Deep Learning-Based Stacked Denoising and Autoencoder for ECG Heartbeat Classification

TinyML and edge intelligence applications in cardiovascular disease: A survey

Tiny machine learning (TinyML) and edge intelligence have emerged as pivotal paradigms for enabling machine learning on resource-constrained devices situated at the extreme edge of networks. In this paper, we explore the transformative potential of TinyML in facilitating pervasive, low-power cardiovascular monitoring and real-time analytics for patients with cardiac anomalies, leveraging wearable devices as the primary interface. To begin with, we provide an overview of TinyML software and hardware enablers, accompanied by an examination of networking solutions such as Low-power Wide area network (LPWAN) that facilitate the seamless deployment of TinyML frameworks. Following this, we delve into the methodologies of knowledge distillation, quantization, and pruning, which represent the cornerstone strategies for optimizing machine learning models to operate efficiently within resource-constrained environments. Furthermore, our discussion extends to the role of efficient deep neural networks tailored specifically for cardiovascular monitoring on wearable devices with limited computational resources. Through a comprehensive review, we analyze the applications of prominent artificial neural network architectures including Convolutional Neural Networks (CNNs), Autoencoders, Deep Belief Networks (DBNs), and Transformers in the domain of Electrocardiogram (ECG) analytics, shedding light on their efficacy and potential in advancing healthcare technology.

Leveraging Mask Autoencoder and Crossover Binary Sand Cat Algorithm for Early Detection of Glaucoma

Glaucoma, a leading cause of irreversible blindness worldwide, can be effectively managed if detected early. Glaucoma is directly associated with aging as it commonly occurs in people over the age of 40 and in elderly people. Glaucoma detection in retinal fundus images typically involves utilizing image processing and machine learning techniques. By leveraging advancements in computer vision, a robust and automated system is developed to assist ophthalmologists in screening and diagnosing glaucoma from retinal fundus images. Furthermore, fundus images can vary significantly in quality due to factors like illumination variations, focus, and artifacts. Ensuring consistent image quality across different datasets and acquisition devices is essential for reliable detection. Addressing these challenges requires interdisciplinary collaboration between ophthalmologists to develop robust and reliable solutions for the detection of glaucoma. Hence a novel mask autoencoder-based crossover binary sand cat (MA-CBSC) algorithm is proposed to detect glaucoma. In this algorithm, the mask autoencoder recognizes the features indicating the presence of glaucoma in the input images and the crossover binary sand cat algorithm is used to fine tune the overall performance of the algorithm by selecting the most appropriate features escaped due to overfitting issues. Preprocessing steps such as image enhancement, filtering, and data cleaning are applied to the extracted ROI for the purpose of increasing the image quality and enhancing the visibility of features relevant to glaucoma detection. ROI extraction attributes namely optic disc, cup-to-disc ratio, bean-pot cupping, and vertical enlargement are derived from the ROI along with some other relevant features. In this work, the crossover-based binary sand cat optimization algorithm is utilized for hyperparameter tuning to enhance the efficiency of the MA-CBSC method. Extensive experimental assessments are conducted, comparing the effectiveness of MA-CBSC algorithms with the Retinal Disease Classification dataset, Fundus Glaucoma Detection Data Dataset, and Glaucoma Dataset. The results obtained by the proposed method are compared with the results obtained by the existing techniques such as DLCNN-MGWO-VW, FRCNN-FKM, ML-DCNN, and DNN-MSO to show its superiority. Seven evaluation parameters are used in assessing the efficiency of the developed model including positive predictive value (PPV), accuracy, precision, F1 score, sensitivity, recall, and specificity. These evaluation measures show that the model has a more promising performance than the existing methods with 98.3% accuracy.

Disentangled representational learning for anomaly detection in single-lead electrocardiogram signals using variational autoencoder

Wearable technology enables the unsupervised recording of electrocardiogram (ECG) signals. Analyzing these high-dimensional ECG data poses challenges regarding statistical approaches and explainability. This work investigates the feasibility of medically explainable anomaly detection through disentangled representational learning of ECGs and personalization to mitigate inter-subject variations. Five open-source ECG datasets were converted into a set of denoised one-second traces of lead I signal, each covering individual features such as wave morphologies and pathologies. A beta total correlation variational autoencoder was optimized on four of these datasets for 68 systematic parameterization variants. The best-performing model revealed disentanglement in the 12-dimensional embedding space, specifically between atrial- and ventricular features. Within the embedding space, a k-nearest neighbor classifier was evaluated on a left-out test set tailored for anomaly detection. The result is a F1 score of 0.94 for the binary prediction of sinus rhythm and the pathological classes: Left bundle branch block, right bundle branch block, myocardial infarction, and AV block (1st degree). The 90.94% accuracy in anomaly detection falls within the range of established detectors (89.38%-99.77%) but offers the advantage of being explainable and largely unsupervised. Model fine-tuning for each of 100 randomly sampled individuals of the Icentia11k dataset mitigated inter-subject variations. The associated F1 score for predicting normal, premature atrial contraction, and premature ventricular contraction from the embedding space was 0.93. The distribution plots of pathologies along the explainable axis were reasonably consistent with medical expertise. The results suggest the presented disentangled variational autoencoder as a robust method for explainable ECG representation.

An improved electrocardiogram arrhythmia classification performance with feature optimization

BACKGROUND

Automatic classification of arrhythmias based on electrocardiography (ECG) data faces several significant challenges, particularly due to the substantial volume of clinical data involved in ECG signal analysis. The volume of clinical data has increased considerably, especially with the emergence of new clinical symptoms and signs in various arrhythmia conditions. These symptoms and signs, which serve as distinguishing features, can number in the tens of thousands. However, the inclusion of irrelevant features can lead to inaccurate classification results.

