as traditional (T) or non-traditional (NT) on whether the FPs were visible or not on the basis of nomenclature previously proposed in literature. The T shape was present in 62% of the patients, with a higher ∓ prevalence in MI (79%). No relationship was found between T prevalence and ejection fraction (EF). In 20 patients with T shape, we checked the FPs correspondence with the real valve movements by concomitant SCG and US measures. When compared with reference values in healthy subjects available in the literature, we observed that the Echo vs. FP differences are significantly more dispersed in the patients than in the healthy population with higher differences for the estimation of the mitral valve closure (-17 vs. 4 ms on average). Our results indicate that not every cardiac patient has an SCG waveform suitable for the CTI estimation, thus before starting an SCG-based CTI monitoring a preliminary check by a simultaneous SCG-US measure is advisable to verify the applicability of the methodology.
ABSTRACT
This study investigates the beat-to-beat relationships among Pulse Transit Times (PTTs) and Pulse Arrival Times (PATs) concomitantly measured from the heart to finger, ear and forehead vascular districts, and their links with continuous finger blood pressure and isovolumic contraction time (IVCT). These aspects were explored in 22 young volunteers at rest and during cold pressure test (CPT, thermal stress), handgrip (HG, isometric exercise) and cyclo-ergometer pedalling (CYC, dynamic exercise). The starting point of the PTT measures was detected by the seismocardiogram. Results indicate that PTTs measured at the ear, forehead and finger districts are uncorrelated each other at rest, and during CPT and HG. The stressors produced district-dependent changes in the PTT variability. Only the dynamic exercise was able to induce significant changes with respect to rest in the PTTs mean values (-40%, -36% and -17%, respectively for PTTear, PTTfore, PTTfinger), and synchronize their modulations. Similar trends were observed in the PATs. IVCT decreased during the application of stressors with a minimum at CYC (-25%) reflecting an augmented heart contractility. The increase in blood pressure (BP) at CPT was greater than that at CYC (137 vs. 128 mmHg), but the correlations between beat-to-beat transit times and BP were maximal at CYC (PAT showed a higher correlation than PTT; correlations were greater for systolic than for diastolic BP). This suggests that pulse transit times do not always depend directly on the beat-to-beat BP values but, under specific conditions, on other factors and mechanisms that concomitantly also influence BP.
Subject(s)
Hand Strength , Pulse Wave Analysis , Blood Pressure/physiology , Blood Pressure Determination/methods , Heart Rate/physiology , Humans , Pulse Wave Analysis/methodsABSTRACT
Commercially available health technologies such as smartphones and smartwatches, activity trackers and eHealth applications, commonly referred to as wearables, are increasingly available and used both in the leisure and healthcare sector for pulse and fitness/activity tracking. The aim of the Position Paper is to identify specific barriers and knowledge gaps for the use of wearables, in particular for heart rate (HR) and activity tracking, in clinical cardiovascular healthcare to support their implementation into clinical care. The widespread use of HR and fitness tracking technologies provides unparalleled opportunities for capturing physiological information from large populations in the community, which has previously only been available in patient populations in the setting of healthcare provision. The availability of low-cost and high-volume physiological data from the community also provides unique challenges. While the number of patients meeting healthcare providers with data from wearables is rapidly growing, there are at present no clinical guidelines on how and when to use data from wearables in primary and secondary prevention. Technical aspects of HR tracking especially during activity need to be further validated. How to analyse, translate, and interpret large datasets of information into clinically applicable recommendations needs further consideration. While the current users of wearable technologies tend to be young, healthy and in the higher sociodemographic strata, wearables could potentially have a greater utility in the elderly and higher-risk population. Wearables may also provide a benefit through increased health awareness, democratization of health data and patient engagement. Use of continuous monitoring may provide opportunities for detection of risk factors and disease development earlier in the causal pathway, which may provide novel applications in both prevention and clinical research. However, wearables may also have potential adverse consequences due to unintended modification of behaviour, uncertain use and interpretation of large physiological data, a possible increase in social inequality due to differential access and technological literacy, challenges with regulatory bodies and privacy issues. In the present position paper, current applications as well as specific barriers and gaps in knowledge are identified and discussed in order to support the implementation of wearable technologies from gadget-ology into clinical cardiology.
