Automatic detection of spiking events in EMFi sheet during sleep

Unobstructive Heartbeat Monitoring of Sleeping Infants and Young Children Using Sheet-Type PVDF Sensors

Techniques for noninvasively acquiring the vital information of infants and young children are considered very useful in the fields of healthcare and medical care. An unobstructive measurement method for sleeping infants and young children under the age of 6 years using a sheet-type vital sensor with a polyvinylidene fluoride (PVDF) pressure-sensitive layer is demonstrated. The signal filter conditions to obtain the ballistocardiogram (BCG) and phonocardiogram (PCG) are discussed from the waveform data of infants and young children. The difference in signal processing conditions was caused by the physique of the infants and young children. The peak-to-peak interval (PPI) extracted from the BCG or PCG during sleep showed an extremely high correlation with the R-to-R interval (RRI) extracted from the electrocardiogram (ECG). The vital changes until awakening in infants monitored using a sheet sensor were also investigated. In infants under one year of age that awakened spontaneously, the distinctive vital changes during awakening were observed. Understanding the changes in the heartbeat and respiration signs of infants and young children during sleep is essential for improving the accuracy of abnormality detection by unobstructive sensors.

Investigating Cardiorespiratory Interaction Using Ballistocardiography and Seismocardiography-A Narrative Review

Ballistocardiography (BCG) and seismocardiography (SCG) are non-invasive techniques used to record the micromovements induced by cardiovascular activity at the body's center of mass and on the chest, respectively. Since their inception, their potential for evaluating cardiovascular health has been studied. However, both BCG and SCG are impacted by respiration, leading to a periodic modulation of these signals. As a result, data processing algorithms have been developed to exclude the respiratory signals, or recording protocols have been designed to limit the respiratory bias. Reviewing the present status of the literature reveals an increasing interest in applying these techniques to extract respiratory information, as well as cardiac information. The possibility of simultaneous monitoring of respiratory and cardiovascular signals via BCG or SCG enables the monitoring of vital signs during activities that require considerable mental concentration, in extreme environments, or during sleep, where data acquisition must occur without introducing recording bias due to irritating monitoring equipment. This work aims to provide a theoretical and practical overview of cardiopulmonary interaction based on BCG and SCG signals. It covers the recent improvements in extracting respiratory signals, computing markers of the cardiorespiratory interaction with practical applications, and investigating sleep breathing disorders, as well as a comparison of different sensors used for these applications. According to the results of this review, recent studies have mainly concentrated on a few domains, especially sleep studies and heart rate variability computation. Even in those instances, the study population is not always large or diversified. Furthermore, BCG and SCG are prone to movement artifacts and are relatively subject dependent. However, the growing tendency toward artificial intelligence may help achieve a more accurate and efficient diagnosis. These encouraging results bring hope that, in the near future, such compact, lightweight BCG and SCG devices will offer a good proxy for the gold standard methods for assessing cardiorespiratory function, with the added benefit of being able to perform measurements in real-world situations, outside of the clinic, and thus decrease costs and time.

An Open Source Classifier for Bed Mattress Signal in Infant Sleep Monitoring

Objective

To develop a non-invasive and clinically practical method for a long-term monitoring of infant sleep cycling in the intensive care unit.

Methods

Forty three infant polysomnography recordings were performed at 1–18 weeks of age, including a piezo element bed mattress sensor to record respiratory and gross-body movements. The hypnogram scored from polysomnography signals was used as the ground truth in training sleep classifiers based on 20,022 epochs of movement and/or electrocardiography signals. Three classifier designs were evaluated in the detection of deep sleep (N3 state): support vector machine (SVM), Long Short-Term Memory neural network, and convolutional neural network (CNN).

Results

Deep sleep was accurately identified from other states with all classifier variants. The SVM classifier based on a combination of movement and electrocardiography features had the highest performance (AUC 97.6%). A SVM classifier based on only movement features had comparable accuracy (AUC 95.0%). The feature-independent CNN resulted in roughly comparable accuracy (AUC 93.3%).

Conclusion

Automated non-invasive tracking of sleep state cycling is technically feasible using measurements from a piezo element situated under a bed mattress.

Significance

An open source infant deep sleep detector of this kind allows quantitative, continuous bedside assessment of infant’s sleep cycling.

