A novel atrial fibrillation automatic detection algorithm based on ensemble learning and multi-feature discrimination.

Atrial fibrillation (AF) is a prevalent cardiac arrhythmia disorder that necessitates long-time electrocardiogram (ECG) data for clinical diagnosis, leading to low detection efficiency. Automatic detection of AF signals within short-time ECG recordings is challenging. To address these issues, this paper proposes a novel algorithm called Ensemble Learning and Multi-Feature Discrimination (ELMD) for the identification and detection of AF signals. Firstly, a robust classifier, BSK-Model, is constructed using ensemble learning. Subsequently, the ECG R-waves are detected, and the ECG signals are segmented into consecutive RR intervals. Time domain, frequency domain, and nonlinear features are extracted from these intervals. Finally, these features are fed into the BSK-Model to discriminate AF. The proposed methodology is evaluated using the MIT-BIH AF database. The results demonstrate that when RR intervals are employed as classification units, the specificity and accuracy of AF detection in long-time ECG data exceed 99%, showcasing a significant improvement over traditional single-model classification. Additionally, the sensitivity and accuracy achieved by testing cardiac segments are both above 96%. With a minimum requirement of only four cardiac segments, AF events can be accurately identified, thereby enabling rapid discrimination of short-time single-lead ECG AF events. Consequently, this approach is suitable for real-time and accurate AF detection using low-computational-power ECG diagnostic analysis devices, such as wearable devices.

Multi-Channel Representation Learning Enhanced Unfolding Multi-Scale Compressed Sensing Network for High Quality Image Reconstruction.

Deep Unfolding Networks (DUNs) serve as a predominant approach for Compressed Sensing (CS) reconstruction algorithms by harnessing optimization. However, a notable constraint within the DUN framework is the restriction to single-channel inputs and outputs at each stage during gradient descent computations. This constraint compels the feature maps of the proximal mapping module to undergo multi-channel to single-channel dimensionality reduction, resulting in limited feature characterization capabilities. Furthermore, most prevalent reconstruction networks rely on single-scale structures, neglecting the extraction of features from different scales, thereby impeding the overall reconstruction network's performance. To address these limitations, this paper introduces a novel CS reconstruction network termed the Multi-channel and Multi-scale Unfolding Network (MMU-Net). MMU-Net embraces a multi-channel approach, featuring the incorporation of Adap-SKConv with an attention mechanism to facilitate the exchange of information between gradient terms and enhance the feature map's characterization capacity. Moreover, a Multi-scale Block is introduced to extract multi-scale features, bolstering the network's ability to characterize and reconstruct the images. Our study extensively evaluates MMU-Net's performance across multiple benchmark datasets, including Urban100, Set11, BSD68, and the UC Merced Land Use Dataset, encompassing both natural and remote sensing images. The results of our study underscore the superior performance of MMU-Net in comparison to existing state-of-the-art CS methods.

[An image classification method for arrhythmias based on Gramian angular summation field and improved Inception-ResNet-v2].

Arrhythmia is a significant cardiovascular disease that poses a threat to human health, and its primary diagnosis relies on electrocardiogram (ECG). Implementing computer technology to achieve automatic classification of arrhythmia can effectively avoid human error, improve diagnostic efficiency, and reduce costs. However, most automatic arrhythmia classification algorithms focus on one-dimensional temporal signals, which lack robustness. Therefore, this study proposed an arrhythmia image classification method based on Gramian angular summation field (GASF) and an improved Inception-ResNet-v2 network. Firstly, the data was preprocessed using variational mode decomposition, and data augmentation was performed using a deep convolutional generative adversarial network. Then, GASF was used to transform one-dimensional ECG signals into two-dimensional images, and an improved Inception-ResNet-v2 network was utilized to implement the five arrhythmia classifications recommended by the AAMI (N, V, S, F, and Q). The experimental results on the MIT-BIH Arrhythmia Database showed that the proposed method achieved an overall classification accuracy of 99.52% and 95.48% under the intra-patient and inter-patient paradigms, respectively. The arrhythmia classification performance of the improved Inception-ResNet-v2 network in this study outperforms other methods, providing a new approach for deep learning-based automatic arrhythmia classification.

Determination of Parameters for an Entropy-Based Atrial Fibrillation Detector.

Entropy algorithm is an important nonlinear method for cardiovascular disease detection due to its power in analyzing short-term time series. In previous a study, we proposed a new entropy-based atrial fibrillation (AF) detector, i.e., EntropyAF, which showed a high classification accuracy in identifying AF and non-AF rhythms. As a variation of entropy measures, EntropyAF has two parameters that need to be initialized before the calculation: (1) tolerance threshold r and (2) similarity weight n. In this study, a comprehensive analysis for the two parameters determination was presented, aiming to achieve a high detection accuracy for AF events. Data were from the MIT-BIH AF database. RR interval recordings were segmented using a 30-beat time window. The parameters r and n were initialized from a relatively small value, then gradually increased, and finally the best parameter combination was determined using grid searching. AUC (area under curve) values from the receiver operator characteristic curve (ROC) were compared under different parameter combinations of parameters r and n, and the results demonstrated that the selection of these two parameters plays an important role in AF/non-AF classification. Small values of parameters r and n can lead to a better detection accuracy than other selections. The best AUC value for AF detection was 98.15%, and the corresponding parameter combinations for EntropyAF were as follows: r = 0.01, n = 0.0625, 0.125, 0.25, or 0.5; r = 0.05 and n = 0.0625, 0.125, or 0.25; and r = 0.10 and n = 0.0625 or 0.125.

