Laurent Giovangrandi

Robust Neural-Network-Based Classification of Premature Ventricular Contractions Using Wavelet Transform and Timing Interval Features

Automatic electrocardiogram (ECG) beat classification is essential to timely diagnosis of dangerous heart conditions. Specifically, accurate detection of premature ventricular contractions (PVCs) is imperative to prepare for the possible onset of life-threatening arrhythmias. Although many groups have developed highly accurate algorithms for detecting PVC beats, results have generally been limited to relatively small data sets. Additionally, many of the highest classification accuracies (>90%) have been achieved in experiments where training and testing sets overlapped significantly. Expanding the overall data set greatly reduces overall accuracy due to significant variation in ECG morphology among different patients. As a result, we believe that morphological information must be coupled with timing information, which is more constant among patients, in order to achieve high classification accuracy for larger data sets. With this approach, we combined wavelet-transformed ECG waves with timing information as our feature set for classification. We used select waveforms of 18 files of the MIT/BIH arrhythmia database, which provides an annotated collection of normal and arrhythmic beats, for training our neural-network classifier. We then tested the classifier on these 18 training files as well as 22 other files from the database. The accuracy was 95.16% over 93,281 beats from all 40 files, and 96.82% over the 22 files outside the training set in differentiating normal, PVC, and other beats

Robust ballistocardiogram acquisition for home monitoring

The ballistocardiogram (BCG) measures the reaction of the body to cardiac ejection forces, and is an effective, non-invasive means of evaluating cardiovascular function. A simple, robust method is presented for acquiring high-quality, repeatable BCG signals from a modified, commercially available scale. The measured BCG waveforms for all subjects qualitatively matched values in the existing literature and physiologic expectations in terms of timing and IJ amplitude. Additionally, the BCG IJ amplitude was shown to be correlated with diastolic filling time for a subject with premature atrial contractions, demonstrating the sensitivity of the apparatus to beat-by-beat hemodynamic changes. The signal-to-noise ratio (SNR) of the BCG was estimated using two methods, and the average SNR over all subjects was greater than 12 for both estimates. The BCG measurement was shown to be repeatable over 50 recordings taken from the same subject over a three week period. This approach could allow patients at home to monitor trends in cardiovascular health.

Lifeguard - a personal physiological monitor for extreme environments

Monitoring vital signs in applications that require the subject to be mobile requires small, lightweight, and robust sensors and electronics. A body-worn system should be unobtrusive, noninvasive, and easy-to-use. It must be able to log vital signs data for several hours as well as transmit it on demand in real-time using secure wireless technologies. The NASA Ames Research Center (Astrobionics) and Stanford University (National Center for Space Biological Technologies) are currently developing a wearable physiological monitoring system for astronauts, called LifeGuard, that meets all of the above requirements and is also applicable to clinical, home-health monitoring, first responder and military applications.

General purpose, field-portable cell-based biosensor platform

There are several groups of researchers developing cell-based biosensors for chemical and biological warfare agents based on electrophysiologic monitoring of cells. In order to transition such sensors from the laboratory to the field, a general-purpose hardware and software platform is required. This paper describes the design, implementation, and field-testing of such a system, consisting of cell-transport and data acquisition instruments. The cell-transport module is a self-contained, battery-powered instrument that allows various types of cell-based modules to be maintained at a preset temperature and ambient CO(2) level while in transit or in the field. The data acquisition module provides 32 channels of action potential amplification, filtering, and real-time data streaming to a laptop computer. At present, detailed analysis of the data acquired is carried out off-line, but sufficient computing power is available in the data acquisition module to enable the most useful algorithms to eventually be run real-time in the field. Both modules have sufficient internal power to permit realistic field-testing, such as the example presented in this paper.

Non-invasive cardiac output trending during exercise recovery on a bathroom-scale-based ballistocardiograph.

Cardiac ejection of blood into the aorta generates a reaction force on the body that can be measured externally via the ballistocardiogram (BCG). In this study, a commercial bathroom scale was modified to measure the BCGs of nine healthy subjects recovering from treadmill exercise. During the recovery, Doppler echocardiogram signals were obtained simultaneously from the left ventricular outflow tract of the heart. The percentage changes in root-mean-square (RMS) power of the BCG were strongly correlated with the percentage changes in cardiac output measured by Doppler echocardiography (R(2) = 0.85, n = 275 data points). The correlation coefficients for individually analyzed data ranged from 0.79 to 0.96. Using Bland-Altman methods for assessing agreement, the mean bias was found to be -0.5% (+/-24%) in estimating the percentage changes in cardiac output. In contrast to other non-invasive methods for trending cardiac output, the unobtrusive procedure presented here uses inexpensive equipment and could be performed without the aid of a medical professional.