Emfit Bed Sensor Activity Shows Strong Agreement with Wrist Actigraphy for the Assessment of Sleep in the Home Setting

PURPOSE

Wrist-worn actigraphy via research-grade devices, a well-established approach to the assessment of rest-activity, is limited by poor compliance, battery life, and lack of direct evidence for time spent physically in the bed. A non-invasive bed sensor (Emfit) may provide advantages over actigraphy for long-term sleep assessment in the home. This study compared sleep-wake measurements between this sensor and a validated actigraph.

PATIENTS AND METHODS

Thirty healthy subjects (6 to 54 years) underwent simultaneous monitoring with both devices for 14 days and filled out a daily sleep diary. Parameters included bed entry time, sleep start, sleep end, bed exit time, rest interval duration, and wake after sleep onset (WASO). The agreement between the two devices was measured using Bland-Altman plots and inter-class correlation coefficients (ICC). In addition, sensitivity, specificity, and accuracy were obtained from epoch-by-epoch comparisons of Emfit and actigraphy.

RESULTS

Fifteen percent of the subjects reported that wearing the actigraph was a burden. None reported that using the bed sensor was a burden. The minimal detectable change between Emfit and actigraphy was 11 minutes for bed entry time, 14 minutes for sleep start, 14 minutes for sleep end, 10 minutes for bed exit time, 20 minutes for rest interval duration, and 110 minutes for WASO. Inter-class correlation coefficients revealed an excellent agreement for all sleep parameters (ICC=0.99, 95% CI 98-99) except for WASO (ICC=0.46, 95% CI 0.33-0.56). Sensitivity, specificity, and accuracy were 0.62, 0.93, and 0.88, respectively. Kappa correlation analysis revealed a moderate correlation between the two devices (κ=0.55, p<0.0001).

CONCLUSION

Emfit is an acceptable alternative to actigraphy for the estimation of bed entry time, sleep start, sleep end, bed exit time, and rest interval duration. However, WASO estimates are poorly correlated between the two devices. Emfit may offer methodological advantages in situations where actigraphy is challenging to implement.

A Ballistographic Approach for Continuous and Non-Obtrusive Monitoring of Movement in Neonates

Continuously monitoring body movement in preterm infants can have important clinical applications since changes in movement-patterns can be a significant marker for clinical deteriorations including the onset of sepsis, seizures, and apneas. This paper proposes a system and method to monitor body movement of preterm infants in a clinical environment using ballistography. The ballistographic signal (BSG) is acquired using a thin and a film-like sensor that is placed underneath an infant. Manual annotations based on video-recordings served as a reference standard for identifying movement. We investigated the performance of multiple features, constructed from the BSG waveform, to discriminate movement from no movement based on data acquired from 10 preterm infants. Since routine cardiorespiratory monitoring is prone to movement artifacts, we also compared the application of these features on the simultaneously acquired cardiorespiratory waveforms, i.e., the electrocardiogram, the chest impedance, and the photoplethysmogram. The BSG-based-features consistently outperformed those based on the routinely acquired cardiorespiratory waveforms. The best performing BSG-based feature-the signal instability index-had a mean (standard deviation) effect size of 0.90 (0.06), as measured by the area under the receiver operating curve. The proposed system for monitoring body movement is robust to noise, non-obtrusive, and has high performance in clinical settings.

Detection of Snores Using Source Separation on an Emfit Signal

Snoring (SN) is an early sign of upper airway dysfunction, and it is strongly associated with obstructive sleep apnea. SN detection is important to monitor SN objectively and to improve the diagnostic sensitivity of sleep-disordered breathing. In this study, an automatic snore detection method using an electromechanical film transducer (Emfit) signal is presented. Representative polysomnographs of normal breathing and SN periods from 30 subjects were selected. Individual SN events were identified using source separation applying nonnegative matrix factorization deconvolution. The algorithm was evaluated using manual annotation of the polysomnographic recordings. According to our results, the sensitivity, and the positive predictive value of the developed method to reveal snoring from the Emfit signal were 82.81% and 86.29%, respectively. Compared to other approaches, our method adapts to the individual spectral snoring profile of the subject rather than matching a particular spectral profile, estimates the snoring intensity, and obtains the specific spectral profile of the snores in the epoch. Additionally, no training is necessary. This study suggests that it is possible to detect individual SN events with Emfit mattress, which can be used as a contactless alternative to more conventional methods such as piezo-snore sensors or microphones.

