Hee Nam Yoon

Nocturnal Awakening and Sleep Efficiency Estimation Using Unobtrusively Measured Ballistocardiogram

Fragmented sleep due to frequent awakenings represents a major cause of impaired daytime performance and adverse health outcomes. Currently, the gold standard for studying and assessing sleep fragmentation is polysomnography (PSG). Here, we propose an alternative method for real-time detection of nocturnal awakening via ballistocardiography using an unobtrusive polyvinylidene fluoride (PVDF) film sensor on a bed mattress. From ballistocardiogram, heart rate and body movement information were extracted to develop an algorithm for classifying sleeping and awakening epochs. In total, ten normal subjects (mean age 38.7 ± 14.6 years) and ten patients with obstructive sleep apnea (OSA) (mean age 44.2 ± 16.5 years) of varying symptom severity participated in this study. Our study detected awakening epochs with an average sensitivity of 85.3% and 85.2%, specificity of 98.4% and 97.7%, accuracy of 97.4% and 96.5%, and Cohens kappa coefficient of 0.83 and 0.81 for normal subjects and OSA patients, respectively. Also, sleep efficiency was estimated using detected awakening epochs and then compared with PSG results. Mean absolute errors in sleep efficiency were 1.08% and 1.44% for normal subjects and OSA patients, respectively. The results presented here indicate that our suggested method could be reliably applied to real-time nocturnal awakening detection and sleep efficiency estimation. Furthermore, our method may ultimately be an effective tool for long-term, home monitoring of sleep-wake behavior.

Heart Rate Variability Monitoring during Sleep Based on Capacitively Coupled Textile Electrodes on a Bed

In this study, we developed and tested a capacitively coupled electrocardiogram (ECG) measurement system using conductive textiles on a bed, for long-term healthcare monitoring. The system, which was designed to measure ECG in a bed with no constraints of sleep position and posture, included a foam layer to increase the contact region with the curvature of the body and a cover to ensure durability and easy installation. Nine healthy subjects participated in the experiment during polysomnography (PSG), and the heart rate (HR) coverage and heart rate variability (HRV) parameters were analyzed to evaluate the system. The experimental results showed that the mean of R-peak coverage was 98.0% (95.5%–99.7%), and the normalized errors of HRV time and spectral measures between the Ag/AgCl system and our system ranged from 0.15% to 4.20%. The root mean square errors for inter-beat (RR) intervals and HR were 1.36 ms and 0.09 bpm, respectively. We also showed the potential of our developed system for rapid eye movement (REM) sleep and wake detection as well as for recording of abnormal states.

Physiological Signal Monitoring Bed for Infants Based on Load-Cell Sensors

Ballistocardiographs (BCGs), which record the mechanical activity of the heart, have been a subject of interest for several years because of their advantages in providing unobtrusive physiological measurements. BCGs could also be useful for monitoring the biological signals of infants without the need for physical confinement. In this study, we describe a physiological signal monitoring bed based on load cells and assess an algorithm to extract the heart rate and breathing rate from the measured load-cell signals. Four infants participated in a total of 13 experiments. As a reference signal, electrocardiogram and respiration signals were simultaneously measured using a commercial device. The proposed automatic algorithm then selected the optimal sensor from which to estimate the heartbeat and respiration information. The results from the load-cell sensor signals were compared with those of the reference signals, and the heartbeat and respiration information were found to have average performance errors of 2.55% and 2.66%, respectively. The experimental results verify the positive feasibility of BCG-based measurements in infants.

Estimation of sleep posture using a patch-type accelerometer based device

In this study, we developed a sleep posture estimation algorithm using 3-axis accelerometer signals measured from a patch-type sensor. Firstly, we inspected the characteristics of accelerometer signals for different sleep postures. Based on the results, we established decision rules to estimate 5 postures containing supine, left, right lateral, prone postures, and non-sleep postures such as sitting and standing. The algorithm was tested by the data from thirteen subjects during night time PSG. As a result, the algorithm estimated sleep postures with an average agreement of 99.16%, and cohens kappa of 0.98 compared with reference sleep postures determined by position sensor and video recording. The proposed method with the device could be used as supportive purpose in routine PSG study and out-of-hospital environment.