Rapid Assessment of Cardiac Contractility on a Home Bathroom Scale

Analyzing systolic time intervals-specifically the preejection-period (PEP)-is widely accepted as one of the few methods for the noninvasive assessment of cardiac contractility. In this paper, we investigated the ballistocardiogram (BCG) as a way to noninvasively measure myocardial contractility when combined with the ECG. Specifically, we derived a parameter from the BCG and ECG that we hypothesized would be highly correlated to PEP. This is the time delay between the J-wave peak of the BCG and the R-wave of the ECG, which we refer to as the RJ interval. The RJ interval was correlated to PEP (r2 = 0.86) for 2126 heart beats across ten subjects, with a y-intercept of 138 ms and slope of 1.05. This suggests that the RJ interval can be reliably used as a noninvasive assessment of cardiac contractility.

Optical Scanner for Immunoassays With Up-Converting Phosphorescent Labels

A 2-D optical scanner was developed for the imaging and quantification of up-converting phosphor (UCP) labels in immunoassays. With resolution better than 500 mum, a scan rate of 0.4 mm/s, and a 1-2% coefficient of variation for repeatability, this scanner achieved a detection limit of fewer than 100 UCP particles in an 8.8 times 104 mum2 area and a dynamic range that covered more than three orders of magnitude. Utilizing this scanner, a microfluidic chip immunoassay for the cytokine interferon-gamma (IFN-gamma) was developed: concentrations as low as 3 pM (50 pg/mL) were detected from 100 muL samples with a total assay time of under an hour, including the 8 min readout. For this UCP-based assay, 2-D images of the capture antibody lines were scanned, image processing techniques were employed to extract the UCP emission signals, a response curve that spanned 3-600 pM IFN-gamma was generated, and a five-parameter logistic mathematical model was fitted to the data for determination of unknown IFN-gamma concentrations. Relative to common single-point or 1-D scanning optical measurements, our results suggest that a simple 2-D imaging system can speed assay development, reduce errors, and improve accuracy by characterizing the spatial distribution and uniformity of surface-captured optical labels as a function of assay conditions and device parameters.

Microfluidic impedance cytometer for platelet analysis

We present the design and performance characteristics of a platelet analysis platform based on a microfluidic impedance cytometer. Dielectrophoretic focusing is used to centre cells in a fluid stream, which then forms the core of a two-phase flow (dielectric focusing). This flow then passes between electrodes for analysis by differential impedance spectroscopy at multiple frequencies from 280 kHz to 4 MHz. This approach increases the signal-to-noise ratio relative to a single-phase, unfocused stream, while minimising the shear forces to which the cells are subjected. The percentage of activated platelets before and after passage through the chip was measured using flow cytometry, and no significant change was measured. Measuring the in-phase amplitude at a single frequency is sufficient to distinguish platelets from erythrocytes. Using multi-frequency impedance measurements and discriminant analysis, resting platelets can be discriminated from activated platelets. This multifrequency impedance cytometer therefore allows ready determination of the degree of platelet activation in blood samples.

Novel methods for estimating the ballistocardiogram signal using a simultaneously acquired electrocardiogram

The ballistocardiogram (BCG) signal represents the movements of the body in response to cardiac ejection of blood. Recently, many groups have developed low-cost instrumentation for facilitating BCG measurement in the home. The standard method used in the literature for estimating the BCG pulse response has generally been ensemble averaging over several beats. Unfortunately, since the BCG pulse response is likely longer than a typical heartbeat interval, this standard approach does not yield a full-length estimate of the response. This paper describes a simple, novel algorithm for estimating the full-length BCG pulse response using the R-wave timing of a simultaneously acquired electrocardiogram (ECG). With this pulse response, the full signal can be reconstructed, enabling the analysis of slow transient effects in the BCG signal, and of the measurement noise. Additionally, while this paper focuses only on the BCG signal, the same algorithm could be applied to other biomedical signals such as the phonocardiogram or impedance cardiogram, particularly when the heartbeat interval is shorter than the duration of the cpulse response.

Motion artifact cancellation to obtain heart sounds from a single chest-worn accelerometer

This paper presents a method of extracting primary heart sound signals from chest-worn accelerometer data in the presence of motion artifacts. The proposed method outperforms noise removal techniques such as wavelet denoising and adaptive filtering. Results from six subjects show a primary heart signal detection rate of 99.36% with a false positive rate of 1.3%.