Emfit movement sensor in evaluating nocturnal breathing

Obstructive sleep apnea (OSA) diagnostics by the movement sensors static charge-sensitive bed (SCSB) and electromechanical film transducer (Emfit) is based on dividing the signal into different breathing patterns. The usage of non-invasive mattress sensors in diagnosing OSA is particularly tempting if patient has many other non sleep-related monitoring sensors. However, a systematic comparison of the apnea-hypopnea index (AHI) with Emfit-parameters is lacking. In addition to periodic breathing, SCSB and Emfit visualize episodes of sustained negative increases in intrathoracic pressure (increased respiratory resistance, IRR), of which relevance is still ambiguous. Our aim is to compare Emfit-parameters with the AHI and to provide a description of the patients suffering from IRR. Time percentage with all obstructive periodic Emfit breathing patterns (OPTotal%) showed the best correlation with the AHI. The OPTotal percentage of 21 yielded to excellent accuracy in detecting subjects with an AHI of 15/h or more. Patients with IRR received high scores in GHQ-12-questionnaire. An Emfit movement sensor might offer additional information in OSA diagnostics especially if nasal pressure transducer cannot be used.

Increased respiratory effort during sleep is non-invasively detected with movement sensor

INTRODUCTION

Measuring breathing effort during sleep with an oesophageal pressure sensor remains technically challenging and has not become routine practice. The aim of the present work was to investigate whether increased thoracic pressure during sleep can be detected with the Emfit movement sensor. Experimental data suggest that increased respiratory efforts with the intrathoracic pressure variation induce high-frequency spikes in the Emfit signal, but this has not been systematically examined.

METHODS

Polysomnography, oesophageal pressure and Emfit signal were recorded in 32 patients with suspected sleep-disordered breathing. Increased respiratory effort was defined as oesophageal pressure below -8 cmH(2)O during inspiration. The epochs of normal breathing, periodic breathing patterns and sustained spiking labelled as increased respiratory resistance (IRR) were defined on the Emfit signal according to established rules.

RESULTS

Compared to normal breathing, the proportion of increased respiratory effort was higher during all periodic breathing with spiking. The highest proportion (18-23%) occurred during IRR, which is characterised by sustained spiking.

CONCLUSION

The Emfit movement sensor is a non-invasive alternative to the oesophageal pressure sensor in the assessment of the respiratory effort during sleep. In particular, the Emfit sensor enhances detection of non-apnoeic sleep-disordered breathing, the significance of which should not be ignored.

Sleep staging based on signals acquired through bed sensor

We describe a system for the evaluation of the sleep macrostructure on the basis of Emfit sensor foils placed into bed mattress and of advanced signal processing. The signals on which the analysis is based are heart-beat interval (HBI) and movement activity obtained from the bed sensor, the relevant features and parameters obtained through a time-variant autoregressive model (TVAM) used as feature extractor, and the classification obtained through a hidden Markov model (HMM). Parameters coming from the joint probability of the HBI features were used as input to a HMM, while movement features are used for wake period detection. A total of 18 recordings from healthy subjects, including also reference polysomnography, were used for the validation of the system. When compared to wake-nonrapid-eye-movement (NREM)-REM classification provided by experts, the described system achieved a total accuracy of 79+/-9% and a kappa index of 0.43+/-0.17 with only two HBI features and one movement parameter, and a total accuracy of 79+/-10% and a kappa index of 0.44+/-0.19 with three HBI features and one movement parameter. These results suggest that the combination of HBI and movement features could be a suitable alternative for sleep staging with the advantage of low cost and simplicity.

Compliance and technical feasibility of long-term health monitoring with wearable and ambient technologies

We developed a system consisting of both wearable and ambient technologies designed to monitor personal wellbeing for several months during daily life. The variables monitored included bodyweight, blood pressure, heart-rate variability and air temperature. Two different user groups were studied: there were 17 working-age subjects participating in a vocational rehabilitation programme and 19 elderly people living in an assisted living facility. The working-age subjects collected data for a total of 1406 days; the average participation period was 83 days (range 43-99). The elderly subjects collected data for a total of 1593 days; the average participation period was 84 days (range 19-107). Usage, technical feasibility and usability of the system were also studied. Some technical and practical problems appeared which we had not expected such as thunder storm damage to equipment in homes and scheduling differences between staff and the subjects. The users gave positive feedback in almost all their responses in a questionnaire. The study suggests that the data-collection rate is likely be 70-90% for typical health monitoring data.