Ventricular repolarization instability quantified by instantaneous frequency of ECG ST intervals.

BACKGROUND: Ventricular repolarization instabilities have been documented to be closely linked to arrhythmia development. The electrocardiogram (ECG) ST interval can be used to measure ventricular repolarization. Analyzing the duration variation of the ST intervals can provide new information about the arrhythmogenic vulnerability. OBJECTIVE: In this work, we propose a new method based on mean instantaneous frequency (IF) of the ST intervals to quantitatively evaluate the risk of sudden cardiac deaths (SCDs). METHODS: Two spectral bands, i.e. the low-frequency band (LF, 0-0.15 Hz) and the high-frequency band (HF, 0.15-0.5 Hz), are considered in this paper. Based on IF estimates, the ECG recordings from three MIT-BIH databases that represent different risk levels of SCD occurrence are used, and their mean IFs in the LF and HF bands are calculated. RESULTS: The statistical results show that healthy subjects have a higher mean IF in the HF band and a lower mean IF in the LF band. The experimental results are the opposite for patients with malignant ventricular arrhythmia. CONCLUSION: The proposed mean IF can represent an indirect measure of intrinsic ventricular repolarization instability and can mark cardiac instability associated with SCDs.

Electrocardiogram Baseline Wander Suppression Based on the Combination of Morphological and Wavelet Transformation Based Filtering.

One of the major noise components in electrocardiogram (ECG) is the baseline wander (BW). Effective methods for suppressing BW include the wavelet-based (WT) and the mathematical morphological filtering-based (MMF) algorithms. However, the T waveform distortions introduced by the WT and the rectangular/trapezoidal distortions introduced by MMF degrade the quality of the output signal. Hence, in this study, we introduce a method by combining the MMF and WT to overcome the shortcomings of both existing methods. To demonstrate the effectiveness of the proposed method, artificial ECG signals containing a clinical BW are used for numerical simulation, and we also create a realistic model of baseline wander to compare the proposed method with other state-of-the-art methods commonly used in the literature. The results show that the BW suppression effect of the proposed method is better than that of the others. Also, the new method is capable of preserving the outline of the BW and avoiding waveform distortions caused by the morphology filter, thereby obtaining an enhanced quality of ECG.

Beamforming for hyperthermia treatment by training a weighted network of an ultrasonic array.

We propose a beamforming algorithm based on waveform diversity for hyperthermia treatment of breast cancer using an ultrasonic array. The introduced array has a structure with a network connecting the feeding nodes and the array elements, and the objective of the algorithm is to train the weight matrix of the network to minimize the difference between the generated beam pattern and the ideal one. The training procedure of the algorithm, which is inspired by the idea of machine learning, comprises three parts: forward calculation, comparison, and backward calculation. The forward calculation maps the weight matrix to the beam pattern, and in the comparison step, the generated beam pattern is modified based on the error, and finally, the backward calculation maps the modified beam pattern to a refined weight matrix which performs better than the original one. An optimal weight matrix is obtained by iterative training. The effectiveness of the algorithm is demonstrated by using numerical simulations.

Classification of epileptic EEG signals based on simple random sampling and sequential feature selection.

Electroencephalogram (EEG) signals are used broadly in the medical fields. The main applications of EEG signals are the diagnosis and treatment of diseases such as epilepsy, Alzheimer, sleep problems and so on. This paper presents a new method which extracts and selects features from multi-channel EEG signals. This research focuses on three main points. Firstly, simple random sampling (SRS) technique is used to extract features from the time domain of EEG signals. Secondly, the sequential feature selection (SFS) algorithm is applied to select the key features and to reduce the dimensionality of the data. Finally, the selected features are forwarded to a least square support vector machine (LS_SVM) classifier to classify the EEG signals. The LS_SVM classifier classified the features which are extracted and selected from the SRS and the SFS. The experimental results show that the method achieves 99.90, 99.80 and 100 % for classification accuracy, sensitivity and specificity, respectively.

[An Improved Cubic Spline Interpolation Method for Removing Electrocardiogram Baseline Drift].

