Search

Physician detection of heart sound abnormality is complicated by the inherent difficulty of detecting critical abnormalities in the presence of noise. Computer-aided heart auscultation provides a promising alternative for more accurate detection, with recent deep learning approaches exceeding expert accuracy. Although combining phonocardiogram (PCG) data with electrocardiogram (ECG) data provides more information to an abnormal heart sound classifier, the scarce presence of labelled datasets with this combination impedes training. This paper explores fine-tuning deep convolutional neural networks such as ResNet, VGG, and inceptionv3, on images of spectrograms, mel-spectrograms, and scalograms. By fine-tuning deep pre-trained models on image representations of ECG and PCG, we achieve 91.25% accuracy on the training-a dataset of the PhysioNet Computing in Cardiology Challenge 2016, compared to a previous result of 81.48%. Interpretation of the model’s learned features is also provided, with the results indicative of clinical significance.

This paper proposes a new adaptive algorithm for the second-order blind signal separation (BSS) problem with convolutive mixtures by utilising a combination of an accelerated gradient and a conjugate gradient method. For each iteration of the adaptive algorithm, the search point and the search direction are obtained based on the current and the previous iterations. The algorithm efficiently calculates the step size for the accelerated conjugate gradient algorithm in each iteration. Simulation results show that the proposed accelerated conjugate gradient algorithm with optimal step size converges faster than the accelerated descent algorithm and the steepest descent algorithm with optimal step size while having lower computational complexity. In particular, the number of iterations required for convergence of the accelerated conjugate gradient algorithm is significantly lower than the accelerated descent algorithm and the steepest descent algorithm. In addition, the proposed system achieves improvement in terms of the signal to interference ratio and signal to noise ratio for the dominant speech outputs.

Abstract In this paper, an adaptive averaging a priori SNR estimation employing critical band processing is proposed. The proposed method modifies the current decision-directed a priori SNR estimation to achieve faster tracking when SNR changes. The decision-directed estimator (DD) employs a fixed weighting with the value close to one, which makes it slow in following the onsets of speech utterances. The proposed SNR estimator provides a means to solve this issue by employing an adaptive weighting factor. This allows an improved tracking of onset changes in the speech signal. As a consequence, it results in better preservation of speech components. This adaptive technique ensures that the weighting between the modified decision-directed a priori estimate and the maximum likelihood a priori estimate is a function of the speech absence probability. The estimate of the speech absence probability is modeled by a sigmoid function. Furthermore, a critical band mapping for the short-time Fourier transform analysis-synthesis system is utilized in the speech enhancement to achieve less musical noise. In addition, to evaluate the ability of the a priori SNR estimation method in preserving speech components, we proposed a modified objective measurement known as modified hamming distance. Evaluations are performed by utilizing both objective and subjective measurements. The experimental results show that the proposed method improves the speech quality under different noise conditions. Moreover, it maintains the advantage of the DD approach in eliminating the musical noise under different SNR conditions. The objective results are supported by subjective listening tests using 10 subjects (5 males and 5 females).

Adaptive noise cancellation is a useful linear technique to attenuate unwanted background noise that cannot be removed using traditional frequency-selective filters. Usually, this is due to the signal and noise co-existing in the same frequency band. This paper tests a weighted least mean squares (WLMS) algorithm on a stethoscope system for use in detecting coronary artery disease in the presence of background noise. Each stethoscope is equipped with two microphones: one used to detect heart signals and one used to detect background noise. The WLMS method was used for four different sources of background noise whilst measuring a heartbeat, including a single tone, multiple tones, hospital/clinic noise, and breathing noise. The magnitude-squared coherence between both microphones was unity for the tone scenarios, resulting in complete attenuation. For the other background noise sources, a less-than-unity magnitude-squared coherence resulted in minor and no attenuation. Thus, the coherence function is a tool that can be used to predict the amount of attenuation achievable by linear adaptive noise-cancellation techniques, such as WLMS, as presented in this article.

Adaptive modulation received significant attention for underwater acoustic (UA) communication systems with the aim of increasing the system efficiency. It is challenging to attain a high data rate in UA communication, as UA channels vary fast, along with the environmental factors. For a time-varying UA channel, a self-adaptive system is an attractive option, which can choose the best method according to the channel condition to guarantee the continuous connectivity and high performance constantly. A real-time orthogonal frequency-division multiplexing (OFDM)-based adaptive UA communication system is presented in this paper, employing the National Instruments (NI) LabVIEW software and NI CompactDAQ device. In this paper, the received SNR is considered as a performance metric to select the transmission parameters, which are sent back to the transmitter for data transmission. In this research, a UA OFDM communication system is developed, employing adaptive modulation schemes for a nonstationary UA environment which allows to select subcarriers, modulation size, and allocate power adaptively to enhance the reliability of communication, guarantee continuous connectivity, and boost data rate. The recent UA communication experiments carried out in the Canning River, Western Australia, verify the performance of the proposed adaptive UA OFDM system, and the experimental results confirm the superiority of the proposed adaptive scheme.

This paper concerns the soft-in soft-out detection in a coded communication system, where the transmitted symbols are discrete valued, and the exact a posteriori probability (APP) detection often involves prohibitive complexity. By using the properties of Gaussian functions, an approximate approach to the APP detection is devised with the idea that, in the computation of the APP of each symbol, the remaining symbols are distinguished based on their contributions to the APP of the concerned symbol, and the symbols with less contributions are approximated as (continuous) Gaussian variables [hence the name partial Gaussian approximation (PGA)] to reduce the computational complexity. The connection between the PGA detector and the reduced dimension maximum a posteriori detector (RDMAP) is investigated. It is shown that, PGA is equivalent to RDMAP, but it has a complexity much lower than that of RDMAP, i.e., PGA can be regarded as an efficient implementation of RDMAP. In addition, the application of PGA in intersymbol interference (ISI) channel equalization is also investigated. We show that PGA allows further significant complexity reduction by exploiting the circulant structure of the system transfer matrix, which makes PGA very attractive in handling severe ISI channels with large memory length.

In designing acoustic broadband beamformers, the complexity can grow significantly when the number of microphones and the filter length increase. It is advantageous if many of the filter coefficients are zeroes so that the implementation can be executed with less computation. Moreover, the size of the array can also be pruned to reduce complexity. These problems are addressed in this paper. A suitable optimization model is proposed. Both array pruning and filter thinning can be solved together as a two-stage optimization problem to yield the final sparse designs. Numerical results show that the complexity of the designed beamformers can be reduced significantly with minimal effect on performance.

This paper investigates the problem of enhancing a single desired speech source from a mixture of signals in multispeaker environments. A beamformer structure is proposed which combines a fixed beamformer with postfiltering. In the first stage, the fixed multiobjective optimal beamformer is designed to spatially extract the desired source by suppressing all other undesired sources. In the second stage, a multichannel power spectral estimator is proposed and incorporated in the postfilter, thus enabling further suppression capability. The combined scheme exploits both spatial and spectral characteristics of the signals. Two new multichannel spectral estimation methods are proposed for the postfiltering using, respectively, inner product and joint diagonalization. Evaluations using recordings from a real-room environment show that the proposed beamformer offers a good interference suppression level whilst maintaining a low-distortion level of the desired source.