Ballistocardiography for nonintrusive sleep structure estimation

Based on the its nonintrusive characteristics, ballistocardiography(BCG) has applied in the estimation of sleep structure without attaching any sensors to the subjects body. Loadcell or polyvinylidenefluoride (PVDF) film sensors are installed on the mattress for the monitoring of BCG. BCG peak was detected and heart rate variability parameters are derived. Parameters representing sleep structure and quality are estimated using these parameters. Sleep efficiency, four stages of sleep structure and sleep onset latency are estimated and results are compared with the results derived from polysomnographic recording.

Automatic Noise Removal and Peak Detection Algorithm for ECG Measured from Capacitively Coupled Electrodes Included within a Cloth Mattress Pad

Abstract: Recent technological advances have increased interest in personal health monitoring. Electrocardio-gram(ECG) monitoring is a basic healthcare activity and can provide decisive information regarding cardiovascularsystem status. In this study, we developed a capacitive ECG measurement system that can be included within a clothmattress pad. The device permits ECG data to be obtained during sleep by using capacitive electrodes. However,it is difficult to detect R-wave peaks automatically because signals obtained from the system can include a high levelof noise from various sources. Because R-peak detection is important in ECG applications, we developed an algorithmthat can reduce noise and improve detection accuracy under noisy conditions. Algorithm reliability was evaluated bydetermining its sensitivity(Se), positive predictivity(+P), and error rate(Er) by using data from the MIT-BIH Poly-somnographic Database and from our capacitive ECG system. The results showed that Se = 99.75%, +P = 99.77%,and Er = 0.47% for MIT-BIH Polysomnographic Database while Se = 96.47%, +P = 99.32%, and Er = 4.34% forour capacitive ECG system. Based on those results, we conclude that our R-peak detection method is capable of pro-viding useful ECG information, even under noisy signal conditions.Key words: Capacitively Measured ECG, R-peak Detection, Ubiquitous Healthcare, Non-intrusive ECG MonitoringSystem, MIT-BIH Polysomnographic Database

Automatic Detection of Slow-Wave Sleep Based on Electrocardiogram

Abstract: The objective of this research is to develop an automatic algorithm based on electrocardiogram (ECG)to estimate slow-wave sleep (SWS). An algorithm is based on 7 indices extracted from heart rate on ECG whichsimultaneously recorded with standard full night polysomnography from 31 subjects. Those 7 indices were thenapplied to independent component analysis to extract a feature that discriminates SWS and other sleep stages. OverallCohen’s kappa, accuracy, sensitivity and specificity of the algorithm to detect 30s epochs of SWS were 0.52, 0.87,0.70 and 0.90, respectively. The automatic SWS detection algorithm could be useful combining with existing REMand wake estimation technique on unattended home-based sleep monitoring. Key words: Slow-wave sleep, Electrocardiogram, Heart rate I. 서 론 서파 수면(Slow-wave sleep, SWS)은 수면에서 나타나는 다양한 수면 단계 중 하나로 생체신호의 특성 변화를 기반으로 수면 단계를 구분한 RKNREM)수면이다[1]. 전체 수면의 15-25%를 차지하는 서파수면은 그 이름에서도 알 수 있듯이 수면 중 1-2 Hz 이하의 뇌전도(EEG)의 크기가 우세하게 나타나는 특징을 가지며, 다양한 생리학적 기능을 수행한다고 알려져 있다. 기본적으로 인간은 서파 수면을 통해 일상 생활에 의한 뇌의 피로를 회복한다고 알려져 있으며[2], 다른 수면 상태보다 서파 수면에서 성장 호르몬의 분비가 증가한다고 밝혀졌다[3].또한, 최근 연구에서 서파 수면은 기억 응고화(Memoryconsolidation)의 기능을 한다고 밝힌바 있다[4,5]. 연구에의하면 서파 수면의 비율과 기억력 사이의 유의미한 양의상관 관계가 있으며, 이는 서파 수면에서 주로 나타나는 뇌파의 느린 진동(Slow oscillation)에 의한 것이라 입증하였다. 나아가 서파 수면의 양(Quantity)은 폐쇄성 수면 무호흡증 환자를 치료를 위한 중요한 특징으로 활용 될 수Corresponding Author : Park Kwang Suk 있다고 밝힌 연구 또한 존재한다[6]. 이처럼 서파 수면은 기능적으로 중요한 역할을 수행할 뿐만 아니라, 수면 관련 질환자들의 처치에도 유용하게 활용할 수 있기 때문에, 전체 수면에서 서파 수면을 검출하고, 이해하는 것은 큰 의미를 갖는다고 말할 수 있다.Seoul National University College of Medicine, 103 Daehak-ro, Jongno-gu, Seoul, KoreaTEL: +82-2-2072-3135 / FAX: +82-2-3676-2821E-mail: pks@bmsil.snu.ac.kr