METHOD

To identify the most relevant and optimal features for ECG arrhythmia classification, common feature extraction techniques have been applied to ECG signals, specifically shallow and deep feature extraction. Additionally, a feature selection technique based on a metaheuristic optimization algorithm is utilized following the ECG feature extraction process.

RESULTS

Our findings indicate that shallow feature extraction based on the time-domain analysis, combined with feature selection using a metaheuristic optimization algorithm, outperformed other ECG feature extraction and selection techniques. Among eight features of time-domain anaylsis, the selected feature is one to three features from RR-interval assesment, achieving 100% accuracy, sensitivity, specificity, and precision for ECG arrhythmia classification.

CONCLUSION

The proposed end-to-end architecture for ECG arrhythmia classification demonstrates simplicity in parameters and low complexity, making it highly effective for practical applications.

Cardiovascular disease detection from cardiac arrhythmia ECG signals using artificial intelligence models with hyperparameters tuning methodologies

Cardiovascular disease (CVD) is connected with irregular cardiac electrical activity, which can be seen in ECG alterations. Due to its convenience and non-invasive aspect, the ECG is routinely exploited to identify different arrhythmias and automatic ECG recognition is needed immediately. In this paper, enhancement for the detection of CVDs such as Ventricular Tachycardia (VT), Premature Ventricular Contraction (PVC) and ST Change (ST) arrhythmia using different dimensionality reduction techniques and multiple classifiers are presented. Three-dimensionality reduction methods, such as Local Linear Embedding (LLE), Diffusion Maps (DM), and Laplacian Eigen (LE), are employed. The dimensionally reduced ECG samples are further feature selected with Cuckoo Search (CS) and Harmonic Search Optimization (HSO) algorithms. A publicly available MIT-BIH (Physionet) - VT database, PVC database, ST Change database and NSR database were used in this work. The cardiac vascular disturbances are classified by using seven classifiers such as Gaussian Mixture Model (GMM), Expectation Maximization (EM), Non-linear Regression (NLR), Logistic Regression (LR), Bayesian Linear Discriminant Analysis (BDLC), Detrended Fluctuation Analysis (Detrended FA), and Firefly. For different classes, the average overall accuracy of the classification techniques is 55.65 % when without CS and HSO feature selection, 64.36 % when CS feature selection is used, and 75.39 % when HSO feature selection is used. Also, to improve the performance of classifiers, the hyperparameters of four classifiers (GMM, EM, BDLC and Firefly) are tuned with the Adam and Grid Search Optimization (GSO) approaches. The average accuracy of classification for the CS feature-based classifiers that used GSO and Adam hyperparameter tuning was 79.92 % and 85.78 %, respectively. The average accuracy of classification for the HSO feature-based classifiers that used GSO and Adam hyperparameter tuning was 86.87 % and 93.77 %, respectively. The performance of the classifier is analyzed based on the accuracy parameter for both with and without feature selection methods and with hyperparameter tuning techniques. In the case of ST vs. NSR, a higher accuracy of 98.92 % is achieved for the LLE dimensionality reduction with HSO feature selection for the GMM classifier with Adam's hyperparameter tuning approach. The GMM classifier with the Adam hyperparameter tuning approach with 98.92 % accuracy in detecting ST vs. NSR cardiac disease is outperforming all other classifiers and methodologies.

Denoising magnetic resonance spectroscopy (MRS) data using stacked autoencoder for improving signal-to-noise ratio and speed of MRS

BACKGROUND

While magnetic resonance imaging (MRI) provides high resolution anatomical images with sharp soft tissue contrast, magnetic resonance spectroscopy (MRS) enables non-invasive detection and measurement of biochemicals and metabolites. However, MRS has low signal-to-noise ratio (SNR) when concentrations of metabolites are in the range of millimolar. Standard approach of using a high number of signal averaging (NSA) to achieve sufficient SNR comes at the cost of a long acquisition time.

PURPOSE

We propose to use deep-learning approaches to denoise MRS data without increasing NSA. This method has potential to reduce the acquisition time as well as improve SNR and quality of spectra, which could enhance the diagnostic value and broaden the clinical applications of MRS.

METHODS

The study was conducted using data collected from the brain spectroscopy phantom and human subjects. We utilized a stack auto-encoder (SAE) network to train deep learning models for denoising low NSA data (NSA = 1, 2, 4, 8, and 16) randomly truncated from high SNR data collected with high NSA (NSA = 192), which were also used to obtain the ground truth. We applied both self-supervised and fully-supervised training approaches and compared their performance of denoising low NSA data based on improvement in SNR. To prevent overfitting, the SAE network was trained in a patch-based manner. We then tested the denoising methods on noise-containing data collected from the phantom and human subjects, including data from brain tumor patients. We evaluated their performance by comparing the SNR levels and mean squared errors (MSEs) calculated for the whole spectra against high SNR "ground truth", as well as the value of chemical shift of N-acetyl-aspartate (NAA) before and after denoising.

RESULTS

With the SAE model, the SNR of low NSA data (NSA = 1) obtained from the phantom increased by 28.5% and the MSE decreased by 42.9%. For low NSA data of the human parietal and temporal lobes, the SNR increased by 32.9% and the MSE decreased by 63.1%. In all cases, the chemical shift of NAA in the denoised spectra closely matched with the high SNR spectra without significant distortion to the spectra after denoising. Furthermore, the denoising performance of the SAE model was more effective in denoising spectra with higher noise levels.

