Sven Ole Aase

Multi-frame compression: theory and design

This paper consist of two parts. The first part concerns approximation capabilities in using an overcomplete dictionary, a frame, for block coding. A frame design technique for use with vector selection algorithms, for example matching pursuits (MP), is presented. We call the technique method of optimal directions (MOD). It is iterative and requires a training set of signal vectors. Experiments demonstrate that the approximation capabilities of the optimized frames are significantly better than those obtained using frames designed by ad hoc techniques or chosen in an ad hoc fashion. Experiments show typical reduction in mean squared error (MSE) by 30–80% for speech and electrocardiogram (ECG) signals. The second part concerns a complete compression scheme using a set of optimized frames, and evaluates both the use of fixed size and variable size frames. A signal compression scheme using frames optimized with the MOD technique is proposed. The technique, called multi-frame compression (MFC) uses several different frames, each optimized for a fixed number of selected frame vectors in each approximation. We apply the MOD and the MFC scheme to ECG signals. The coding results are compared with results obtained when using transform-based compression schemes like the discrete cosine transform (DCT) in combination with run-length and entropy coding. The experiments demonstrate improved rate-distortion performance by 2–4 dB for the MFC scheme when compared to the DCT at low bit-rates. They also show that variable sized frames in the compression scheme perform better than fixed sized frames.

Predicting Outcome of Defibrillation by Spectral Characterization and Nonparametric Classification of Ventricular Fibrillation in Patients With Out-of-Hospital Cardiac Arrest

BackgroundIn 156 patients with out-of-hospital cardiac arrest of cardiac cause, we analyzed the ability of 4 spectral features of ventricular fibrillation before a total of 868 shocks to discriminate or not between segments that correspond to return of spontaneous circulation (ROSC). Methods and ResultsCentroid frequency, peak power frequency, spectral flatness, and energy were studied. A second decorrelated feature set was generated with the coefficients of the principal component analysis transformation of the original feature set. Each feature set was split into training and testing sets for improved reliability in the evaluation of nonparametric classifiers for each possible feature combination. The combination of centroid frequency and peak power frequency achieved a mean±SD sensitivity of 92±2% and specificity of 27±2% in testing. The highest performing classifier corresponded to the combination of the 2 dominant decorrelated spectral features with sensitivity and specificity equal to 92±2% and 42±1% in testing or a positive predictive value of 0.15 and a negative predictive value of 0.98. Using the highest performing classifier, 328 of 781 shocks not leading to ROSC would have been avoided, whereas 7 of 87 shocks leading to ROSC would not have been administered. ConclusionsThe ECG contained information predictive of shock therapy. This could reduce the delivery of unsuccessful shocks and thereby the duration of unnecessary “hands-off” intervals during cardiopulmonary resuscitation. The low specificity and positive predictive value indicate that other features should be added to improve performance.

Compression depth estimation for CPR quality assessment using DSP on accelerometer signals

Chest compression is a vital part of cardiopulmonary resuscitation (CPR). This paper demonstrates how the compression depth can be estimated using the principles of inertia navigation. The proposed method uses accelerometer sensors, one placed on the patients chest, the other beside the patient. The acceleration-to-position conversion is performed using discrete-time digital signal processing (DSP). Instability problems due to integration are combated using a set of boundary conditions. The proposed algorithm is tested on a mannequin in harsh environments, where the patient is exposed to external forces as in a boat or car, as well as improper sensor/patient alignment. The overall performance is an estimation depth error of 4.3 mm in these environments, which is reduced to 1.6 mm in a regular, flat-floor controlled environment.

Frame based signal compression using method of optimal directions (MOD)

The method of optimal directions (MOD) is an iterative method for designing frames for sparse representation purposes using a training set. In this paper we use frames designed by MOD in a multiframe compression (MFC) scheme. Both the MOD and the MFC need a vector selection algorithm, and orthogonal matching pursuit (OMP) is used in this paper. In the MFC scheme several different frames are used, each optimized for a fixed number of selected frame vectors in each approximation. We apply the MOD and the MFC scheme to ECG signals, and do experiments with both fixed size and variable size on the different frames used in the MFC scheme. Compared to traditional transform based compression, the experiments demonstrate improved rate-distortion performance by 1-4 dB, and that variable sized frames perform better than fixed sized frames.

