Multiple Instance Dictionary Learning for Beat-to-Beat Heart Rate Monitoring From Ballistocardiograms

CHMMConvScaleNet: a hybrid convolutional neural network and continuous hidden Markov model with multi-scale features for sleep posture detection

Sleep posture, a vital aspect of sleep wellness, has become a crucial focus in sleep medicine. Studies show that supine posture can lead to a higher occurrence of obstructive sleep apnea, while lateral posture might reduce it. For bedridden patients, frequent posture changes are essential to prevent ulcers and bedsores, highlighting the importance of monitoring sleep posture. This paper introduces CHMMConvScaleNet, a novel method for sleep posture recognition using pressure signals from limited piezoelectric ceramic sensors. It employs a Movement Artifact and Rollover Identification (MARI) module to detect critical rollover events and extracts multi-scale spatiotemporal features using six sub-convolution networks with different-length adjacent segments. To optimize performance, a Continuous Hidden Markov Model (CHMM) with rollover features is presented. We collected continuous real sleep data from 22 participants, yielding a total of 8583 samples from a 32-sensor array. Experiments show that CHMMConvScaleNet achieves a recall of 92.91%, precision of 91.87%, and accuracy of 93.41%, comparable to state-of-the-art methods that require ten times more sensors to achieve a slightly improved accuracy of 96.90% on non-continuous datasets. Thus, CHMMConvScaleNet demonstrates potential for home sleep monitoring using portable devices.

Self-Supervised, Non-Contact Heartbeat Detection Based on Ballistocardiograms Utilizing Physiological Information Guidance

Ballistocardiograms (BCG) is a passive, non-contact heart rate detection technology that requires no action on the part of the individual. However, during the BCG signal acquisition process, the surface pressure generated by cardiac contraction is easily disturbed by external factors, and as people's health deteriorates, the j-peak (the main peak of the BCG signal) is no longer prominent. Our aim is to establish a non-contact, self-supervised heart rate detection method based on physiological information, to improve the accuracy and robustness of BCG heart rate detection under wider and more adverse conditions. The algorithm is guided by the heart rate estimation based on BCG itself, thereby reconstructing a signal with physiological significance. We also propose a heartbeat mapping algorithm based on Bidirectional Long Short-Term Memory Network (BiLSTM) for extracting global deep features, achieving real-time heartbeat prediction, and eliminating local deviations brought about by reconstruction. To verify the effectiveness of the proposed method, this paper evaluated 40 young subjects and 4 elderly subjects. Compared with the existing state-of-the-art methods, beat-to-beat heart rate estimation and heartbeat detection both performed excellently, surpassing most methods using precise labels. The experimental results show that the proposed method achieves effective heartbeat detection, demonstrating robustness and effectiveness in the face of unavoidable noise and variations.

Smart Sleep Monitoring: Sparse Sensor-Based Spatiotemporal CNN for Sleep Posture Detection

Sleep quality is heavily influenced by sleep posture, with research indicating that a supine posture can worsen obstructive sleep apnea (OSA) while lateral postures promote better sleep. For patients confined to beds, regular changes in posture are crucial to prevent the development of ulcers and bedsores. This study presents a novel sparse sensor-based spatiotemporal convolutional neural network (S3CNN) for detecting sleep posture. This S3CNN holistically incorporates a pair of spatial convolution neural networks to capture cardiorespiratory activity maps and a pair of temporal convolution neural networks to capture the heart rate and respiratory rate. Sleep data were collected in actual sleep conditions from 22 subjects using a sparse sensor array. The S3CNN was then trained to capture the spatial pressure distribution from the cardiorespiratory activity and temporal cardiopulmonary variability from the heart and respiratory data. Its performance was evaluated using three rounds of 10 fold cross-validation on the 8583 data samples collected from the subjects. The results yielded 91.96% recall, 92.65% precision, and 93.02% accuracy, which are comparable to the state-of-the-art methods that use significantly more sensors for marginally enhanced accuracy. Hence, the proposed S3CNN shows promise for sleep posture monitoring using sparse sensors, demonstrating potential for a more cost-effective approach.