Ballistocardiogaphic studies with acceleration and electromechanical film sensors

The purpose of this research is to demonstrate and compare the utilization of electromechanical film (EMFi) and two acceleration sensors, ADXL202 and MXA2500U, for ballistocardiographic (BCG) and pulse transit time (PTT) studies. We have constructed a mobile physiological measurement station including amplifiers and a data collection system to record the previously mentioned signals and an electrocardiogram signal. Various versions of the measuring systems used in BCG studies in the past are also presented and evaluated. We have showed the ability of the EMFi sensor to define the elastic properties of the cardiovascular system and to ensure the functionality of the proposed instrumentation in different physiological loading conditions, before and after exercise and sauna bath. The EMFi sensor provided a BCG signal of good quality in the study of the human heart and function of the cardiovascular system with different measurement configurations. EMFi BCG measurements provided accurate and repeatable results for the different components of the heart cycle. In multiple-channel EMFi measurements, the carotid and limb pulse signals acquired were detailed and distinctive, allowing accurate PTT measurements. Changes in blood pressure were clearly observed and easily determined with EMFi sensor strips in pulse wave velocity (PWV) measurements. In conclusion, the configuration of the constructed device provided reliable measurements of the electrocardiogram, BCG, heart sound, and carotid and ankle pulse wave signals. Attached EMFi sensor strips on the neck and limbs yield completely new applications of the EMFi sensors aside from the conventional seat and supine recordings. Higher sensitivity, ease of utilization, and minimum discomfort of the EMFi sensor compared with acceleration sensors strengthen the status of the EMFi sensor for accurate and reliable BCG and PWV measurements, providing novel evaluation of the elastic properties of the cardiovascular system.

Nonintrusive measurement of biological signals for ubiquitous healthcare

In order to monitor biological signals during our daily lives for ubiquitous healthcare, nonintrusive biological signal monitoring methods have been developed. Without contacting sensors and connecting wires to the subjects, the biological signals are monitored using specially designed methods. ECG is measured using capacitive coupling over clothes and PPG is measured nonintrusively during ordinary activities. Blood pressure is also estimated from ECG and PPG by calculating pulse arrival time (PAT). These methods can be applied to evaluate the health levels of subjects without intervening in their ordinary daily activities.

Periodic limb movement screening as an additional feature of Emfit sensor in sleep-disordered breathing studies

BACKGROUND

The standard method for recording periodic limb movements is anterior tibialis electromyography (EMG) but other methods are also used. A new movement sensor Emfit (ElectroMechanical Film) provides information about sleep-disordered breathing but also shows movements in bed. The aim of the study was to investigate the usability of a small Emfit sensor in revealing periodic movements.

METHODS

Twenty seven consecutive patients were studied. Periodic movements in EMG and Emfit were scored blindly and periodic leg movement index (PLMI) for EMG and periodic movement index (PMI) for Emfit were counted. Spearman's correlation coefficient was used to assess the relationship between Emfit data and EMG results. Sensitivities and specificities were computed for PLMI and PMI levels of 5 and 15 movements/h. Additionally, receiver operating characteristic (ROC) curves were derived and the area under the curve (AUC) was calculated.

RESULTS

The Spearman's correlation coefficient between the PMI of Emfit and the PLMI of EMG was 0.87. The sensitivity of the Emfit sensor to detect periodic limb movements was 0.91 at the level of 5 movements/h and 0.73 when the cut-off level was 15 movements/h. The specificities were 0.75 and 1.00, respectively. AUC in ROC analysis was 0.96 and 0.98 in the levels of 5 and 15 movements/h.

CONCLUSIONS

The results suggest that the Emfit sensor might be suitable for screening of periodic limb movements even if the sensor is placed under the thoracic area of the patient in sleep-disordered breathing studies.

