Christoph Brüser

Adaptive Beat-to-Beat Heart Rate Estimation in Ballistocardiograms

A ballistocardiograph records the mechanical activity of the heart. We present a novel algorithm for the detection of individual heart beats and beat-to-beat interval lengths in ballistocardiograms (BCGs) from healthy subjects. An automatic training step based on unsupervised learning techniques is used to extract the shape of a single heart beat from the BCG. Using the learned parameters, the occurrence of individual heart beats in the signal is detected. A final refinement step improves the accuracy of the estimated beat-to-beat interval lengths. Compared to many existing algorithms, the new approach offers heart rate estimates on a beat-to-beat basis. The agreement of the proposed algorithm with an ECG reference has been evaluated. A relative beat-to-beat interval error of 1.79% with a coverage of 95.94% was achieved on recordings from 16 subjects.

Robust Inter-beat Interval Estimation in Cardiac Vibration Signals

Reliable and accurate estimation of instantaneous frequencies of physiological rhythms, such as heart rate, is critical for many healthcare applications. Robust estimation is especially challenging when novel unobtrusive sensors are used for continuous health monitoring in uncontrolled environments, because these sensors can create significant amounts of potentially unreliable data. We propose a new flexible algorithm for the robust estimation of local (beat-to-beat) intervals from cardiac vibration signals, specifically ballistocardiograms (BCGs), recorded by an unobtrusive bed-mounted sensor. This sensor allows the measurement of motions of the body which are caused by cardiac activity. Our method requires neither a training phase nor any prior knowledge about the morphology of the heart beats in the analyzed waveforms. Instead, three short-time estimators are combined using a Bayesian approach to continuously estimate the inter-beat intervals. We have validated our method on over-night BCG recordings from 33 subjects (8 normal, 25 insomniacs). On this dataset, containing approximately one million heart beats, our method achieved a mean beat-to-beat interval error of 0.78% with a coverage of 72.69%.

Automatic Detection of Atrial Fibrillation in Cardiac Vibration Signals

We present a study on the feasibility of the automatic detection of atrial fibrillation (AF) from cardiac vibration signals (ballistocardiograms/BCGs) recorded by unobtrusive bed-mounted sensors. The proposed system is intended as a screening and monitoring tool in home-healthcare applications and not as a replacement for ECG-based methods used in clinical environments. Based on the BCG data recorded in a study with ten AF patients, we evaluate and rank seven popular machine learning algorithms (naive Bayes, linear and quadratic discriminant analysis, support vector machines, random forests as well as bagged and boosted trees) for their performance in separating 30-s long BCG epochs into one of three classes: sinus rhythm, AF, and artifact. For each algorithm, feature subsets of a set of statistical time-frequency-domain and time-domain features were selected based on the mutual information between features and class labels as well as the first- and second-order interactions among features. The classifiers were evaluated on a set of 856 epochs by means of tenfold cross validation. The best algorithm (random forests) achieved a Matthews correlation coefficient, mean sensitivity, and mean specificity of 0.921, 0.938, and 0.982, respectively.

Ambient and Unobtrusive Cardiorespiratory Monitoring Techniques

Monitoring vital signs through unobtrusive means is a goal which has attracted a lot of attention in the past decade. This review provides a systematic and comprehensive review over the current state of the field of ambient and unobtrusive cardiorespiratory monitoring. To this end, nine different sensing modalities which have been in the focus of current research activities are covered: capacitive electrocardiography, seismo- and ballistocardiography, reflective photoplethysmography (PPG) and PPG imaging, thermography, methods relying on laser or radar for distance-based measurements, video motion analysis, as well as methods using high-frequency electromagnetic fields. Current trends in these subfields are reviewed. Moreover, we systematically analyze similarities and differences between these methods with respect to the physiological and physical effects they sense as well as the resulting implications. Finally, future research trends for the field as a whole are identified.

Applying machine learning to detect individual heart beats in ballistocardiograms

Ballistocardiography is a technique in which the mechanical activity of the heart is recorded. We present a novel algorithm for the detection of individual heart beats in ballistocardiograms (BCGs). In a training step, unsupervised learning techniques are used to identify the shape of a single heart beat in the BCG. The learned parameters are combined with so-called “heart valve components” to detect the occurrence of individual heart beats in the signal. A refinement step improves the accuracy of the estimated beat-to-beat interval lengths. Compared to other algorithms this new approach offers heart rate estimates on a beat-to-beat basis and is designed to cope with arrhythmias. The proposed algorithm has been evaluated in laboratory and home settings for its agreement with an ECG reference. A beat-to-beat interval error of 14.16 ms with a coverage of 96.87% was achieved. Averaged over 10 s long epochs, the mean heart rate error was 0.39 bpm.