CONCLUSIONS

The reported SAE denoising method is a model-free approach to enhance the SNR of MRS data collected with low NSA. With the denoising capability, it is possible to acquire MRS data with a few NSA, shortening the scan time while maintaining adequate spectroscopic information for detecting and quantifying the metabolites of interest. This approach has the potential to improve the efficiency and effectiveness of clinical MRS data acquisition by reducing the scan time and increasing the quality of spectroscopic data.

Deep learning for ECG Arrhythmia detection and classification: an overview of progress for period 2017-2023

Cardiovascular diseases are a leading cause of mortality globally. Electrocardiography (ECG) still represents the benchmark approach for identifying cardiac irregularities. Automatic detection of abnormalities from the ECG can aid in the early detection, diagnosis, and prevention of cardiovascular diseases. Deep Learning (DL) architectures have been successfully employed for arrhythmia detection and classification and offered superior performance to traditional shallow Machine Learning (ML) approaches. This survey categorizes and compares the DL architectures used in ECG arrhythmia detection from 2017-2023 that have exhibited superior performance. Different DL models such as Convolutional Neural Networks (CNNs), Multilayer Perceptrons (MLPs), Transformers, and Recurrent Neural Networks (RNNs) are reviewed, and a summary of their effectiveness is provided. This survey provides a comprehensive roadmap to expedite the acclimation process for emerging researchers willing to develop efficient algorithms for detecting ECG anomalies using DL models. Our tailored guidelines bridge the knowledge gap allowing newcomers to align smoothly with the prevailing research trends in ECG arrhythmia detection. We shed light on potential areas for future research and refinement in model development and optimization, intending to stimulate advancement in ECG arrhythmia detection and classification.

Hygeia: A Multilabel Deep Learning-Based Classification Method for Imbalanced Electrocardiogram Data

Electrocardiogram (ECG) is a common diagnostic indicator of heart disease in hospitals. Because of the low price and noninvasiveness of ECG diagnosis, it is widely used for prescreening and physical examination of heart diseases. In several studies on ECG analysis, only rough diagnoses are made to determine whether ECGs are abnormal or on a few kinds of ECG. In actual scenarios, doctors must analyze ECG samples in detail, which is a multilabel classification problem. Herein, we propose Hygeia, a multilabel deep learning-based ECG classification method that can analyze and classify 55 types of ECG. First, a guidance model is constructed to transform the multilabel classification problem into multiple interrelated two-classification models. This method ensures the good performance of each ECG analysis model, and the relationship between various types of ECG can be used in the analysis. The imbalance of samples in ECG datasets makes it difficult to analyze abnormal ECGs with high sensitivity and accuracy. We used data generation and mixed sampling methods for 11 ECG types with imbalanced problems to improve the average accuracy, sensitivity, F1 value, and accuracy from 87.74%, 43.11%, 0.3929, and 0.3929, to 92.68%, 96.92, 0.9287, and 99.47%, respectively. The average accuracy, sensitivity, F1 value, and accuracy of 44 of the 55 tags of the abnormal ECG analysis model are 99.69%, 95.81%, 0.9758, and 99.72%, respectively.

Deep learning-based protoacoustic signal denoising for proton range verification

Proton therapy is a type of radiation therapy that can provide better dose distribution compared to photon therapy by delivering most of the energy at the end of range, which is called the Bragg peak (BP). The protoacoustic technique was developed to determine the BP locationsin vivo, but it requires a large dose delivery to the tissue to obtain a high number of signal averaging (NSA) to achieve a sufficient signal-to-noise ratio (SNR), which is not suitable for clinical use. A novel deep learning-based technique has been proposed to denoise acoustic signals and reduce BP range uncertainty with much lower doses. Three accelerometers were placed on the distal surface of a cylindrical polyethylene (PE) phantom to collect protoacoustic signals. In total, 512 raw signals were collected at each device. Device-specific stack autoencoder (SAE) denoising models were trained to denoise the noise-containing input signals, which were generated by averaging only 1, 2, 4, 8, 16, or 24 raw signals (low NSA signals), while the clean signals were obtained by averaging 192 raw signals (high NSA). Both supervised and unsupervised training strategies were employed, and the evaluation of the models was based on mean squared error (MSE), SNR, and BP range uncertainty. Overall, the supervised SAEs outperformed the unsupervised SAEs in BP range verification. For the high accuracy detector, it achieved a BP range uncertainty of 0.20 ± 3.44 mm by averaging over 8 raw signals, while for the other two low accuracy detectors, they achieved the BP uncertainty of 1.44 ± 6.45 mm and -0.23 ± 4.88 mm by averaging 16 raw signals, respectively. This deep learning-based denoising method has shown promising results in enhancing the SNR of protoacoustic measurements and improving the accuracy in BP range verification. It greatly reduces the dose and time for potential clinical applications.

Current and Future Use of Artificial Intelligence in Electrocardiography

Artificial intelligence (AI) is increasingly used in electrocardiography (ECG) to assist in diagnosis, stratification, and management. AI algorithms can help clinicians in the following areas: (1) interpretation and detection of arrhythmias, ST-segment changes, QT prolongation, and other ECG abnormalities; (2) risk prediction integrated with or without clinical variables (to predict arrhythmias, sudden cardiac death, stroke, and other cardiovascular events); (3) monitoring ECG signals from cardiac implantable electronic devices and wearable devices in real time and alerting clinicians or patients when significant changes occur according to timing, duration, and situation; (4) signal processing, improving ECG quality and accuracy by removing noise/artifacts/interference, and extracting features not visible to the human eye (heart rate variability, beat-to-beat intervals, wavelet transforms, sample-level resolution, etc.); (5) therapy guidance, assisting in patient selection, optimizing treatments, improving symptom-to-treatment times, and cost effectiveness (earlier activation of code infarction in patients with ST-segment elevation, predicting the response to antiarrhythmic drugs or cardiac implantable devices therapies, reducing the risk of cardiac toxicity, etc.); (6) facilitating the integration of ECG data with other modalities (imaging, genomics, proteomics, biomarkers, etc.). In the future, AI is expected to play an increasingly important role in ECG diagnosis and management, as more data become available and more sophisticated algorithms are developed.

