Preserving shape details of pulse signals for video-based blood pressure estimation.

In recent years, imaging photoplethysmograph (iPPG) pulse signals have been widely used in the research of non-contact blood pressure (BP) estimation, in which BP estimation based on pulse features is the main research direction. Pulse features are directly related to the shape of pulse signals while iPPG pulse signals are easily disturbed during the extraction process. To mitigate the impact of pulse feature distortion on BP estimation, it is necessary to eliminate interference while retaining valuable shape details in the iPPG pulse signal. Contact photoplethysmograph (cPPG) pulse signals measured at rest can be considered as the undisturbed reference signal. Transforming the iPPG pulse signal to the corresponding cPPG pulse signal is a method to ensure the effectiveness of shape details. However, achieving the required shape accuracy through direct transformation from iPPG to the corresponding cPPG pulse signals is challenging. We propose a method to mitigate this challenge by replacing the reference signal with an average cardiac cycle (ACC) signal, which can approximately represent the shape information of all cardiac cycles in a short time. A neural network using multi-scale convolution and self-attention mechanisms is developed for this transformation. Our method demonstrates a significant improvement in the maximal information coefficient (MIC) between pulse features and BP values, indicating a stronger correlation. Moreover, pulse signals transformed by our method exhibit enhanced performance in BP estimation using different model types. Experiments are conducted on a real-world database with 491 subjects in the hospital, averaging 60 years of age.

Predicting Arterial Stiffness From Single-Channel Photoplethysmography Signal: A Feature Interaction-Based Approach.

Arterial stiffness (AS) serves as a crucial indicator of arterial elasticity and function, typically requiring expensive equipment for detection. Given the strong correlation between AS and various photoplethysmography (PPG) features, PPG emerges as a convenient method for assessing AS. However, the limitations of independent PPG features hinder detection accuracy. This study introduces a feature selection method leveraging the interactive relationships between features to enhance the accuracy of predicting AS from a single-channel PPG signal. Initially, an adaptive signal interception method was employed to capture high-quality signal fragments from PPG sequences. 58 PPG features, deemed to have potential contributions to AS estimation, were extracted and analyzed. Subsequently, the interaction factor (IF) was introduced to redefine the interaction and redundancy between features. A feature selection algorithm (IFFS) based on the IF was then proposed, resulting in a combination of interactive features. Finally, the Xgboost model is utilized to estimate AS from the selected features set. The proposed approach is evaluated on datasets of 268 male and 124 female subjects, respectively. The results of AS estimation indicate that IFFS yields interacting features from numerous sources, rejects redundant ones, and enhances the association. The interaction features combined with the Xgboost model resulted in an MAE of 122.42 and 142.12 cm/sec, an SDE of 88.16 and 102.56 cm/sec, and a PCC of 0.88 and 0.85 for the male and female groups, respectively. The findings of this study suggest that the stated method improves the accuracy of predicting AS from single-channel PPG, which can be used as a non-invasive and cost-effective screening tool for atherosclerosis.

VidBP: Detecting Blood Pressure from Facial Videos with Personalized Calibration.

Recent studies have found that blood volume pulse (BVP) in facial videos contains features highly correlated to blood pressure (BP). However, the mapping from BVP features to BP varies from person to person. To address this issue, VidBP has been proposed as a BP detector that can be calibrated based on an individual's data. VidBP is pre-trained on a large dataset to extract BP-related features from BVP. Then, BVP samples and BP labels of an individual are fed into the pre-trained VidBP to create a personal dictionary of BP-related features. When estimating the individual's BP, the current BP-related feature is compared to the features saved in the dictionary, and the BP labels of the similar features are considered as the BP estimate. The performance of VidBP was evaluated on 640 samples of 16 subjects, and it demonstrated significantly lower errors in BP estimation compared to state-of-the-art methods. The personalized calibration of VidBP is a significant advantage, enabling it to better capture the unique mapping from BVP features to BP for each individual.Clinical relevance This study reports a feasible method to estimate BP from facial videos, providing a convenient and cost-effective way for home BP monitoring.

PFDNet: A Pulse Feature Disentanglement Network for Atrial Fibrillation Screening From Facial Videos.

