Measuring dissimilarity between respiratory effort signals based on uniform scaling for sleep staging.

Polysomnography (PSG) has been extensively studied for sleep staging, where sleep stages are usually classified as wake, rapid-eye-movement (REM) sleep, or non-REM (NREM) sleep (including light and deep sleep). Respiratory information has been proven to correlate with autonomic nervous activity that is related to sleep stages. For example, it is known that the breathing rate and amplitude during NREM sleep, in particular during deep sleep, are steadier and more regular compared to periods of wakefulness that can be influenced by body movements, conscious control, or other external factors. However, the respiratory morphology has not been well investigated across sleep stages. We thus explore the dissimilarity of respiratory effort with respect to its signal waveform or morphology. The dissimilarity measure is computed between two respiratory effort signal segments with the same number of consecutive breaths using a uniform scaling distance. To capture the property of signal morphological dissimilarity, we propose a novel window-based feature in a framework of sleep staging. Experiments were conducted with a data set of 48 healthy subjects using a linear discriminant classifier and a ten-fold cross validation. It is revealed that this feature can help discriminate between sleep stages, but with an exception of separating wake and REM sleep. When combining the new feature with 26 existing respiratory features, we achieved a Cohen's Kappa coefficient of 0.48 for 3-stage classification (wake, REM sleep and NREM sleep) and of 0.41 for 4-stage classification (wake, REM sleep, light sleep and deep sleep), which outperform the results obtained without using this new feature.

An adaptive Kalman filter approach for cardiorespiratory signal extraction and fusion of non-contacting sensors.

BACKGROUND: Extracting cardiorespiratory signals from non-invasive and non-contacting sensor arrangements, i.e. magnetic induction sensors, is a challenging task. The respiratory and cardiac signals are mixed on top of a large and time-varying offset and are likely to be disturbed by measurement noise. Basic filtering techniques fail to extract relevant information for monitoring purposes. METHODS: We present a real-time filtering system based on an adaptive Kalman filter approach that separates signal offsets, respiratory and heart signals from three different sensor channels. It continuously estimates respiration and heart rates, which are fed back into the system model to enhance performance. Sensor and system noise covariance matrices are automatically adapted to the aimed application, thus improving the signal separation capabilities. We apply the filtering to two different subjects with different heart rates and sensor properties and compare the results to the non-adaptive version of the same Kalman filter. Also, the performance, depending on the initialization of the filters, is analyzed using three different configurations ranging from best to worst case. RESULTS: Extracted data are compared with reference heart rates derived from a standard pulse-photoplethysmographic sensor and respiration rates from a flowmeter. In the worst case for one of the subjects the adaptive filter obtains mean errors (standard deviations) of -0.2 min(-1) (0.3 min(-1)) and -0.7 bpm (1.7 bpm) (compared to -0.2 min(-1) (0.4 min(-1)) and 42.0 bpm (6.1 bpm) for the non-adaptive filter) for respiration and heart rate, respectively. In bad conditions the heart rate is only correctly measurable when the Kalman matrices are adapted to the target sensor signals. Also, the reduced mean error between the extracted offset and the raw sensor signal shows that adapting the Kalman filter continuously improves the ability to separate the desired signals from the raw sensor data. The average total computational time needed for the Kalman filters is under 25% of the total signal length rendering it possible to perform the filtering in real-time. CONCLUSIONS: It is possible to measure in real-time heart and breathing rates using an adaptive Kalman filter approach. Adapting the Kalman filter matrices improves the estimation results and makes the filter universally deployable when measuring cardiorespiratory signals.

Automatic detection of overnight deep sleep based on heart rate variability: a preliminary study.