DAE-ConvBiLSTM: End-to-end learning single-lead electrocardiogram signal for heart abnormalities detection

BACKGROUND

The electrocardiogram (ECG) is a widely used diagnostic that observes the heart activities of patients to ascertain a heart abnormality diagnosis. The artifacts or noises are primarily associated with the problem of ECG signal processing. Conventional denoising techniques have been proposed in previous literature; however, some lacks, such as the determination of suitable wavelet basis function and threshold, can be a time-consuming process. This paper presents end-to-end learning using a denoising auto-encoder (DAE) for denoising algorithms and convolutional-bidirectional long short-term memory (ConvBiLSTM) for ECG delineation to classify ECG waveforms in terms of the PQRST-wave and isoelectric lines. The denoising reconstruction using unsupervised learning based on the encoder-decoder process can be proposed to improve the drawbacks. First, The ECG signals are reduced to a low-dimensional vector in the encoder. Second, the decoder reconstructed the signals. The last, the reconstructed signals of ECG can be processed to ConvBiLSTM. The proposed architecture of DAE-ConvBiLSTM is the end-to-end diagnosis of heart abnormality detection.

RESULTS

As a result, the performance of DAE-ConvBiLSTM has obtained an average of above 98.59% accuracy, sensitivity, specificity, precision, and F1 score from the existing studies. The DAE-ConvBiLSTM has also experimented with detecting T-wave (due to ventricular repolarisation) morphology abnormalities.

CONCLUSION

The development architecture for detecting heart abnormalities using an unsupervised learning DAE and supervised learning ConvBiLSTM can be proposed for an end-to-end learning algorithm. In the future, the precise accuracy of the ECG main waveform will affect heart abnormalities detection in clinical practice.

A robust multiple heartbeats classification with weight-based loss based on convolutional neural network and bidirectional long short-term memory

Background: Analysis of electrocardiogram (ECG) provides a straightforward and non-invasive approach for cardiologists to diagnose and classify the nature and severity of variant cardiac diseases including cardiac arrhythmia. However, the interpretation and analysis of ECG are highly working-load demanding, and the subjective may lead to false diagnoses and heartbeats classification. In recent years, many deep learning works showed an excellent role in accurate heartbeats classification. However, the imbalance of heartbeat classes is universal in most of the available ECG databases since abnormal heartbeats are always relatively rare in real life scenarios. In addition, many existing approaches achieved prominent results by removing noise and extracting features in data preprocessing, which relies heavily on powerful computers. It is a pressing need to develop efficient and automatic light weighted algorithms for accurate heartbeats classification that can be used in portable ECG sensors. Objective: This study aims at developing a robust and efficient deep learning method, which can be embedded into wearable or portable ECG monitors for classifying heartbeats. Methods: We proposed a novel and light weighted deep learning architecture with weight-based loss based on a convolutional neural network (CNN) and bidirectional long short-term memory (Bi-LSTM) that can automatically identify five types of ECG heartbeats according to the AAMI EC57 standard. It was also true that the raw ECG signals were simply segmented without noise removal and other feature extraction processing. Moreover, to tackle the challenge of classification bias due to imbalanced ECG datasets for different types of arrhythmias, we introduced a weight-based loss function to reduce the influence of over-weighted categories in the ECG dataset. For avoiding the influence of the division of validation dataset, k-fold method was adopted to improve the reliability of the model. Results: The proposed algorithm is trained and tested on MIT-BIH Arrhythmia Database, and achieves an average of 99.33% accuracy, 93.67% sensitivity, 99.18% specificity, 89.85% positive prediction, and 91.65% F₁ score.

A Reparameterization Multifeature Fusion CNN for Arrhythmia Heartbeats Classification

Aiming at arrhythmia heartbeats classification, a novel multifeature fusion deep learning-based method is proposed. The stationary wavelet transforms (SWT) and RR interval features are firstly extracted. Based on the traditional one-dimensional convolutional neural network (1D-CNN), a parallel multibranch convolutional network is designed for training. The subband of SWT is input into the multiscale 1D-CNN separately. The output fused with RR interval features are fed to the fully connected layer for classification. To achieve the lightweight network while maintaining the powerful inference capability of the multibranch structure, the redundant branches of the network are removed by reparameterization. Experimental results and analysis show that it outperforms existing methods by many in arrhythmic heartbeat classification.