A Review of Patient Bed Sensors for Monitoring of Vital Signs

The analysis of biomedical signals is a very challenging task. This review paper is focused on the presentation of various methods where biomedical data, in particular vital signs, could be monitored using sensors mounted to beds. The presented methods to monitor vital signs include those combined with optical fibers, camera systems, pressure sensors, or other sensors, which may provide more efficient patient bed monitoring results. This work also covers the aspects of interference occurrence in the above-mentioned signals and sleep quality monitoring, which play a very important role in the analysis of biomedical signals and the choice of appropriate signal-processing methods. The provided information will help various researchers to understand the importance of vital sign monitoring and will be a thorough and up-to-date summary of these methods. It will also be a foundation for further enhancement of these methods.

A semi-supervised approach to unobtrusively predict abnormality in breathing patterns using hydraulic bed sensor data in older adults aging in place

Shortness of breath is often considered a repercussion of aging in older adults, as respiratory illnesses like COPD1 or respiratory illnesses due to heart-related issues are often misdiagnosed, under-diagnosed or ignored at early stages. Continuous health monitoring using ambient sensors has the potential to ameliorate this problem for older adults at aging-in-place facilities. In this paper, we leverage continuous respiratory health data collected by using ambient hydraulic bed sensors installed in the apartments of older adults in aging-in-place Americare facilities to find data-adaptive indicators related to shortness of breath. We used unlabeled data collected unobtrusively over the span of three years from a COPD-diagnosed individual and used data mining to label the data. These labeled data are then used to train a predictive model to make future predictions in older adults related to shortness of breath abnormality. To pick the continuous changes in respiratory health we make predictions for shorter time windows (60-s). Hence, to summarize each day's predictions we propose an abnormal breathing index (ABI) in this paper. To showcase the trajectory of the shortness of breath abnormality over time (in terms of days), we also propose trend analysis on the ABI quarterly and incrementally. We have evaluated six individual cases retrospectively to highlight the potential and use cases of our approach.

Monitoring of Cardiorespiratory Parameters during Sleep Using a Special Holder for the Accelerometer Sensor

Sleep is extremely important for physical and mental health. Although polysomnography is an established approach in sleep analysis, it is quite intrusive and expensive. Consequently, developing a non-invasive and non-intrusive home sleep monitoring system with minimal influence on patients, that can reliably and accurately measure cardiorespiratory parameters, is of great interest. The aim of this study is to validate a non-invasive and unobtrusive cardiorespiratory parameter monitoring system based on an accelerometer sensor. This system includes a special holder to install the system under the bed mattress. The additional aim is to determine the optimum relative system position (in relation to the subject) at which the most accurate and precise values of measured parameters could be achieved. The data were collected from 23 subjects (13 males and 10 females). The obtained ballistocardiogram signal was sequentially processed using a sixth-order Butterworth bandpass filter and a moving average filter. As a result, an average error (compared to reference values) of 2.24 beats per minute for heart rate and 1.52 breaths per minute for respiratory rate was achieved, regardless of the subject's sleep position. For males and females, the errors were 2.28 bpm and 2.19 bpm for heart rate and 1.41 rpm and 1.30 rpm for respiratory rate. We determined that placing the sensor and system at chest level is the preferred configuration for cardiorespiratory measurement. Further studies of the system's performance in larger groups of subjects are required, despite the promising results of the current tests in healthy subjects.

HTIDB: Hierarchical Time-Indexed Database for Efficient Storage and Access to Irregular Time-series Health Sensor Data

With the enormous amount of data collected by unobtrusive sensors, the potential of utilizing these data and applying various multi-modal advanced analytics on them is numerous and promising. However, taking advantage of the ever-growing data requires high-performance data-handling systems to enable high data scalability and easy data accessibility. This paper demonstrates robust design, developments, and techniques of a hierarchical time-indexed database for decision support systems leveraging irregular and sporadic time series data from sensor systems, e.g., wearables or environmental. We propose a technique that leverages the flexibility of general purpose, high-scalability database systems, while integrating data analytics focused column stores that leverage hierarchical time indexing, compression, and dense raw numeric data storage. We have evaluated the performance characteristics and tradeoffs of each to understand the data access latencies and storage requirements, which are key elements for capacity planning for scalable systems.