ABSTRACT
This article presents a new wearable platform, SeisMote, for the monitoring of cardiovascular function in controlled conditions and daily life. It consists of a wireless network of sensorized nodes providing simultaneous multiple measures of electrocardiogram (ECG), acceleration, rotational velocity, and photoplethysmogram (PPG) from different body areas. A custom low-power transmission protocol was developed to allow the concomitant real-time monitoring of 32 signals (16 bit @200 Hz) from up to 12 nodes with a jitter in the among-node time synchronization lower than 0.2 ms. The BluetoothLE protocol may be used when only a single node is needed. Data can also be collected in the off-line mode. Seismocardiogram and pulse transit times can be derived from the collected data to obtain additional information on cardiac mechanics and vascular characteristics. The employment of the system in the field showed recordings without data gaps caused by transmission errors, and the duration of each battery charge exceeded 16 h. The system is currently used to investigate strategies of hemodynamic regulation in different vascular districts (through a multisite assessment of ECG and PPG) and to study the propagation of precordial vibrations along the thorax. The single-node version is presently exploited to monitor cardiac patients during telerehabilitation.
Subject(s)
Monitoring, Physiologic/methods , Wireless Technology/instrumentation , Activities of Daily Living , Cardiovascular Diseases/diagnosis , Computer Communication Networks , Electric Power Supplies , Electrocardiography , Equipment Design , Heart Failure/rehabilitation , Humans , Monitoring, Physiologic/instrumentation , Photoplethysmography , Pulse Wave Analysis , Signal Processing, Computer-Assisted , Telemedicine/instrumentation , Telemedicine/methods , Wearable Electronic DevicesABSTRACT
Cardiac time intervals are important hemodynamic indices and provide information about left ventricular performance. Phonocardiography (PCG), impedance cardiography (ICG), and recently, seismocardiography (SCG) have been unobtrusive methods of choice for detection of cardiac time intervals and have potentials to be integrated into wearable devices. The main purpose of this study was to investigate the accuracy and precision of beat-to-beat extraction of cardiac timings from the PCG, ICG and SCG recordings in comparison to multimodal echocardiography (Doppler, TDI, and M-mode) as the gold clinical standard. Recordings were obtained from 86 healthy adults and in total 2,120 cardiac cycles were analyzed. For estimation of the pre-ejection period (PEP), 43% of ICG annotations fell in the corresponding echocardiography ranges while this was 86% for SCG. For estimation of the total systolic time (TST), these numbers were 43, 80, and 90% for ICG, PCG, and SCG, respectively. In summary, SCG and PCG signals provided an acceptable accuracy and precision in estimating cardiac timings, as compared to ICG.
ABSTRACT
BACKGROUND: Fingolimod, an oral drug used in multiple sclerosis (MS) treatment, exerts its action through S1P-receptor engagement. These receptors are also expressed in heart and endothelial cells. The engagement of receptors on the atrial heart myocytes may cause a slowing effect on heart rate (HR). We aimed to explore the acute effect of fingolimod on the cardiac autonomic control, a side-effect of the drug that still needs to be clarified. METHODS: In 10 MS patients, we investigated the influence of the first administration of fingolimod (0.5 mg) on sympathetic and parasympathetic indexes via the analysis of the HR variability, and on the baroreflex sensitivity via sequence and alpha coefficient techniques. RESULTS: Fingolimod produced an average HR maximal drop of 12.7 (7.8) beats/min and the minimal HR occurred after 2.73 (0.38) hours from the dose administration. The pulse interval (PI) mean value and the pNN50 and RMSSD indexes of parasympathetic drive to the heart significantly increased. Interestingly, in 6 out of 10 patients also the power in the low-frequency band (LF) increased. The baroreflex sensitivity was not modified by the first dose of the drug. CONCLUSIONS: Our findings indicate that although the first dose of fingolimod invariably activates the parasympathetic system, in several subjects, it may induce also a surge in the sympathetic cardiac drive. This suggests that not only the vagal, as usually assumed, but also the sympathetic autonomic branch should be considered in the risk profile assessment of MS patients starting treatment with fingolimod.