A lightweight U-net for ECG denoising using knowledge distillation

Objective.Electrocardiogram (ECG) signals are easily polluted by various noises which are likely to have adverse effects on subsequent interpretations. Research on model lightweighting can promote the practical application of deep learning-based ECG denoising methods in real-time processing.Approach.Firstly, grouped convolution and conventional convolution are combined to replace the continuous conventional convolution in the model, and the depthwise convolution with stride is used to compress the feature map in the encoder modules. Secondly, additional identity connections and a local maximum and minimum enhancement module are designed, which can retain the detailed information and characteristic waveform in the ECG waveform while effectively denoising. Finally, we develop knowledge distillation in the experiments, which further improves the ECG denoising performance without increasing the model complexity. The ground-truth ECG is from The China Physiological Signal Challenge (CPSC) 2018, and the noise signal is from the MIT-BIH Noise Stress Test Database (NSTDB). We evaluate denoising performance using the signal-to-noise ratio (SNR), the root mean square error (RMSE) and the Pearson correlation coefficient (P). We use the floating point of operations (FLOPs) and parameters to calculate computational complexity.Main Results.Different data generation processes are used to conduct experiments: group 1, group 2 and group 3. The results show that the proposed model (U_Lde-net) can improve SNRs by 10.30 dB, 12.16 dB and 12.61 dB; reduce RMSEs by 9.88 × 10^-2, 20.63 × 10^-2and 15.25 × 10^-2; and increasePs by 14.77 × 10^-2, 27.74 × 10^-2and 21.32 × 10^-2. Moreover, the denoising performance after knowledge distillation is further improved. The U_Lde-net has parameters of 6.9 K and FLOPs of 6.6 M, which are much smaller than the compared models.Significance.We designed a lightweight model, but also retain adequate ECG denoising performance. We believe that this method can be successfully applied to practical applications under time or memory limits.

Heartbeat Classification and Arrhythmia Detection Using a Multi-Model Deep-Learning Technique

Cardiac arrhythmias pose a significant danger to human life; therefore, it is of utmost importance to be able to efficiently diagnose these arrhythmias promptly. There exist many techniques for the detection of arrhythmias; however, the most widely adopted method is the use of an Electrocardiogram (ECG). The manual analysis of ECGs by medical experts is often inefficient. Therefore, the detection and recognition of ECG characteristics via machine-learning techniques have become prevalent. There are two major drawbacks of existing machine-learning approaches: (a) they require extensive training time; and (b) they require manual feature selection. To address these issues, this paper presents a novel deep-learning framework that integrates various networks by stacking similar layers in each network to produce a single robust model. The proposed framework has been tested on two publicly available datasets for the recognition of five micro-classes of arrhythmias. The overall classification sensitivity, specificity, positive predictive value, and accuracy of the proposed approach are 98.37%, 99.59%, 98.41%, and 99.35%, respectively. The results are compared with state-of-the-art approaches. The proposed approach outperformed the existing approaches in terms of sensitivity, specificity, positive predictive value, accuracy and computational cost.

Channel-Wise Average Pooling and 1D Pixel-Shuffle Denoising Autoencoder for Electrode Motion Artifact Removal in ECG

This paper presents a channel-wise average pooling and one dimension pixel-shuffle architecture for a denoising autoencoder (CPDAE) design that can be applied to efficiently remove electrode motion (EM) artifacts in an electrocardiogram (ECG) signal. The three advantages of the proposed design are as follows: (1) In the skip connection layer, less memory is needed to transfer the features extracted by the neural network; (2) Pixel shuffle and pixel unshuffle techniques with point-wise convolution are used to effectively reserve the key features generated from each layer in both the encoder and decoder; (3) Overall, fewer parameters are required to reconstruct the ECG signal. This paper describes three deep neural network models, namely CPDAELite, CPDAERegular, and CPDAEFull, which support various computational capacity and hardware arrangements. The three proposed structures involve an encoder and decoder with six, seven, and eight layers, respectively. Furthermore, the CPDAELite, CPDAERegular, and CPDAEFull structures require fewer multiply-accumulate operations—355.01, 56.96, and 14.69 million, respectively—and less parameter usage—2.69 million, 149.7 thousand, and 55.5 thousand, respectively. To evaluate the denoising performance, the MIT–BIH noise stress test database containing six signal-to-noise ratios (SNRs) of noisy ECGs was employed. The results demonstrated that the proposed models had a higher improvement of SNR and lower percentage root-mean-square difference than other state-of-the-art methods under various conditions of SNR.

A Survey on Deep Learning-Based Change Detection from High-Resolution Remote Sensing Images

Change detection based on remote sensing images plays an important role in the field of remote sensing analysis, and it has been widely used in many areas, such as resources monitoring, urban planning, disaster assessment, etc. In recent years, it has aroused widespread interest due to the explosive development of artificial intelligence (AI) technology, and change detection algorithms based on deep learning frameworks have made it possible to detect more delicate changes (such as the alteration of small buildings) with the help of huge amounts of remote sensing data, especially high-resolution (HR) data. Although there are many methods, we still lack a deep review of the recent progress concerning the latest deep learning methods in change detection. To this end, the main purpose of this paper is to provide a review of the available deep learning-based change detection algorithms using HR remote sensing images. The paper first describes the change detection framework and classifies the methods from the perspective of the deep network architectures adopted. Then, we review the latest progress in the application of deep learning in various granularity structures for change detection. Further, the paper provides a summary of HR datasets derived from different sensors, along with information related to change detection, for the potential use of researchers. Simultaneously, representative evaluation metrics for this task are investigated. Finally, a conclusion of the challenges for change detection using HR remote sensing images, which must be dealt with in order to improve the model’s performance, is presented. In addition, we put forward promising directions for future research in this area.