This preliminary study investigated the use of cardiac information or more specifically, heart rate variability (HRV), for automatic deep sleep detection throughout the night. The HRV data can be derived from cardiac signals, which were obtained from polysomnography (PSG) recordings. In total 42 features were extracted from the HRV data of 15 single-night PSG recordings (from 15 healthy subjects) for each 30-s epoch, used to perform epoch-by-epoch classification of deep sleep and non-deep sleep (including wake state and all the other sleep stages except deep sleep). To reduce variation of cardiac physiology between subjects, we normalized each feature per subject using a simple Z-score normalization method by subtracting the mean and dividing by the standard deviation of the feature values. A correlation-based feature selection (CFS) method was employed to select informative features as well as removing feature redundancy and a linear discriminant (LD) classifier was applied for deep and non-deep sleep classification. Results show that the use of Z-score normalization can significantly improve the classification performance. A Cohen's Kappa coefficient of 0.42 and an overall accuracy of 81.3% based on a leave-one-subject-out cross-validation were achieved.

Sleep and wake classification with actigraphy and respiratory effort using dynamic warping.

This paper proposes the use of dynamic warping (DW) methods for improving automatic sleep and wake classification using actigraphy and respiratory effort. DW is an algorithm that finds an optimal nonlinear alignment between two series allowing scaling and shifting. It is widely used to quantify (dis)similarity between two series. To compare the respiratory effort between sleep and wake states by means of (dis)similarity, we constructed two novel features based on DW. For a given epoch of a respiratory effort recording, the features search for the optimally aligned epoch within the same recording in time and frequency domain. This is expected to yield a high (or low) similarity score when this epoch is sleep (or wake). Since the comparison occurs throughout the entire-night recording of a subject, it may reduce the effects of within- and between-subject variations of the respiratory effort, and thus help discriminate between sleep and wake states. The DW-based features were evaluated using a linear discriminant classifier on a dataset of 15 healthy subjects. Results show that the DW-based features can provide a Cohen's Kappa coefficient of agreement κ = 0.59 which is significantly higher than the existing respiratory-based features and is comparable to actigraphy. After combining the actigraphy and the DW-based features, the classifier achieved a κ of 0.66 and an overall accuracy of 95.7%, outperforming an earlier actigraphy- and respiratory-based feature set ( κ = 0.62). The results are also comparable with those obtained using an actigraphy- and cardiorespiratory-based feature set but have the important advantage that they do not require an ECG signal to be recorded.

A novel low-complexity post-processing algorithm for precise QRS localization.

Precise localization of QRS complexes is an essential step in the analysis of small transient changes in instant heart rate and before signal averaging in QRS morphological analysis. Most localization algorithms reported in literature are either not robust to artifacts, depend on the sampling rate of the ECG recordings or are too computationally expensive for real-time applications, especially in low-power embedded devices. This paper proposes a localization algorithm based on the intersection of tangents fitted to the slopes of R waves detected by any QRS detector. Despite having a lower complexity, this algorithm achieves comparable trigger jitter to more complex localization methods without requiring the data to first be upsampled. It also achieves high localization precision regardless of which QRS detector is used as input. It is robust to clipping artifacts and to noise, achieving an average localization error below 2 ms and a trigger jitter below 1 ms on recordings where no additional artifacts were added, and below 8 ms for recordings where the signal was severely degraded. Finally, it increases the accuracy of template-based false positive rejection, allowing nearly all mock false positives added to a set of QRS detections to be removed at the cost of a very small decrease in sensitivity. The localization algorithm proposed is particularly well-suited for implementation in embedded, low-power devices for real-time applications.

Respiration amplitude analysis for REM and NREM sleep classification.

In previous work, single-night polysomnography recordings (PSG) of respiratory effort and electrocardiogram (ECG) signals combined with actigraphy were used to classify sleep and wake states. In this study, we aim at classifying rapid-eye-movement (REM) and non-REM (NREM) sleep states. Besides the existing features used for sleep and wake classification, we propose a set of new features based on respiration amplitude. This choice is motivated by the observation that the breathing pattern has a more regular amplitude during NREM sleep than during REM sleep. Experiments were conducted with a data set of 14 healthy subjects using a linear discriminant (LD) classifier. Leave-one-subject-out cross-validations show that adding the new features into the existing feature set results in an increase in Cohen's Kappa coefficient to a value of κ = 0.59 (overall accuracy of 87.6%) compared to that obtained without using these features (κ of 0.54 and overall accuracy of 86.4%). In addition, we compared the results to those reported in some other studies with different features and signal modalities.