Removal of cardiopulmonary resuscitation artifacts from human ECG using an efficient matching pursuit-like algorithm

We present a computationally efficient and numerically robust solution to the problem of removing artifacts due to precordial compressions and ventilations from the human electrocardiogram (ECG) in an emergency medicine setting. Incorporated into automated external defibrillators, this would allow for simultaneous ECG signal analysis and administration of precordial compressions and ventilations, resulting in significant clinical improvement to the treatment of cardiac arrest patients. While we have previously demonstrated the feasibility of such artifact removal using a multichannel Wiener filter, we here focus on an efficient matching pursuit-like approach making practical real-time implementations of such a scheme feasible for a wide variety of sampling rates and filter lengths. Using more realistic data than what have been previously available, we present evidence showing the excellent performance of our approach and quantify its computational complexity.

CPR artifact removal from human ECG using optimal multichannel filtering

The purpose of this study was to assess whether the artifacts presented by precordial compressions during cardiopulmonary resuscitation could be removed from the human electrocardiogram (ECG) using a filtering approach. This would allow analysis and defibrillator charging during ongoing precordial compressions yielding a very important clinical improvement to the treatment of cardiac arrest patients. In this investigation the authors started with noise-free human ECGs with ventricular fibrillation (VF) and ventricular tachycardia (VT) records. To simulate a realistic resuscitation situation, they added a weighted artifact signal to the human ECG, where the weight factor was chosen to provide the desired signal-to-noise ratio (SNR) level. As artifact signals the authors used ECGs recorded from animals in asystole during precordial compressions at rates 60, 90, and 120 compressions/min. The compression depth and the thorax impedance was also recorded. In a real-life situation such reference signals are available and, using an adaptive multichannel Wiener filter, the authors construct an estimate of the artifact signal, which subsequently can be subtracted from the noisy human ECG signal. The success of the proposed method is demonstrated through graphic examples, SNR, and rhythm classification evaluations.

On the optimality of nonunitary filter banks in subband coders

This paper investigates the energy compaction capabilities of nonunitary filter banks in subband coding. It is shown that nonunitary filter banks have larger coding gain than unitary filter banks because of the possibility of performing half-whitening in each channel. For long filter unit pulse responses, optimization of subband coding gain for stationary input signals results in a filter bank decomposition, where each channel works as an optimal open-loop DPCM system. We derive a formula giving the optimal filter response for each channel as a function of the input power spectral density (PSD). For shorter filter bank responses, good gain is obtained by suboptimal half-whitening responses, but the impact on the theoretical coding gain is still highly significant. Image coding examples demonstrate that better performance is achieved using nonunitary filter banks when the input images are in correspondence with the signal model.

A critique of SVD-based image coding systems

During the past couple of decades several proposals for image coders using singular value decomposition (SVD) have been put forward. The results using SVD in this context have never been spectacular. The main problem with the SVD is that the transform itself must be transmitted as side information. We demonstrate through some simple experiments that for a given image reconstruction quality, more scalar parameters must be transmitted using the SVD, than when using the discrete cosine transform (DCT). Also, using an alternative interpretation of the SVD we show that the SVD representation necessitates quantization of individual factors as compared to quantization of the associated product. This is clearly suboptimal.

Impedance-Based Ventilation Detection During Cardiopulmonary Resuscitation

It has been suggested to develop automated external defibrillators with the ability to monitor cardiopulmonary resuscitation (CPR) performance online and give corrective feedback in order to improve the resuscitation quality. Thoracic impedance changes are closely correlated to lung volume changes and can be used to monitor the ventilatory activity. We developed a pattern-recognition-based detection system that uses thoracic impedance to accurately detect ventilation during ongoing CPR. The detection system was developed and evaluated on recordings of real-world resuscitation efforts of cardiac arrest patients where ventilations were manually annotated by human experts. The annotated ventilations were detected with an overall positive predictive value of 95.5% for a sensitivity of 90.4%. During chest compressions, the detection system achieved a mean positive predictive value of 94.8% for a sensitivity of 88.7%. The results suggest that accurate ventilation detection during CPR based on the proposed approach is feasible, and that the performance is not significantly degraded in the presence of chest compressions.

Automatic Identification of Return of Spontaneous Circulation During Cardiopulmonary Resuscitation

The main problem during pulse check in out-of-hospital cardiac arrest is the discrimination between normal pulse-generating rhythm (PR) and pulseless electrical activity (PEA). It has been suggested that circulatory information can be acquired by measuring the thoracic impedance via the defibrillator pads. To investigate this, we performed an experimental study where we retrospectively analyzed 127 PEA segments and 91 PR segments out of 219 and 113 segments. A PEA versus PR classification framework was developed, that uses short segments (< 10 s) of ECG and impedance measurements to discriminate between the two rhythms. Using realistic data analyzed over a duration of 3 s, our system correctly identifies 90.0% of the segments with rhythm being pulseless electrical activity, and 91.5% of the normal pulse rhythm segments. Automatic identification of pulse could avoid unnecessary pulse checks and thereby reduce no-flow time and potentially increase the chance of survival.