Short Single-Lead ECG Signal Delineation-Based Deep Learning: Implementation in Automatic Atrial Fibrillation Identification

Physicians manually interpret an electrocardiogram (ECG) signal morphology in routine clinical practice. This activity is a monotonous and abstract task that relies on the experience of understanding ECG waveform meaning, including P-wave, QRS-complex, and T-wave. Such a manual process depends on signal quality and the number of leads. ECG signal classification based on deep learning (DL) has produced an automatic interpretation; however, the proposed method is used for specific abnormality conditions. When the ECG signal morphology change to other abnormalities, it cannot proceed automatically. To generalize the automatic interpretation, we aim to delineate ECG waveform. However, the output of delineation process only ECG waveform duration classes for P-wave, QRS-complex, and T-wave. It should be combined with a medical knowledge rule to produce the abnormality interpretation. The proposed model is applied for atrial fibrillation (AF) identification. This study meets the AF criteria with RR irregularities and the absence of P-waves in essential oscillations for even more accurate identification. The QT database by Physionet is utilized for developing the delineation model, and it validates with The Lobachevsky University Database. The results show that our delineation model works properly, with 98.91% sensitivity, 99.01% precision, 99.79% specificity, 99.79% accuracy, and a 98.96% F1 score. We use about 4058 normal sinus rhythm records and 1804 AF records from the experiment to identify AF conditions that are taken from three datasets. The comprehensive testing has produced higher negative predictive value and positive predictive value. This means that the proposed model can identify AF conditions from ECG signal delineation. Our approach can considerably contribute to AF diagnosis with these results.

Automated Detection of Caffeinated Coffee-Induced Short-Term Effects on ECG Signals Using EMD, DWT, and WPD

The effect of coffee (caffeinated) on electro-cardiac activity is not yet sufficiently researched. In the current study, the occurrence of coffee-induced short-term changes in electrocardiogram (ECG) signals was examined. Further, a machine learning model that can efficiently detect coffee-induced alterations in cardiac activity is proposed. The ECG signals were decomposed using three different joint time-frequency decomposition methods: empirical mode decomposition, discrete wavelet transforms, and wavelet packet decomposition with varying decomposition parameters. Various statistical and entropy-based features were computed from the decomposed coefficients. The statistical significance of these features was computed using Wilcoxon's signed-rank (WSR) test for significance testing. The results of the WSR tests infer a significant change in many of these parameters after the consumption of coffee (caffeinated). Further, the analysis of the frequency bands of the decomposed coefficients reveals that most of the significant change was localized in the lower frequency band (<22.5 Hz). Herein, the performance of nine machine learning models is compared and a gradient-boosted tree classifier is proposed as the best model. The results suggest that the gradient-boosted tree (GBT) model that was developed using a db2 mother wavelet at level 2 decomposition shows the highest mean classification accuracy of 78%. The outcome of the current study will open up new possibilities in detecting the effects of drugs, various food products, and alcohol on cardiac functionality.

A Proposal for a Data-Driven Approach to the Influence of Music on Heart Dynamics

Electrocardiographic signals (ECG) and heart rate viability measurements (HRV) provide information in a range of specialist fields, extending to musical perception. The ECG signal records heart electrical activity, while HRV reflects the state or condition of the autonomic nervous system. HRV has been studied as a marker of diverse psychological and physical diseases including coronary heart disease, myocardial infarction, and stroke. HRV has also been used to observe the effects of medicines, the impact of exercise and the analysis of emotional responses and evaluation of effects of various quantifiable elements of sound and music on the human body. Variations in blood pressure, levels of stress or anxiety, subjective sensations and even changes in emotions constitute multiple aspects that may well-react or respond to musical stimuli. Although both ECG and HRV continue to feature extensively in research in health and perception, methodologies vary substantially. This makes it difficult to compare studies, with researchers making recommendations to improve experiment planning and the analysis and reporting of data. The present work provides a methodological framework to examine the effect of sound on ECG and HRV with the aim of associating musical structures and noise to the signals by means of artificial intelligence (AI); it first presents a way to select experimental study subjects in light of the research aims and then offers possibilities for selecting and producing suitable sound stimuli; once sounds have been selected, a guide is proposed for optimal experimental design. Finally, a framework is introduced for analysis of data and signals, based on both conventional as well as data-driven AI tools. AI is able to study big data at a single stroke, can be applied to different types of data, and is capable of generalisation and so is considered the main tool in the analysis.

Indoor Positioning System in Learning Approach Experiments

The positioning system research strongly supports the development of location-based services used by related business organizations. However, location-based services with user experience still have many obstacles to overcome, including how to maintain a high level of position accuracy. From the literature studies reviewed, it is necessary to develop an indoor positioning system using fingerprinting based on Received Signal Strength (RSS). So far, the testing of the indoor positioning system has been carried out with an algorithm. But, in this research, with the proposed parameters, we will conduct experiments with a learning approach. The data tested is the signal service data on the device in the Dinamika Bangsa University building. The test was conducted with a deep learning approach using a deep neural network (DNN) algorithm. The DNN method can estimate the actual space and get better position results, whereas machine learning methods such as the DNN algorithm can handle more effectively large data and produce more accurate data. From the results of comparative testing with the learning approach between DNN, KNN, and SVM, it can be concluded that the evaluation with KNN is slightly better than the use of DNN in a single case. However, the results of KNN have low consistency; this is seen from the fluctuations in the movements of the R2 score and MSE values produced. Meanwhile, DNN gives a consistent value even though it has varied hidden layers. The Support Vector Machine (SVM) gives the worst value of these experiments, although, in the past, SVM was known as one of the favorite methods.