Video-based Photoplethysmography (VPPG) can identify arrhythmic pulses during atrial fibrillation (AF) from facial videos, providing a convenient and cost-effective way to screen for occult AF. However, facial motions in videos always distort VPPG pulse signals and thus lead to the false detection of AF. Photoplethysmography (PPG) pulse signals offer a possible solution to this problem due to the high quality and resemblance to VPPG pulse signals. Given this, a pulse feature disentanglement network (PFDNet) is proposed to discover the common features of VPPG and PPG pulse signals for AF detection. Taking a VPPG pulse signal and a synchronous PPG pulse signal as inputs, PFDNet is pre-trained to extract the motion-robust features that the two signals share. The pre-trained feature extractor of the VPPG pulse signal is then connected to an AF classifier, forming a VPPG-driven AF detector after joint fine-tuning. PFDNet has been tested on 1440 facial videos of 240 subjects (50% AF absence and 50% AF presence). It achieves a Cohen's Kappa value of 0.875 (95% confidence interval: 0.840-0.910, P<0.001) on the video samples with typical facial motions, which is 6.8% higher than that of the state-of-the-art method. PFDNet shows significant robustness to motion interference in the video-based AF detection task, promoting the development of opportunistic screening for AF in the community.

TC-Net: A Transformer Capsule Network for EEG-based emotion recognition.

Deep learning has recently achieved remarkable success in emotion recognition based on Electroencephalogram (EEG), in which convolutional neural networks (CNNs) are the mostly used models. However, due to the local feature learning mechanism, CNNs have difficulty in capturing the global contextual information involving temporal domain, frequency domain, intra-channel and inter-channel. In this paper, we propose a Transformer Capsule Network (TC-Net), which mainly contains an EEG Transformer module to extract EEG features and an Emotion Capsule module to refine the features and classify the emotion states. In the EEG Transformer module, EEG signals are partitioned into non-overlapping windows. A Transformer block is adopted to capture global features among different windows, and we propose a novel patch merging strategy named EEG-PatchMerging (EEG-PM) to better extract local features. In the Emotion Capsule module, each channel of the EEG feature maps is encoded into a capsule to better characterize the spatial relationships among multiple features. Experimental results on two popular datasets (i.e., DEAP and DREAMER) demonstrate that the proposed method achieves the state-of-the-art performance in the subject-dependent scenario. Specifically, on DEAP (DREAMER), our TC-Net achieves the average accuracies of 98.76% (98.59%), 98.81% (98.61%) and 98.82% (98.67%) at valence, arousal and dominance dimensions, respectively. Moreover, the proposed TC-Net also shows high effectiveness in multi-state emotion recognition tasks using the popular VA and VAD models. The main limitation of the proposed model is that it tends to obtain relatively low performance in the cross-subject recognition task, which is worthy of further study in the future.

EEG-Based Seizure Prediction via Model Uncertainty Learning.

Deep neural networks (DNNs) have the powerful ability to automatically extract efficient features, which makes them prominent in electroencephalogram (EEG) based seizure prediction tasks. However, current research in this field cannot take the model uncertainty into account, causing the prediction less credible. To this end, we introduce a novel end-to-end patient-specific seizure prediction framework via model uncertainty learning. Specifically, we propose a reparameterized EEG-based lightweight CNN architecture and a modified Monte Carlo dropout (RepNet-MMCD) strategy to improve the reliability of the DNNs-based model. In RepNet, we obtain multi-scale feature representations by applying depthwise separable convolutions of different kernels. After training, depthwise convolutions with different scales are equivalently converted into a single convolution layer, which can greatly reduce computational budgets without losing model performance. In addition, we propose a modified Monte Carlo (MMCD) strategy, leveraging the samples-based temporal information in EEG signals to simulate the Monte Carlo dropout sampling. Sensitivity, false-positive rate (FPR), and area under curve (AUC) of the proposed RepNet-MMCD achieve 93.1%, 0.033/h, 0.950 and 81.6%, 0.056/h, 0.903 on two public datasets, respectively. We further extend the MMCD strategy to the other baseline methods, which can improve the performance of seizure prediction by a clear margin.

Patient-Specific Seizure Prediction via Adder Network and Supervised Contrastive Learning.

Deep learning (DL) methods have been widely used in the field of seizure prediction from electroencephalogram (EEG) in recent years. However, DL methods usually have numerous multiplication operations resulting in high computational complexity. In addtion, most of the current approaches in this field focus on designing models with special architectures to learn representations, ignoring the use of intrinsic patterns in the data. In this study, we propose a simple and effective end-to-end adder network and supervised contrastive learning (AddNet-SCL). The method uses addition instead of the massive multiplication in the convolution process to reduce the computational cost. Besides, contrastive learning is employed to effectively use label information, points of the same class are clustered together in the projection space, and points of different class are pushed apart at the same time. Moreover, the proposed model is trained by combining the supervised contrastive loss from the projection layer and the cross-entropy loss from the classification layer. Since the adder networks uses the l1 -norm distance as the similarity measure between the input feature and the filters, the gradient function of the network changes, an adaptive learning rate strategy is employed to ensure the convergence of AddNet-CL. Experimental results show that the proposed method achieves 94.9% sensitivity, an area under curve (AUC) of 94.2%, and a false positive rate of (FPR) 0.077/h on 19 patients in the CHB-MIT database and 89.1% sensitivity, an AUC of 83.1%, and an FPR of 0.120/h in the Kaggle database. Competitive results show that this method has broad prospects in clinical practice.