Mechanism-Driven Modeling to Aid Non-invasive Monitoring of Cardiac Function via Ballistocardiography

Left ventricular (LV) catheterization provides LV pressure-volume (P-V) loops and it represents the gold standard for cardiac function monitoring. This technique, however, is invasive and this limits its applicability in clinical and in-home settings. Ballistocardiography (BCG) is a good candidate for non-invasive cardiac monitoring, as it is based on capturing non-invasively the body motion that results from the blood flowing through the cardiovascular system. This work aims at building a mechanistic connection between changes in the BCG signal, changes in the P-V loops and changes in cardiac function. A mechanism-driven model based on cardiovascular physiology has been used as a virtual laboratory to predict how changes in cardiac function will manifest in the BCG waveform. Specifically, model simulations indicate that a decline in LV contractility results in an increase of the relative timing between the ECG and BCG signal and a decrease in BCG amplitude. The predicted changes have subsequently been observed in measurements on three swine serving as pre-clinical models for pre- and post-myocardial infarction conditions. The reproducibility of BCG measurements has been assessed on repeated, consecutive sessions of data acquisitions on three additional swine. Overall, this study provides experimental evidence supporting the utilization of mechanism-driven mathematical modeling as a guide to interpret changes in the BCG signal on the basis of cardiovascular physiology, thereby advancing the BCG technique as an effective method for non-invasive monitoring of cardiac function.

ResNet-BiLSTM: A Multiscale Deep Learning Model for Heartbeat Detection Using Ballistocardiogram Signals

As the heartbeat detection from ballistocardiogram (BCG) signals using force sensors is interfered by respiratory effort and artifact motion, advanced signal processing algorithms are required to detect the J-peak of each BCG signal so that beat-to-beat interval can be identified. However, existing methods generally rely on rule-based detection of a fixed size, without considering the rhythm features in a large time scale covering multiple BCG signals. Methods. This paper develops a deep learning framework based on ResNet and bidirectional long short-term memory (BiLSTM) to conduct beat-to-beat detection of BCG signals. Unlike the existing methods, the proposed network takes multiscale features of BCG signals as the input and, thus, can enjoy the complementary advantages of both morphological features of one BCG signal and rhythm features of multiple BCG signals. Different time scales of multiscale features for the proposed model are validated and analyzed through experiments. Results. The BCG signals recorded from 21 healthy subjects are conducted to verify the performance of the proposed heartbeat detection scheme using leave-one-out cross-validation. The impact of different time scales on the detection performance and the performance of the proposed model for different sleep postures are examined. Numerical results demonstrate that the proposed multiscale model performs robust to sleep postures and achieves an averaged absolute error (E abs) and an averaged relative error (E rel) of the heartbeat interval relative to the R-R interval of 9.92 ms and 2.67 ms, respectively, which are superior to those of the state-of-the-art detection protocol. Conclusion. In this work, a multiscale deep-learning model for heartbeat detection using BCG signals is designed. We demonstrate through the experiment that the detection with multiscale features of BCG signals can provide a superior performance to the existing works. Further study will examine the ultimate performance of the multiscale model in practical scenarios, i.e., detection for patients suffering from cardiovascular disorders with night-sleep monitoring.

Convolutional Autoencoding and Gaussian Mixture Clustering for Unsupervised Beat-to-Beat Heart Rate Estimation of Electrocardiograms from Wearable Sensors

Heart rate is one of the most important diagnostic bases for cardiovascular disease. This paper introduces a deep autoencoding strategy into feature extraction of electrocardiogram (ECG) signals, and proposes a beat-to-beat heart rate estimation method based on convolution autoencoding and Gaussian mixture clustering. The high-level heartbeat features were first extracted in an unsupervised manner by training the convolutional autoencoder network, and then the adaptive Gaussian mixture clustering was applied to detect the heartbeat locations from the extracted features, and calculated the beat-to-beat heart rate. Compared with the existing heartbeat classification/detection methods, the proposed unsupervised feature learning and heartbeat clustering method does not rely on accurate labeling of each heartbeat location, which could save a lot of time and effort in human annotations. Experimental results demonstrate that the proposed method maintains better accuracy and generalization ability compared with the existing ECG heart rate estimation methods and could be a robust long-time heart rate monitoring solution for wearable ECG devices.

A Practical Application for Quantitative Brain Fatigue Evaluation Based on Machine Learning and Ballistocardiogram