Automated detection of coronary artery disease, myocardial infarction and congestive heart failure using GaborCNN model with ECG signals

Cardiovascular diseases (CVDs) are main causes of death globally with coronary artery disease (CAD) being the most important. Timely diagnosis and treatment of CAD is crucial to reduce the incidence of CAD complications like myocardial infarction (MI) and ischemia-induced congestive heart failure (CHF). Electrocardiogram (ECG) signals are most commonly employed as the diagnostic screening tool to detect CAD. In this study, an automated system (AS) was developed for the automated categorization of electrocardiogram signals into normal, CAD, myocardial infarction (MI) and congestive heart failure (CHF) classes using convolutional neural network (CNN) and unique GaborCNN models. Weight balancing was used to balance the imbalanced dataset. High classification accuracies of more than 98.5% were obtained by the CNN and GaborCNN models respectively, for the 4-class classification of normal, coronary artery disease, myocardial infarction and congestive heart failure classes. GaborCNN is a more preferred model due to its good performance and reduced computational complexity as compared to the CNN model. To the best of our knowledge, this is the first study to propose GaborCNN model for automated categorizing of normal, coronary artery disease, myocardial infarction and congestive heart failure classes using ECG signals. Our proposed system is equipped to be validated with bigger database and has the potential to aid the clinicians to screen for CVDs using ECG signals.

An ECG Denoising Method Based on the Generative Adversarial Residual Network

High-quality and high-fidelity removal of noise in the Electrocardiogram (ECG) signal is of great significance to the auxiliary diagnosis of ECG diseases. In view of the single function of traditional denoising methods and the insufficient performance of signal details after denoising, a new method of ECG denoising based on the combination of the Generative Adversarial Network (GAN) and Residual Network is proposed. The method adopted in this paper is based on the GAN structure, and it restructures the generator and discriminator. In the generator network, residual blocks and Skip-Connecting are used to deepen the network structure and better capture the in-depth information in the ECG signal. In the discriminator network, the ResNet framework is used. In order to optimize the noise reduction process and solve the lack of local relevance considering the global ECG problem, the differential function and overall function of the maximum local difference are added in the loss function in this paper. The experimental results prove that the method used in this article has better performance than the current excellent S-Transform (S-T) algorithm, Wavelet Transform (WT) algorithm, Stacked Denoising Autoencoder (S-DAE) algorithm, and Improved Denoising Autoencoder (I-DAE) algorithm. Experiments show that the Root Mean Square Error (RMSE) of this method in the Massachusetts Institute of Technology and Beth Israel Hospital (MIT-BIH) noise pressure database is 0.0102, and the Signal-to-Noise Ratio (SNR) is 40.8526 dB, which is compared with that of the most advanced experimental methods. Our method improves the SNR by 88.57% on average. Besides the three noise intensities for comparison experiments, additional noise reduction experiments are also performed under four noise intensities in our paper. The experimental results verify the scientific nature of the model, which is that our method can effectively retain the important information conveyed by the original signal.

Intelligent Diagnosis Method Based on 2DECG Model

Electrophysiological signals can effectively reflect various physiological states of human body, and provide favorable basis for medical diagnosis. However, the correct analysis of electrophysiological signals requires professional medical diagnosis experience. With the rapid development of artificial intelligence, intelligent diagnosis methods based on deep learning are gradually applied in the medical field in order to reduce the dependence of diagnosis results on medical experience. Deep learning has made remarkable achievements in the field of image processing, through which deeper information can be extracted than through time-series signals. Therefore, this paper proposes a method of 2DECG diagnosis based on Faster R-CNN (Faster Region-based Convolutional Neural Network). First, the time-series ECG signal is transformed into two-dimensional curve. Then, the Faster R-CNN model based on beat is obtained by using dataset training. Finally, three kinds of ECG diseases are diagnosed by the Faster R-CNN model. The test results show that compared with the effect of one-dimensional CNN, the method proposed in this paper has high diagnosis accuracy and can help doctors to diagnose diseases more intuitively.

A Review of Plant Phenotypic Image Recognition Technology Based on Deep Learning

Plant phenotypic image recognition (PPIR) is an important branch of smart agriculture. In recent years, deep learning has achieved significant breakthroughs in image recognition. Consequently, PPIR technology that is based on deep learning is becoming increasingly popular. First, this paper introduces the development and application of PPIR technology, followed by its classification and analysis. Second, it presents the theory of four types of deep learning methods and their applications in PPIR. These methods include the convolutional neural network, deep belief network, recurrent neural network, and stacked autoencoder, and they are applied to identify plant species, diagnose plant diseases, etc. Finally, the difficulties and challenges of deep learning in PPIR are discussed.

Implementation of Autoencoders with Systolic Arrays through OpenCL

In the world of algorithm acceleration and the implementation of deep neural networks’ recall phase, OpenCL based solutions have a clear tendency to produce perfectly adapted kernels in graphic processor unit (GPU) architectures. However, they fail to obtain the same results when applied to field-programmable gate array (FPGA) based architectures. This situation, along with an enormous advance in new GPU architectures, makes it unfeasible to defend an acceleration solution based on FPGA, even in terms of energy efficiency. Our goal in this paper is to demonstrate that multikernel structures can be written based on classic systolic arrays in OpenCL, trying to extract the most advanced features of FPGAs without having to resort to traditional FPGA development using lower level hardware description languages (HDLs) such as Verilog or VHDL. This OpenCL methodology is based on the intensive use of channels (IntelFPGA extension of OpenCL) for the communication of both data and control and on the refinement of the OpenCL libraries using register transfer logic (RTL) code to improve the performance of the implementation of the base and activation functions of the neurons and, above all, to reflect the importance of adequate communication between the layers when implementing neuronal networks.