Emotion recognition from EEG based on multi-task learning with capsule network and attention mechanism.

Deep learning (DL) technologies have recently shown great potential in emotion recognition based on electroencephalography (EEG). However, existing DL-based EEG emotion recognition methods are built on single-task learning, i.e., learning arousal, valence, and dominance individually, which may ignore the complementary information of different tasks. In addition, single-task learning involves a new round of training every time a new task appears, which is time consuming. To this end, we propose a novel method for EEG-based emotion recognition based on multi-task learning with capsule network (CapsNet) and attention mechanism. First, multi-task learning can learn multiple tasks simultaneously while exploiting commonalities and differences across tasks, it can also obtain more data from different tasks, which can improve generalization and robustness. Second, the innovative structure of the CapsNet enables it to effectively characterize the intrinsic relationship among various EEG channels. Finally, the attention mechanism can change the weight of different channels to extract important information. In the DEAP dataset, the average accuracy reached 97.25%, 97.41%, and 98.35% on arousal, valence, and dominance, respectively. In the DREAMER dataset, average accuracy reached 94.96%, 95.54%, and 95.52% on arousal, valence, and dominance, respectively. Experimental results demonstrate the efficiency of the proposed method for EEG emotion recognition.

Bi-CapsNet: A Binary Capsule Network for EEG-based Emotion Recognition.

In recent years, deep learning has gained widespread attention in electroencephalogram (EEG)-based emotion recognition. However, deep learning methods are usually time-consuming with a large amount of memory usage, which obstructs their practical usage on resource-constrained devices. In this paper, we propose a binary capsule network (Bi-CapsNet) for EEG emotion recognition with low computational cost and memory usage. The Bi-CapsNet binarizes 32-bit weights and activations to 1 bit, and replaces floating-point operations with efficient bitwise operations. To address the issue of function discontinuity in backward propagation, we use a continuous function to approximate the binarization process. Two popular EEG emotion databases, namely, DEAP and DREAMER, are used for performance evaluation. In comparison to its full-precision counterpart, the Bi-CapsNet achieves a reduction on the computational cost and a reduction on the memory usage, while with only a 1% drop on the recognition accuracy. Compared to some state-of-the-art EEG emotion recognition methods, the proposed method obtains more competitive performance. In addition, the Bi-CapsNet is implemented on a mobile phone via an open-source binary inference framework named Bolt, and it achieves an  âˆ¼ 5× inference acceleration in comparison to its full-precision counterpart.

Prediction of arterial blood pressure waveforms from photoplethysmogram signals via fully convolutional neural networks.

Cardiovascular disease (CVD) is one of the most serious diseases threatening human health. Arterial blood pressure (ABP) waveforms, containing vivid cardiovascular information, are of great significance for the diagnosis and the prevention of CVD. This paper proposes a deep learning model, named ABP-Net, to transform photoplethysmogram (PPG) signals into ABP waveforms that contain vital physiological information related to cardiovascular systems. In order to guarantee the quality of the predicted ABP waveforms, the structure of the network, the input signals and the loss functions are carefully designed. Specifically, a Wave-U-Net, one kind of fully convolutional neural networks (CNN), is taken as the core architecture of the ABP-Net. Besides the original PPG signals, its first derivative and second derivative signals are all utilized as the inputs of the ABP-Net. Additionally, the maximal absolute loss, accompany with the mean squared error loss is employed to ensure the match of the predicted ABP waveform with the reference one. The performance of the proposed ABP network is tested on the public MIMIC II database both in subject-dependent and subject-independent manners. Both results verify the superior performance of the proposed model over those existing methods accordingly. The mean absolute error (MAE) and the root-mean-square error (RMSE) between the predicted waveforms via the ABP-Net and the reference ones are 3.20 mmHg and 4.38 mmHg during the subject-dependent experiments while those are 5.57 mmHg and 7.15 mmHg during the subject-independent experiments. Benefiting from the predicted high-quality ABP waveforms, more ABP related physiological parameters can be better obtained, which effectively expands the application scope of PPG devices.

Motion robust ICG measurements using a two-step spectrum denoising method.