Contrast between Spectral and Connectivity Features for Electroencephalography based Authentication

Biometrics using electroencephalography (EEG) have received attention as a strong security method and has been investigated by many researchers. Studies applied spectral and connectivity features to identify individuals. However, comparison of spectral and connectivity features are not yet conducted in the aspect of stability. In this paper, we present contrast between spectral and connectivity features for EEG based authentication with signals measured in different days. Spectral features are represented as power spectrum density (PSD) over 2-40Hz with 1Hz resolution provided from each channel. Connectivity features are presented as coherence (COH) of two channels combined, frequency range of 2-40Hz with 1Hz resolution. Total of 20 subjects participated and measured 32 channels of EEG for 10 seconds in eyes-closed resting state in three different days. We evaluated false authentication rate (FAR), false rejection rate (FRR) and half total error rate (HTER) as performance of authentication system designed: by using data measured in first day as train data (600 trials) and others as test data (1,173 trials). The similarity of data is measured using correlation modified Euclidean distance. During the decision making process, two values of threshold were set. The results were achieved with minimum of 10.45% HTER when using PSD, and 17.45% of HTER when using COH. It is well known that PSD features are relatively stable over time thus we post-analyzed coherence characteristics of EEG measured over three different days to evaluate stability. To assure stability, those that failed to reject ANOVA and highly correlated (over 0.8) were filtered in each subject in alpha band (8-13Hz) and composed coherence map for each participant. We concluded that considering both PSD and COH, feature filtering is necessary in order to guarantee efficient EEG based authentication.

Validation of Algorithm with Noise Tolerance Methods to Detect R-wave

Long-term monitoring of electrocardiogram (ECG) is one of the basic measurement in healthcare and provides decisive information regarding cardiovascular system status. In all ECG applications, the R-wave detection is important. However, it is difficult to detect R-wave automatically because signals obtained in daily life frequently include noise from various sources. Daily life monitoring ECG signal is particularly measured over cloth during walking and sleeping states by using non-invasive sensor, so that it usually presents higher noise level. To improve detection accuracy under noisy condition, we developed algorithm which has some noise tolerance techniques that analyze characteristics of R-wave. The proposed algorithm was evaluated by using the records of the MIT-BIH Polysomnographic database and the data from nonintrusive vital sign monitoring system previously developed in our laboratory. Algorithm reliability was assessed by detection error rate (De), sensitivity (Se) and positive predictivity (P+). The result shows average De of 2.62%, average Se of 98.29% and average P+ of 99.03%. We suggest that our R-wave detection method is useful in the presence of noisy signals.

Unconstrained Estimation of Body Postures on Bed Using Polyvinylidene Fluoride Film-based Sensor

Abstract: As body postures on bed affects various sleep related diseases, it is considered as important informationwhen monitoring sleeping in daily life. Though there have already been a few approaches to monitor body postureson bed conventionally, the development for simple and unconstrained methods is still needed to realize the long-termdaily monitoring. Focusing on the fact that ballistocardiogram changes depending on the body postures on bed, wedeveloped a novel method to estimate body posturesusing extremely simple, film-type ballistocardiogram sensorwhich is based on polyvinylidene fluoride(PVDF) film. With 10 subjects, we performed two experiments. One wasfor an estimation test to show that body postures on bed can be estimated by ballistocardiogram, and the other wasfor a reproducibility test to present the feasibility of ballistocardiogram based body postures monitoring. To estimatebody postures on bed, we made an individual template set of body postures by designating one ballistocardiogram(BCG) sample as a template in each postures. Then, we calculated Pearson’s correlation coefficients between a sam-ple and each templates and estimated the body posture of the sample by choosing a posture which corresponds tothe most significant correlation coefficients. As a result, we estimated body postures on bed with 99.2% accuracyin average and found that the estimation using ballistocardiogram is reproducible. Key words: Ballistocardiogram, sleeping position, PVDF film sensor