ABSTRACT
Closed-loop models of the interactions between blood pressure (BP) and heart rate variations allow for estimation of baroreflex sensitivity (feedback effects of BP changes on heart rate) while also considering the feedforward effects of heart rate on BP. Our study is aimed at comparing modulations of feedback and feedforward couplings over 24 h in normotensive and hypertensive subjects, by assessing closed-loop baroreflex models in ambulatory conditions. Continuous intra-arterial BP recordings were performed for 24 h in eight normotensive and eight hypertensive subjects. Systolic BP (SBP) and pulse interval (PI) beat-by-beat series were analyzed by an autoregressive moving average model over consecutive 6-min running windows, estimating closed-loop feedback and feedforward gains in each window. The open-loop feedback gain was estimated for comparison. Normotensive and hypertensive patients were compared during wake (18:00-22:00) and sleep (23:00-5:00) periods by a mixed-effect linear model at p < 0.05. In both groups feedback (feedforward) gain averaged values were higher (lower) in sleep than in wake. Moreover, the closed-loop feedback gain was higher in normotensive subjects both in wake and sleep, whereas the closed-loop feedforward gain was higher in hypertensive subjects during sleep. By contrast, no significant differences were found between the normotensive and hypertensive groups for the open-loop feedback gain. Therefore, the closed-loop SBP-PI model can detect circadian alterations in the feedforward gain of PI on SBP and derangements of spontaneous baroreflex sensitivity in hypertension not detectable with the open-loop approach. These findings may help to obtain a more comprehensive assessment of the autonomic dysfunction underlying hypertension and for the in-depth evaluation of the benefits of rehabilitation procedures on autonomic cardiovascular modulation.
ABSTRACT
Seismocardiogram, SCG, is the measure of precordial vibrations produced by the beating heart, from which cardiac mechanics may be explored on a beat-to-beat basis. We recently collected a large amount of SCG data (>69 recording hours) from an astronaut to investigate cardiac mechanics during sleep aboard the International Space Station and on Earth. SCG sleep recordings are characterized by a prolonged duration and wide heart rate swings, thus a specific algorithm was developed for their analysis. In this article we describe the new algorithm and its performance. The algorithm is composed of three parts: 1) artifacts removal, 2) identification in each SCG waveform of four fiducial points associated with the opening and closure of the aortic and mitral valves, 3) beat-to-beat computation of indexes of cardiac mechanics from the SCG fiducial points. The algorithm was tested on two sleep recordings and yielded the identification of the fiducial points in more than 36,000 beats with a precision, quantified by the Positive Predictive Value, ≥99.2%. These positive findings provide the first evidence that cardiac mechanics may be explored by the automatic analysis of SCG long-lasting recordings, taken out of the laboratory setting, and in presence of significant heart rate modulations.
Subject(s)
Aerospace Medicine , Heart/physiology , Sleep/physiology , Weightlessness , Algorithms , Ballistocardiography/methods , Earth, Planet , Electrocardiography/methods , Heart Rate/physiology , Humans , Signal Processing, Computer-Assisted , Space FlightABSTRACT
It is known from physiology that heartbeat and respiration influence venous return, but little is known regarding the extent to which these two factors affect flow. In this study, we estimated the prevalence of cardiac- and breathing-induced venous flow modulations in the internal jugular vein (IJV) and vertebral vein (VV) and the effects of posture. In 19 healthy patients, neck vein flow was examined with pulsed wave Doppler. Electrocardiogram and respiratory signals were simultaneously acquired. In supine position, heart contraction always influenced venous flow, whereas breathing influenced 68% of IJV and 34% of VV flow. In sitting position, heart contraction influenced 74% of IJV and 42% of VV flow; breathing influenced 68% of IJV and 61% of VV measures. Thus, cardiac influence is greatly present in supine position, whereas breathing influence prevails in the VV while sitting. This setup allowed us to observe that in some patients, expiration may cause an unexpected increase in venous flow.