Robust Sensor Fusion of Unobtrusively Measured Heart Rate

Contactless vital sign measurement technologies often have the drawback of severe motion artifacts and periods in which no signal is available. However, using several identical or physically different sensors, redundancy can be used to decrease the error in noncontact heart rate estimation, while increasing the time period during which reliable data are available. In this paper, we show for the first time two major results in case of contactless heart rate measurements deduced from a capacitive ECG and optical pulse signals. First, an artifact detection is an essential preprocessing step to allow a reliable fusion. Second, the robust but computationally efficient median already provides good results; however, using a Bayesian approach, and a short time estimation of the variance, best results in terms of difference to reference heart rate and temporal coverage can be achieved. In this paper, six sensor signals were used and coverage increased from 0-90% to 80-94%, while the difference between the estimated heart rate and the gold standard was less than ±2 BPM.

Multi-channel optical sensor-array for measuring ballistocardiograms and respiratory activity in bed

Our work covers improvements in sensors and signal processing for unobtrusive, long-term monitoring of cardiac (and respiratory) rhythms using only non-invasive vibration sensors. We describe a system for the unobtrusive monitoring of vital signs by means of an array of novel optical ballistocardiography (BCG) sensors placed underneath a regular bed mattress. Furthermore, we analyze the systems spatial sensitivity and present proof-of-concept results comparing our system to a more conventional BCG system based on a single electromechanical-film (EMFi) sensor. Our preliminary results suggest that the proposed optical multi-channel system could have the potential to reduce beat-to-beat heart rate estimation errors, as well as enable the analysis of more complex breathing patterns.

Non-contact monitoring techniques - Principles and applications

This work gives an overview about some non-contact methods for monitoring of physiological activity. In particular, the focus is on ballistocardiography, capacitive ECG, Infrared Thermography, Magnetic Impedance Monitroing and Photoplethymographic Imaging. The principles behind the methods are described and an inside into possible medical applications is offered.

Improvement of Force-Sensor-Based Heart Rate Estimation Using Multichannel Data Fusion

The aim of this paper is to present and evaluate algorithms for heartbeat interval estimation from multiple spatially distributed force sensors integrated into a bed. Moreover, the benefit of using multichannel systems as opposed to a single sensor is investigated. While it might seem intuitive that multiple channels are superior to a single channel, the main challenge lies in finding suitable methods to actually leverage this potential. To this end, two algorithms for heart rate estimation from multichannel vibration signals are presented and compared against a single-channel sensing solution. The first method operates by analyzing the cepstrum computed from the average spectra of the individual channels, while the second method applies Bayesian fusion to three interval estimators, such as the autocorrelation, which are applied to each channel. This evaluation is based on 28 night-long sleep lab recordings during which an eight-channel polyvinylidene fluoride-based sensor array was used to acquire cardiac vibration signals. The recruited patients suffered from different sleep disorders of varying severity. From the sensor array data, a virtual single-channel signal was also derived for comparison by averaging the channels. The single-channel results achieved a beat-to-beat interval error of 2.2% with a coverage (i.e., percentage of the recording which could be analyzed) of 68.7%. In comparison, the best multichannel results attained a mean error and coverage of 1.0% and 81.0%, respectively. These results present statistically significant improvements of both metrics over the single-channel results (p <; 0.05).

Beat-to-beat heart rate estimation fusing multimodal video and sensor data

Coverage and accuracy of unobtrusively measured biosignals are generally relatively low compared to clinical modalities. This can be improved by exploiting redundancies in multiple channels with methods of sensor fusion. In this paper, we demonstrate that two modalities, skin color variation and head motion, can be extracted from the video stream recorded with a webcam. Using a Bayesian approach, these signals are fused with a ballistocardiographic signal obtained from the seat of a chair with a mean absolute beat-to-beat estimation error below 25 milliseconds and an average coverage above 90% compared to an ECG reference.