Brain fatigue is often associated with inattention, mental retardation, prolonged reaction time, decreased work efficiency, increased error rate, and other problems. In addition to the accumulation of fatigue, brain fatigue has become one of the important factors that harm our mental health. Therefore, it is of great significance to explore the practical and accurate brain fatigue detection method, especially for quantitative brain fatigue evaluation. In this study, a biomedical signal of ballistocardiogram (BCG), which does not require direct contact with human body, was collected by optical fiber sensor cushion during the whole process of cognitive tasks for 20 subjects. The heart rate variability (HRV) was calculated based on BCG signal. Machine learning classification model was built based on random forest to quantify and recognize brain fatigue. The results showed that: Firstly, the heart rate obtained from BCG signal was consistent with the result displayed by the medical equipment, and the absolute difference was less than 3 beats/min, and the mean error is 1.30 ± 0.81 beats/min; secondly, the random forest classifier for brain fatigue evaluation based on HRV can effectively identify the state of brain fatigue, with an accuracy rate of 96.54%; finally, the correlation between HRV and the accuracy was analyzed, and the correlation coefficient was as high as 0.98, which indicates that the accuracy can be used as an indicator for quantitative brain fatigue evaluation during the whole task. The results suggested that the brain fatigue quantification evaluation method based on the optical fiber sensor cushion and machine learning can carry out real-time brain fatigue detection on the human brain without disturbance, reduce the risk of human accidents in human–machine interaction systems, and improve mental health among the office and driving personnel.

Non-Invasive Heart Rate Estimation From Ballistocardiograms Using Bidirectional LSTM Regression

Non-invasive heart rate estimation is of great importance in daily monitoring of cardiovascular diseases. In this paper, a bidirectional long short term memory (bi-LSTM) regression network is developed for non-invasive heart rate estimation from the ballistocardiograms (BCG) signals. The proposed deep regression model provides an effective solution to the existing challenges in BCG heart rate estimation, such as the mismatch between the BCG signals and ground-truth reference, multi-sensor fusion and effective time series feature learning. Allowing label uncertainty in the estimation can reduce the manual cost of data annotation while further improving the heart rate estimation performance. Compared with the state-of-the-art BCG heart rate estimation methods, the strong fitting and generalization ability of the proposed deep regression model maintains better robustness to noise (e.g., sensor noise) and perturbations (e.g., body movements) in the BCG signals and provides a more reliable solution for long term heart rate monitoring.

Accurate Ballistocardiogram Based Heart Rate Estimation Using an Array of Load Cells in a Hospital Bed

The ballistocardiogram (BCG), a cardiac vibration signal, has been widely investigated for continuous monitoring of heart rate (HR). Among BCG sensing modalities, a hospital bed with multi-channel load-cells could provide robust HR estimation in hospital setups. In this work, we present a novel array processing technique to improve the existing HR estimation algorithm by optimizing the fusion of information from multiple channels. The array processing includes a Gaussian curve to weight the joint probability according to the reference value obtained from the previous inter-beat-interval (IBI) estimations. Additionally, the probability density functions were selected and combined according to their reliability measured by q-values. We demonstrate that this array processing significantly reduces the HR estimation error compared to state-of-the-art multi-channel heartbeat detection algorithms in the existing literature. In the best case, the average mean absolute error (MAE) of 1.76 bpm in the supine position was achieved compared to 2.68 bpm and 1.91 bpm for two state-of-the-art methods from the existing literature. Moreover, the lowest error was found in the supine posture (1.76 bpm) and the highest in the lateral posture (3.03 bpm), thus elucidating the postural effects on HR estimation. The IBI estimation capability was also evaluated, with a MAE of 16.66 ms and confidence interval (95%) of 38.98 ms. The results demonstrate that improved HR estimation can be obtained for a bed-based BCG system with the multi-channel data acquisition and processing approach described in this work.

A review of wearable and unobtrusive sensing technologies for chronic disease management

With the rapidly increasing number of patients with chronic disease, numerous recent studies have put great efforts into achieving long-term health monitoring and patient management. Specifically, chronic diseases including cardiovascular disease, chronic respiratory disease and brain disease can threaten patients' health conditions over a long period of time, thus effecting their daily lives. Vital health parameters, such as heart rate, respiratory rate, SpO2 and blood pressure, are closely associated with patients’ conditions. Wearable devices and unobtrusive sensing technologies can detect such parameters in a convenient way and provide timely predictions on health condition deterioration by tracking these biomedical signals and health parameters. In this paper, we review current advancements in wearable devices and unobtrusive sensing technologies that can provides possible tools and technological supports for chronic disease management. Current challenges and future directions of related techniques are addressed accordingly.

Human-centered approaches that integrate sensor technology across the lifespan: Opportunities and challenges

Children, parents, older adults, and caregivers routinely use sensor technology as a source of health information and health monitoring. The purpose of this paper is to describe three exemplars of research that used a human-centered approach to engage participants in the development, design, and usability of interventions that integrate technology to promote health. The exemplars are based on current research studies that integrate sensor technology into pediatric, adult, and older adult populations living with a chronic health condition. Lessons learned and considerations for future studies are discussed. Nurses have successfully implemented interventions that use technology to improve health and detect, prevent, and manage diseases in children, families, individuals and communities. Nurses are key stakeholders to inform clinically relevant health monitoring that can support timely and personalized intervention and recommendations.