FFT averaging of multichannel BCG signals from bed mattress sensor to improve estimation of heart beat interval

A multichannel pressure sensing Emfit foil was integrated to a bed mattress for measuring ballistocardiograph signals during sleep. We calculated the heart beat interval with cepstrum method, by applying FFT for short time windows including pair of consequent heart beats. We decreased the variance of FFT by averaging the multichannel data in the frequency domain. Relative error of our method in reference to electrocardiograph RR interval was only 0.35% for 15 night recordings with six normal subjects, when 12% of data was automatically removed due to movement artifacts. Background motivation for this work is given from the studies applying heart rate variability for the sleep staging.

Broadband EMFi-based transducers for ultrasonic air applications

In this work, we explore the possibilities of electromechanical film (EMFi) as a new material for developing broadband transducers for ultrasonic air applications. The advantages of the EMFi film are its wide usable frequency range and easiness to use, making it highly suitable for self made, customizable ultrasonic sensors. This paper presents theoretical and experimental information focused on the needs of the sensor's end user, namely, frequency response, actual dynamic mass and Young's modulus, bandwidth, sensitivity, electromechanical dynamical model, acoustic response, and directivity. It is found empirically that the behavior of the film as an almost ideal piston-like acoustic source permits accurate prediction of the characteristics of transducers built on a developable surface. The results obtained represent the first step to more complex geometries, and, ultimately, to completely customizable field ultrasonic transducers.

Prolonged spiking in the Emfit sensor in patients with sleep-disordered breathing is characterized by increase in transcutaneous carbon dioxide

A phenomenon of prolonged spiking in movement sensors, such as static-charge-sensitive bed or Emfit (electromechanical film) sensors, has been connected to an increase in carbon dioxide tension in wakefulness. Spiking is also a common finding in sleep studies. This made us hypothesize that carbon dioxide changes might also happen in sleep during prolonged spiking episodes in Emfit sheet. We examined four different kinds of breathing pattern episodes: normal breathing, episodes of repetitive apnea, episodes of repetitive hypopnea and episodes with prolonged spiking lasting at least 3 min. One hundred and fifteen episodes from 19 polysomnograms were finally admitted to the study according to the protocol. The changes in the transcutaneous carbon dioxide tension (TcCO(2)) were defined for different breathing patterns. During prolonged spiking episodes the TcCO(2) increased significantly and differed statistically from the TcCO(2) changes of normal breathing and periodic breathing patterns (episodes of apnea and hypopnea). The rise in TcCO(2) during prolonged spiking episodes might suggest that prolonged spiking is representing another type of breathing disturbance during sleep differing from periodic breathing patterns. The Emfit sensor as a small, flexible and non-invasive sensor might provide useful additional information about breathing during sleep.

Algorithms to qualify respiratory data collected during the transport of trauma patients

We developed a quality indexing system to numerically qualify respiratory data collected by vital-sign monitors in order to support reliable post-hoc mining of respiratory data. Each monitor-provided (reference) respiratory rate (RR(R)) is evaluated, second-by-second, to quantify the reliability of the rate with a quality index (QI(R)). The quality index is calculated from: (1) a breath identification algorithm that identifies breaths of 'typical' sizes and recalculates the respiratory rate (RR(C)); (2) an evaluation of the respiratory waveform quality (QI(W)) by assessing waveform ambiguities as they impact the calculation of respiratory rates and (3) decision rules that assign a QI(R) based on RR(R), RR(C) and QI(W). RR(C), QI(W) and QI(R) were compared to rates and quality indices independently determined by human experts, with the human measures used as the 'gold standard', for 163 randomly chosen 15 s respiratory waveform samples from our database. The RR(C) more closely matches the rates determined by human evaluation of the waveforms than does the RR(R) (difference of 3.2 +/- 4.6 breaths min(-1) versus 14.3 +/- 19.3 breaths min(-1), mean +/- STD, p < 0.05). Higher QI(W) is found to be associated with smaller differences between calculated and human-evaluated rates (average differences of 1.7 and 8.1 breaths min(-1) for the best and worst QI(W), respectively). Establishment of QI(W) and QI(R), which ranges from 0 for the worst-quality data to 3 for the best, provides a succinct quantitative measure that allows for automatic and systematic selection of respiratory waveforms and rates based on their data quality.