Objective. Impedance cardiography (ICG) is a noninvasive and continuous method for evaluating stroke volume and cardiac output. However, the ICG measurement is easily interfered due to respiration and body movements. Taking into consideration about the spectral correlations between the simultaneously collected ICG, electrocardiogram (ECG), and acceleration signals, this paper introduces a two-step spectrum denoising method to remove motion artifacts of ICG measurements in both resting and exercising scenarios.Approach. First, the major motion artifacts of ECG and ICG are separately suppressed by the spectral subtraction with respect to acceleration signals. The obtained ECG and ICG are further decomposed into two sets of intrinsic mode functions (IMFs) through the ensemble empirical mode decomposition. We then extract the shared spectral information between the two sets of IMFs using the canonical correlation analysis in a spectral domain. Finally, the ICG signal is reconstructed using those canonical variates with largest spectral correlations with ECG IMFs.Main results. The denoising method was evaluated for 30 subjects under both resting and cycling scenarios. Experimental results show that the beat contribution factor of ICG signals increases from its original 80.1%-97.4% after removing the motion artifacts.Significance. The proposed denoising scheme effectively improves the reliability of diagnosis and analysis on cardiovascular diseases relying on ICG signals.

PulseGAN: Learning to Generate Realistic Pulse Waveforms in Remote Photoplethysmography.

Remote photoplethysmography (rPPG) is a non-contact technique for measuring cardiac signals from facial videos. High-quality rPPG pulse signals are urgently demanded in many fields, such as health monitoring and emotion recognition. However, most of the existing rPPG methods can only be used to get average heart rate (HR) values due to the limitation of inaccurate pulse signals. In this paper, a new framework based on generative adversarial network, called PulseGAN, is introduced to generate realistic rPPG pulse signals through denoising the chrominance (CHROM) signals. Considering that the cardiac signal is quasi-periodic and has apparent time-frequency characteristics, the error losses defined in time and spectrum domains are both employed with the adversarial loss to enforce the model generating accurate pulse waveforms as its reference. The proposed framework is tested on three public databases. The results show that the PulseGAN framework can effectively improve the waveform quality, thereby enhancing the accuracy of HR, the interbeat interval (IBI) and the related heart rate variability (HRV) features. The proposed method significantly improves the quality of waveforms compared to the input CHROM signals, with the mean absolute error of AVNN (the average of all normal-to-normal intervals) reduced by 41.19%, 40.45%, 41.63%, and the mean absolute error of SDNN (the standard deviation of all NN intervals) reduced by 37.53%, 44.29%, 58.41%, in the cross-database test on the UBFC-RPPG, PURE, and MAHNOB-HCI databases, respectively. This framework can be easily integrated with other existing rPPG methods to further improve the quality of waveforms, thereby obtaining more reliable IBI features and extending the application scope of rPPG techniques.

Emotion Recognition From Multi-Channel EEG via Deep Forest.

Recently, deep neural networks (DNNs) have been applied to emotion recognition tasks based on electroencephalography (EEG), and have achieved better performance than traditional algorithms. However, DNNs still have the disadvantages of too many hyperparameters and lots of training data. To overcome these shortcomings, in this article, we propose a method for multi-channel EEG-based emotion recognition using deep forest. First, we consider the effect of baseline signal to preprocess the raw artifact-eliminated EEG signal with baseline removal. Secondly, we construct 2 D frame sequences by taking the spatial position relationship across channels into account. Finally, 2 D frame sequences are input into the classification model constructed by deep forest that can mine the spatial and temporal information of EEG signals to classify EEG emotions. The proposed method can eliminate the need for feature extraction in traditional methods and the classification model is insensitive to hyperparameter settings, which greatly reduce the complexity of emotion recognition. To verify the feasibility of the proposed model, experiments were conducted on two public DEAP and DREAMER databases. On the DEAP database, the average accuracies reach to 97.69% and 97.53% for valence and arousal, respectively; on the DREAMER database, the average accuracies reach to 89.03%, 90.41%, and 89.89% for valence, arousal and dominance, respectively. These results show that the proposed method exhibits higher accuracy than the state-of-art methods.

Multi-channel EEG-based emotion recognition via a multi-level features guided capsule network.