On the impact of arousals on the performance of sleep and wake classification using actigraphy.

We evaluated the impact of arousals on the performance of actigraphy-based sleep/wake classification. Using a dataset of 15 healthy adults and a threshold optimized for this task we found that the percentage of sleep epochs with activity counts above that threshold was significantly larger in epochs with and following arousals. We also found that 41.1% of all false positive classifications occurred in these epochs. Finally, we determined that excluding these epochs from the evaluation led to a maximum precision increase of 17.2%. Considering wake detections in those epochs as correct led to a maximum precision increase of 31.3%. We concluded that unless arousals can be automatically identified or at least distinguished from wake, the performance of actigraphy-based sleep/wake classifiers is limited by their presence.

Noncontact monitoring of cardiorespiratory activity by electromagnetic coupling.

In this paper, the method of noncontact monitoring of cardiorespiratory activity by electromagnetic coupling with human tissue is investigated. Two measurement modalities were joined: an inductive coupling sensor based on magnetic eddy current induction and a capacitive coupling sensor based on displacement current induction. The system's sensitivity to electric tissue properties and its dependence on motion are analyzed theoretically as well as experimentally for the inductive and capacitive coupling path. The potential of both coupling methods to assess respiration and pulse without contact and a minimum of thoracic wall motion was verified by laboratory experiments. The demonstrator was embedded in a chair to enable recording from the back part of the thorax.

A portable magnetic induction measurement system (PIMS).

For contactless monitoring of ventilation and heart activity, magnetic induction measurements are applicable. As the technique is harmless for the human body, it is well suited for long-term monitoring solutions, e.g., bedside monitoring, monitoring of home care patients, and the monitoring of persons in critical occupations. For such settings, a two-channel portable magnetic induction system has been developed, which is small and light enough to be fitted in a chair or bed. Because demodulation, control, and filtering are implemented on a front-end digital signal processor, a PC is not required (except for visualization/data storage during research and development). The system can be connected to a local area network (LAN) or wireless network (WiFi), allowing to connect several devices to a large monitoring system, e.g., for a residential home for the elderly or a hospital with low-risk patients not requiring standard ECG monitoring. To visualize data streams, a Qt-based (Qt-framework by Nokia, Espoo, Finland) monitoring application has been developed, which runs on Netbook computers, laptops, or standard PCs. To induce and measure the magnetic fields, external coils and amplifiers are required. This article describes the system and presents results for monitoring respiration and heart activity in a (divan) bed and for respiration monitoring in a chair. Planar configurations and orthogonal coil setups were examined during the measurement procedures. The measurement data were streamed over a LAN to a monitoring PC running Matlab (The MathWorks Inc, Natick, MA, USA).

Breathing detection with a portable impedance measurement system: first measurements.

For monitoring the health status of individuals, detection of breathing and heart activity is important. From an electrical point of view, it is known that breathing and heart activity change the electrical impedance distribution in the human body over the time due to ventilation (high impedance) and blood shifts (low impedance). Thus, it is possible to detect both important vital parameters by measuring the impedance of the thorax or the region around lung and heart. For some measurement scenarios it is also essential to detect these parameters contactless. For instance, monitoring bus drivers health could help to limit accidents, but directly connected systems limit the drivers free moving space. One measurement technology for measuring the impedance changes in the chest without cables is the magnetic impedance tomography (MIT). This article describes a portable measurement system we developed for this scenario that allows to measure breathing contactless. Furthermore, first measurements with five volunteers were performed and analyzed.