Heartbeat Detection and Rate Estimation from Ballistocardiograms using the Gated Recurrent Unit Network

Inspired by the application of recurrent neural networks (RNNs) to image recognition, in this paper, we propose a heartbeat detection framework based on the Gated Recurrent Unit (GRU) network. In this contribution, the heartbeat detection task from ballistocardiogram (BCG) signals was modeled as a classification problem where the segments of BCG signals were formulated as images fed into the GRU network for feature extraction. The proposed framework has advantages in fusion of multi-channel BCG signals and effective extraction of the temporal and waveform characteristics of the heartbeat signal, thereby enhancing heart rate estimation accuracy. In laboratory collected BCG data, the proposed method achieved the best heart rate estimation results compared to previous algorithms.

Heart rate detection with accelerometric sensors under the mattress

The ballistocardiography is a technique that measures the heart rate from the mechanical vibrations of the body due to the heart movement. In this work a novel noninvasive device placed under the mattress of a bed estimates the heart rate using the ballistocardiography. Different algorithms for heart rate estimation have been developed.

A Correlation-Based Algorithm for Beat-to-Beat Heart Rate Estimation from Ballistocardiograms

Ballistocardiography (BCG) is a type of non-contact measurement technique that measures the mechanical reaction of the body resulting from heart contraction and the subsequent cardiac ejection of blood. Herein, we present an algorithm for beat-to-beat heart rate estimation from BCG signals that is both highly universal and easy to implement. The algorithm is based on the correlation between heartbeats in the same section of BCG. It first generates patterns by autocorre-lation, which are then matched with the remaining signals to determine heartbeats. The agreement of the proposed algorithm with synchronized electrocardiogram has been evaluated, and a relative beat-to-beat interval error of 1.66% and a relative average heart rate error of 1.25% were observed. The proposed algorithm is a promising candidate for a non-contact, long-term cardiac monitoring system at home.

Multiple Instance Learning for Multiple Diverse Hyperspectral Target Characterizations

A practical hyperspectral target characterization task estimates a target signature from imprecisely labeled training data. The imprecisions arise from the characteristics of the real-world tasks. First, accurate pixel-level labels on training data are often unavailable. Second, the subpixel targets and occluded targets cause the training samples to contain mixed data and multiple target types. To address these imprecisions, this paper proposes a new hyperspectral target characterization method to produce diverse multiple hyperspectral target signatures under a multiple instance learning (MIL) framework. The proposed method uses only bag-level training samples and labels, which solves the problems arising from the mixed data and lack of pixel-level labels. Moreover, by formulating a multiple characterization MIL and including a diversity-promoting term, the proposed method can learn a set of diverse target signatures, which solves the problems arising from multiple target types in training samples. The experiments on hyperspectral target detections using the learned multiple target signatures over synthetic and real-world data show the effectiveness of the proposed method.

Phyjama

Unobtrusive and continuous monitoring of cardiac and respiratory rhythm, especially during sleeping, can have significant clinical utility. An exciting new possibility for such monitoring is the design of textiles that use all-textile sensors that can be woven or stitched directly into a textile or garment. Our work explores how we can make such monitoring possible by leveraging something that is already familiar, such as pyjama made of cotton/silk fabric, and imperceptibly adapt it to enable sensing of physiological signals to yield natural fitting, comfortable, and less obtrusive smart clothing.

We face several challenges in enabling this vision including requiring new sensor design to measure physiological signals via everyday textiles and new methods to deal with the inherent looseness of normal garments, particularly sleepwear like pyjamas. We design two types of textile-based sensors that obtain a ballistic signal due to cardiac and respiratory rhythm ---the first a novel resistive sensor that leverages pressure between the body and various surfaces and the second is a triboelectric sensor that leverages changes in separation between layers to measure ballistics induced by the heart. We then integrate several instances of such sensors on a pyjama and design a signal processing pipeline that fuses information from the different sensors such that we can robustly measure physiological signals across a range of sleep and stationary postures. We show that the sensor and signal processing pipeline has high accuracy by benchmarking performance both under restricted settings with twenty one users as well as more naturalistic settings with seven users.