Improving Classification Accuracy of Hand Gesture Recognition Based on 60 GHz FMCW Radar with Deep Learning Domain Adaptation

With the recent development of small radars with high resolution, various human–computer interaction (HCI) applications using them have been developed. In particular, a method of applying a user’s hand gesture recognition using a short-range radar to an electronic device is being actively studied. In general, the time delay and Doppler shift characteristics that occur when a transmitted signal that is reflected off an object returns are classified through deep learning to recognize the motion. However, the main obstacle in the commercialization of radar-based hand gesture recognition is that even for the same type of hand gesture, recognition accuracy is degraded due to a slight difference in movement for each individual user. To solve this problem, in this paper, the domain adaptation is applied to hand gesture recognition to minimize the differences among users’ gesture information in the learning and the use stage. To verify the effectiveness of domain adaptation, a domain discriminator that cheats the classifier was applied to a deep learning network with a convolutional neural network (CNN) structure. Seven different hand gesture data were collected for 10 participants and used for learning, and the hand gestures of 10 users that were not included in the training data were input to confirm the recognition accuracy of an average of 98.8%.

Recommendation System Using Autoencoders

The magnitude of the daily explosion of high volumes of data has led to the emergence of the Big Data paradigm. The ever-increasing amount of information available on the Internet makes it increasingly difficult for individuals to find what they need quickly and easily. Recommendation systems have appeared as a solution to overcome this problem. Collaborative filtering is widely used in this type of systems, but high dimensions and data sparsity are always a main problem. With the idea of deep learning gaining more importance, several works have emerged to improve this type of filtering. In this article, a product recommendation system is proposed where an autoencoder based on a collaborative filtering method is employed. A comparison of this model with the Singular Value Decomposition is made and presented in the results section. Our experiment shows a very low Root Mean Squared Error (RMSE) value, considering that the recommendations presented to the users are in line with their interests and are not affected by the data sparsity problem as the datasets are very sparse, 0.996. The results are quite promising achieving an RMSE value of 0.029 in the first dataset and 0.010 in the second one.

Optimizing the Performance of Breast Cancer Classification by Employing the Same Domain Transfer Learning from Hybrid Deep Convolutional Neural Network Model

Breast cancer is a significant factor in female mortality. An early cancer diagnosis leads to a reduction in the breast cancer death rate. With the help of a computer-aided diagnosis system, the efficiency increased, and the cost was reduced for the cancer diagnosis. Traditional breast cancer classification techniques are based on handcrafted features techniques, and their performance relies upon the chosen features. They also are very sensitive to different sizes and complex shapes. However, histopathological breast cancer images are very complex in shape. Currently, deep learning models have become an alternative solution for diagnosis, and have overcome the drawbacks of classical classification techniques. Although deep learning has performed well in various tasks of computer vision and pattern recognition, it still has some challenges. One of the main challenges is the lack of training data. To address this challenge and optimize the performance, we have utilized a transfer learning technique which is where the deep learning models train on a task, and then fine-tune the models for another task. We have employed transfer learning in two ways: Training our proposed model first on the same domain dataset, then on the target dataset, and training our model on a different domain dataset, then on the target dataset. We have empirically proven that the same domain transfer learning optimized the performance. Our hybrid model of parallel convolutional layers and residual links is utilized to classify hematoxylin–eosin-stained breast biopsy images into four classes: invasive carcinoma, in-situ carcinoma, benign tumor and normal tissue. To reduce the effect of overfitting, we have augmented the images with different image processing techniques. The proposed model achieved state-of-the-art performance, and it outperformed the latest methods by achieving a patch-wise classification accuracy of 90.5%, and an image-wise classification accuracy of 97.4% on the validation set. Moreover, we have achieved an image-wise classification accuracy of 96.1% on the test set of the microscopy ICIAR-2018 dataset.

Deep Learning Models for Classification of Red Blood Cells in Microscopy Images to Aid in Sickle Cell Anemia Diagnosis

Sickle cell anemia, which is also called sickle cell disease (SCD), is a hematological disorder that causes occlusion in blood vessels, leading to hurtful episodes and even death. The key function of red blood cells (erythrocytes) is to supply all the parts of the human body with oxygen. Red blood cells (RBCs) form a crescent or sickle shape when sickle cell anemia affects them. This abnormal shape makes it difficult for sickle cells to move through the bloodstream, hence decreasing the oxygen flow. The precise classification of RBCs is the first step toward accurate diagnosis, which aids in evaluating the danger level of sickle cell anemia. The manual classification methods of erythrocytes require immense time, and it is possible that errors may be made throughout the classification stage. Traditional computer-aided techniques, which have been employed for erythrocyte classification, are based on handcrafted features techniques, and their performance relies on the selected features. They also are very sensitive to different sizes, colors, and complex shapes. However, microscopy images of erythrocytes are very complex in shape with different sizes. To this end, this research proposes lightweight deep learning models that classify the erythrocytes into three classes: circular (normal), elongated (sickle cells), and other blood content. These models are different in the number of layers and learnable filters. The available datasets of red blood cells with sickle cell disease are very small for training deep learning models. Therefore, addressing the lack of training data is the main aim of this paper. To tackle this issue and optimize the performance, the transfer learning technique is utilized. Transfer learning does not significantly affect performance on medical image tasks when the source domain is completely different from the target domain. In some cases, it can degrade the performance. Hence, we have applied the same domain transfer learning, unlike other methods that used the ImageNet dataset for transfer learning. To minimize the overfitting effect, we have utilized several data augmentation techniques. Our model obtained state-of-the-art performance and outperformed the latest methods by achieving an accuracy of 99.54% with our model and 99.98% with our model plus a multiclass SVM classifier on the erythrocytesIDB dataset and 98.87% on the collected dataset.