The selection of fiducial points has an important effect on electrocardiogram(ECG)denoise with cubic spline interpolation.An improved cubic spline interpolation algorithm for suppressing ECG baseline drift is presented in this paper.Firstly the first order derivative of original ECG signal is calculated,and the maximum and minimum points of each beat are obtained,which are treated as the position of fiducial points.And then the original ECG is fed into a high pass filter with 1.5Hz cutoff frequency.The difference between the original and the filtered ECG at the fiducial points is taken as the amplitude of the fiducial points.Then cubic spline interpolation curve fitting is used to the fiducial points,and the fitting curve is the baseline drift curve.For the two simulated case test,the correlation coefficients between the fitting curve by the presented algorithm and the simulated curve were increased by 0.242and0.13 compared with that from traditional cubic spline interpolation algorithm.And for the case of clinical baseline drift data,the average correlation coefficient from the presented algorithm achieved 0.972.

A time-domain hybrid analysis method for detecting and quantifying T-wave alternans.

T-wave alternans (TWA) in surface electrocardiograph (ECG) signals has been recognized as a marker of cardiac electrical instability and is hypothesized to be associated with increased risk for ventricular arrhythmias among patients. A novel time-domain TWA hybrid analysis method (HAM) utilizing the correlation method and least squares regression technique is described in this paper. Simulated ECGs containing artificial TWA (cases of absence of TWA and presence of stationary or time-varying or phase-reversal TWA) under different baseline wanderings are used to test the method, and the results show that HAM has a better ability of quantifying TWA amplitude compared with the correlation method (CM) and adapting match filter method (AMFM). The HAM is subsequently used to analyze the clinical ECGs, and results produced by the HAM have, in general, demonstrated consistency with those produced by the CM and the AMFM, while the quantifying TWA amplitudes by the HAM are universally higher than those by the other two methods.

[The detection of non-stationary T-wave alternans: a modified correlation method].

T-wave alternans (TWA) refers to a phenomenon appearing in the surface electrocardiograph (ECG) as a consistent fluctuation in morphology and amplitude of the T wave on an "every-other-beat" basis. Correlation method (CM) has a certain ability to detect the non-stationary TWA, but it is very sensitive to noise. In this paper we propose a modified correlation method to ensure a stable and accurate detection of non-stationary TWA. Compared to the CM, the method modifies the judge condition and uses the linear fitting to limit the noise to gain the ability of detecting of non-stationary TWA. Our simulation and clinical data assessment study demonstrates the improved performance of the proposed algorithm.

A combined algorithm for T-wave alternans qualitative detection and quantitative measurement.

BACKGROUND: T-wave alternans (TWA) provides a noninvasive and clinically useful marker for the risk of sudden cardiac death (SCD). Current most widely used TWA detection algorithms work in two different domains: time and frequency. The disadvantage of the spectral analytical techniques is that they treat the alternans signal as a stationary wave with a constant amplitude and a phase. They cannot detect non-stationary characteristics of the signal. The temporal domain methods are sensitive to the alignment of the T-waves. In this study, we sought to develop a robust combined algorithm (CA) to assess T-wave alternans, which can qualitatively detect and quantitatively measure TWA in time domain. METHODS: The T wave sequences were extracted and the total energy of each T wave within the specified time-frequency region was calculated. The rank-sum test was applied to the ranked energy sequences of T waves to detect TWA qualitatively. The ECG containing TWA was quantitatively analyzed with correlation method. RESULTS: Simulation test result proved a mean sensitivity of 91.2% in detecting TWA, and for the SNR not less than 30 dB, the accuracy rate of detection achieved 100%. The clinical data experiment showed that the results from this method vs. spectral method had the correlation coefficients of 0.96. CONCLUSIONS: A novel TWA analysis algorithm utilizing the wavelet transform and correlation technique is presented in this paper. TWAs are not only correctly detected qualitatively in frequency domain by energy value of T waves, but the alternans frequency and amplitude in temporal domain are measured quantitatively.

[Research on electrocardiogram de-noising algorithm based on wavelet neural networks].

In this paper, the ECG de-noising technology based on wavelet neural networks (WNN) is used to deal with the noises in Electrocardiogram (ECG) signal. The structure of WNN, which has the outstanding nonlinear mapping capability, is designed as a nonlinear filter used for ECG to cancel the baseline wander, electromyo-graphical interference and powerline interference. The network training algorithm and de-noising experiments results are presented, and some key points of the WNN filter using ECG de-noising are discussed.

[A new approach to wavelet-based P-wave detection].

According to the characteristics of four basic P morphologies, combining the wavelet transform and the amplitude and slope of transformed P wave, a new P-wave detecting method based on "wavelet-amplitude-slope" algorithm is presented: First search out all modulus maximum pairs to satisfy the threshold after wavelet transform, and then applying the amplitude and slope criterion exclude the interferes and detect the P peak and its shape, last determine the onset and end of P wave respectively which should be separately calculated for single-peak and double-peak P wave (or biphasic P wave). The approach is applied in experiments of data from MIT/BIH database and randomly collected data of clinical ECG. The experimental statistical results shows that the correct detecting rate is as high as 96% compared to manual annotation.