Subject(s)
Blood Flow Velocity/physiology , Heart Rate/physiology , Jugular Veins/physiology , Myocardial Contraction/physiology , Posture/physiology , Respiratory Mechanics/physiology , Adolescent , Adult , Echocardiography, Doppler, Pulsed/methods , Female , Humans , Male , Middle Aged , Young AdultSubject(s)
Autonomic Nervous System/drug effects , Baroreflex/drug effects , Fingolimod Hydrochloride/therapeutic use , Heart/drug effects , Immunosuppressive Agents/therapeutic use , Multiple Sclerosis, Relapsing-Remitting/drug therapy , Adult , Female , Fingolimod Hydrochloride/pharmacology , Humans , Immunosuppressive Agents/pharmacology , Longitudinal Studies , Male , Middle AgedABSTRACT
INTRODUCTION: Uncontrolled hypertension is largely attributed to unsatisfactory doctor's engagement in its optimal management and to poor patients' compliance to therapeutic interventions. ICT and mobile Health solutions might improve these conditions, being widely available and providing highly effective communication strategies. OBJECTIVE: To evaluate whether ICT and mobile Health tools are able to improve hypertension control by improving doctors' engagement and by increasing patients' education and involvement, and their compliance to lifestyle modification and prescribed drug therapy. METHODS: In a pilot study, we have included 690 treated hypertensive patients with uncontrolled office blood pressure (BP), consecutively recruited by 9 general practitioners over 3 months. Patients were alternatively assigned to routine management based on repeated office visits or to an integrated ICT-based Patients Optimal Strategy for Treatment (POST) system including Home BP monitoring teletransmission, a dedicated web-based platform for patients' management by physicians (Misuriamo platform), and a smartphone mobile application (Eurohypertension APP, E-APP), over a follow-up of 6 months. BP values, demographic and clinical data were collected at baseline and at all follow-up visits (at least two). BP control and cardiovascular risk level have been evaluated at the beginning and at the end of the study. RESULTS: 89 patients did not complete the follow-up, thus data analysis was carried out in 601 of them (303 patients in the POST group and 298 in the control group). Office BP control (<;149/90 mmHg) was 40.0% in control group, and 72.3% in POST group at 6 month follow-up. At the same time Home BP control (<;135/85 mmHg average of 6 days) in POST group was 87.5%. CONCLUSION: this pilot study suggests that ICT based tools might be effective in improving hypertension management, implementing positive patients' involvement with better adherence to treatment prescriptions and providing the physicians with dynamic control of patients' home BP measurements, resulting in lesser clinical inertia.
Subject(s)
Hypertension/drug therapy , Patient Education as Topic/methods , Physicians , Telemedicine/methods , Antihypertensive Agents/pharmacology , Antihypertensive Agents/therapeutic use , Blood Pressure/drug effects , Female , Humans , Hypertension/physiopathology , Life Style , Male , Middle Aged , Mobile Applications , Patient Compliance/statistics & numerical data , Pilot ProjectsABSTRACT
BACKGROUND: Cardiovascular side effects such as bradycardia and atrioventricular block were observed during the early clinical trials of fingolimod in multiple sclerosis, and one cardiovascular- linked death has been reported in the post-marketing period. OBJECTIVE: To investigate the medium-term effects of fingolimod on heart function in order to obtain further insights into its cardiac safety profile. METHODS: The study involved 53 patients starting treatment with fingolimod 0.5 mg daily and 25 patients treated with natalizumab 300 mg monthly. Cardiac function was assessed by means of echocardiography at baseline (T0), and after one (T1), six (T6), and (in the case of the fingolimod group) 12 months (T12). RESULTS: Mean left ventricular ejection fraction significantly decreased and end-systolic volume increased from T0 to T1 (p=0.005) and T6 (p=0.0001) in the fingolimod but not the natalizumab group, although a slight increase was observed at T12. A similar decrease in ejection fraction was also observed after six months in nine patients switched from natalizumab to fingolimod. CONCLUSION: Fingolimod significantly reduces left ventricular systolic function in MS patients. This effect has no clinical consequences in subjects without previous cardiac disorders, but suggests that more caution is required in patients with current or previous heart failure.