In recent years, deep learning (DL) techniques, and in particular convolutional neural networks (CNNs), have shown great potential in electroencephalograph (EEG)-based emotion recognition. However, existing CNN-based EEG emotion recognition methods usually require a relatively complex stage of feature pre-extraction. More importantly, the CNNs cannot well characterize the intrinsic relationship among the different channels of EEG signals, which is essentially a crucial clue for the recognition of emotion. In this paper, we propose an effective multi-level features guided capsule network (MLF-CapsNet) for multi-channel EEG-based emotion recognition to overcome these issues. The MLF-CapsNet is an end-to-end framework, which can simultaneously extract features from the raw EEG signals and determine the emotional states. Compared with original CapsNet, it incorporates multi-level feature maps learned by different layers in forming the primary capsules so that the capability of feature representation can be enhanced. In addition, it uses a bottleneck layer to reduce the amount of parameters and accelerate the speed of calculation. Our method achieves the average accuracy of 97.97%, 98.31% and 98.32% on valence, arousal and dominance of DEAP dataset, respectively, and 94.59%, 95.26% and 95.13% on valence, arousal and dominance of DREAMER dataset, respectively. These results show that our method exhibits higher accuracy than the state-of-the-art methods.

New insights on super-high resolution for video-based heart rate estimation with a semi-blind source separation method.

Remote photoplethysmography (rPPG), a non-contact technique to estimate heart rates (HR) from video recordings, has attracted much attention from researchers in recent years. It is well-known that rPPG signals can be extracted from low-resolution videos. However, the measurement quality may degrade due to camera quantization noise if only a small number of pixels are within the skin region of interest. The purpose of this paper is to comprehensively investigate the benefit of using a super-high resolution for the rPPG-based HR estimation under various shooting distances. A new semi-blind source separation (semi-BSS) rPPG method, which is proposed to combine the advantages of BSS and model-based methods, is fully tested on both the public UBFC-RPPG and self-collected video datasets. The experimental results demonstrate that the new semi-BSS method outperforms several existing techniques. A consistent and remarkable improvement on the rPPG signal quality has been observed with the super-high resolution when the shooting distance is no less than 1.0 m. This indicates that selecting an appropriate resolution based on a given shooting distance also plays a crucial role to improve the quality of rPPG measurements.

Application of T-matrix method in solving mixed boundary separable obstacle problem.

The practical problem of imaging scatterers enclosed by separable obstacles with mixed boundary is addressed. Both the unknown scatterers and the known obstacle media can be mixture of dielectric and perfect electric conducting (PEC) materials. The scattering phenomenon of such problem is well modeled by T-matrix method. By usage of separable prior information, the obstacle media are treated as known scatterers rather than part of the background. The number of unknowns is thus reduced greatly. After recovering the profiles of scatterers by T-matrix method, a criterion is further provided to classify the PEC and dielectric scatterers. Various numerical examples are presented to show the effectiveness and good performance of the method.

Imaging three-dimensional anisotropic scatterers in multilayered medium by multiple signal classification method with enhanced resolution.

This paper investigates the resolution and robustness of the multiple signal classification (MUSIC) method to locate small three-dimensional (3D) anisotropic scatterers near the medium interface in a multilayered background. An enhanced MUSIC algorithm developed for free-space background is extended to solve such a problem. Because its indicator is built in a stable signal subspace, which is continuous across the medium interface, better stability and higher resolution against noise are observed for the proposed method compared to the known standard MUSIC method. Numerical simulations with various medium interfaces and noise levels are conducted to verify the performance of the introduced MUSIC method.

Imaging small three-dimensional elastic inclusions by an enhanced multiple signal classification method.

In this paper, an enhanced multiple signal classification (MUSIC) algorithm is introduced to retrieve small three-dimensional elastic inclusions. First, the multistatic response (MSR) matrix is built by two different ways depending on considering the multiple scattering effect or not. The eigenvalue structure of the MSR matrix is analyzed to identify the degenerate case of inclusions. Second, the pseudo-spectrum function is built with an optimal test direction on each node, where a balancing technique is employed to ensure the numerical stability. Benefitting from this MUSIC indicator, the degenerate inclusions can be located with a good performance against noise. Numerical simulations show the proposed method has wider applicability, better resolution, and more robust in the presence of noise than the standard MUSIC methods.

Electromagnetic imaging of separable obstacle problem.

The practical problem of imaging scatterers that are separable from the known obstacles is addressed. Using such a priori information, the obstacle is regarded as a known scatterer rather than part of the background and can be excluded from the retrieving process by reformulating the cost function. As a result, the proposed method transforms the problem into an inverse scattering problem with homogeneous background, and avoids the computationally intensive calculation of Green's function for inhomogeneous background (bases of the physical model of the problem). Meanwhile, the factors that influence the imaging quality for such kind of problem are also analyzed. Various difficult numerical examples are presented to show the good performance of our method. In addition, a data set of scattering experiments from the Institut Fresnel is tested to verify the validity of our method.