Subject(s)
Fingolimod Hydrochloride/therapeutic use , Immunosuppressive Agents/therapeutic use , Multiple Sclerosis, Relapsing-Remitting/drug therapy , Stroke Volume , Ventricular Function, Left/drug effects , Adult , Case-Control Studies , Echocardiography , Female , Humans , Immunologic Factors/therapeutic use , Male , Middle Aged , Natalizumab/therapeutic use , Prospective Studies , Systole , Ventricular Function, Left/physiologyABSTRACT
The study compares permutation-based and coarse-grained entropy approaches for the assessment of complexity of short heart period (HP) variability recordings. Shannon permutation entropy (SPE) and conditional permutation entropy (CPE) are computed as examples of permutation-based entropies, while the k-nearest neighbor conditional entropy (KNNCE) is calculated as an example of coarse-grained conditional entropy. SPE, CPE and KNNCE were applied to ad-hoc simulated autoregressive processes corrupted by increasing amounts of broad band noise and to real HP variability series recorded after complete vagal blockade obtained via administration of a high dose of atropine (AT) in nine healthy volunteers and during orthostatic challenge induced by 90° head-up tilt (T90) in 15 healthy individuals. Over the simulated series the performances of SPE and CPE degraded more rapidly with the amplitude of the superimposed broad band noise than those of KNNCE. Over real data KNNCE identified the expected decrease of the HP variability complexity both after AT and during T90. Conversely SPE and CPE detected the decrease of HP variability complexity solely during T90 as a likely result of the more favorable signal-to-noise ratio during T90 than after AT. Results derived from both simulations and real data indicated that permutation-based entropies had a larger susceptibility to broad band noise than KNNCE. We recommend caution in applying permutation-based entropies in presence of short HP variability series characterized by a low signal-to-noise ratio.
Subject(s)
Electrocardiography/methods , Heart Rate/physiology , Signal Processing, Computer-Assisted , Adult , Computer Simulation , Humans , Male , Middle Aged , Models, Cardiovascular , Posture/physiology , Signal-To-Noise RatioABSTRACT
We propose a new methodology for the estimation of Pulse Transit Time, PTT, based on the use of the seismocardiogram for the identification of the aortic valve opening, AO. This method has been implemented to obtain a first description of the AO-derived PTT beat-to-beat variability at rest and during the recovery after a cycloergometer exercise at 25W and 100W, its relation with systolic blood pressure, S(BP), and its difference with respect to variability of the Pulse Arrival Time, PAT (i.e. the BP transit time estimated by considering the ECG R peak instead of AO as proximal site). Our preliminary data indicate that 1) the fast components of the PTT variability are only marginally influenced by respiration; 2) only the slower components of the PTT variability are correlated with systolic BP; 3) major differences exist in the dynamics of PTT and PAT, being PAT variability significantly larger and importantly influenced by the beat-to-beat changes occurring in the Pre Ejection Period.
Subject(s)
Electrocardiography/instrumentation , Pulse Wave Analysis/methods , Blood Pressure/physiology , Exercise/physiology , Heart Rate/physiology , Humans , Pilot Projects , Respiration , Rest/physiologyABSTRACT
In the past decade, there has been a resurgence in the field of unobtrusive cardiomechanical assessment, through advancing methods for measuring and interpreting ballistocardiogram (BCG) and seismocardiogram (SCG) signals. Novel instrumentation solutions have enabled BCG and SCG measurement outside of clinical settings, in the home, in the field, and even in microgravity. Customized signal processing algorithms have led to reduced measurement noise, clinically relevant feature extraction, and signal modeling. Finally, human subjects physiology studies have been conducted using these novel instruments and signal processing tools with promising results. This paper reviews the recent advances in these areas of modern BCG and SCG research.