Search

• We investigate the application of the contactless BCG signal in emotion recognition. Emotion recognition is a major research point in the field of affective computing. Existing research on the application of physiological signals to emotion recognition mainly focuses on the processing of contact signals. However, there are issues with contact signal acquisition equipment, such as limited portability and poor user compliance, which make it difficult to promote its use. To explore a new method for emotion recognition based on contactless ballistocardiogram (BCG), we proposed a SE-CNN model with a multi-class focal loss function. To construct the dataset, we used audio-visual stimuli to evoke the subjects' emotions and collected data on the subjects' three discrete emotions, positive, neutral, and negative, through our established BCG signal acquisition system based on a piezoelectric ceramics sensor. Root mean square filter and thresholding were used to detect and eliminate motion artifacts of BCG signals. We did two kinds of preprocessing on BCG signals: wavelet transform and bandpass filtering, to explore the effect of different components of BCG on emotion recognition. Subsequently, we verified the model's performance and cross-time working ability through traditional K-Fold and our proposed K-Session cross-validation methods. The results showed that the band-pass filtering method was more beneficial to the current classification task. Under K-Fold cross-validation, the model's accuracy, precision, and recall were 97.21%, 97.00%, and 97.11%. Under K-Session cross-validation, the model's accuracy, precision, and recall were 94.66%, 93.92%, and 94.86%, respectively, all of which were better than the classification effect of synchronous ECG. The reliability of BCG in contactless emotion recognition was proved. [ABSTRACT FROM AUTHOR]

Silent speech recognition (SSR) is a system that can recognize verbal expressions when speech signals are not accessible. It has found versatile applications not only in helping those who suffer a speech impediment, but also in human–computer interaction. In this paper, we propose an SSR system based on surface electromyography (sEMG) signals. We first design a Mandarin corpus which contains 135-class utterances and collect 8-channel sEMG data from 12 subjects. After preprocessing the raw sEMG signals, we extract a time–frequency hybrid feature map for each sample. Then, we use two data augmentation strategies to obtain more data in order to improve the performance of our model. Finally, we test our proposed models on two tasks: the multi-subject mixed classification task and the single subject classification task. We use a 5-fold cross-validation to evaluate our model on both tasks. The GRU model performs best, achieving a recognition accuracy of 88.01% on the multi-subject mixed classification task and a maximum recognition accuracy of 97.19% on the single subject classification task. Experiments results demonstrate the effectiveness of our proposed system. • A silent speech system based on surface electromyography signals and deep learning models. • Time–frequency hybrid features are extracted from surface electromyography signals. • Speed perturbation and time–frequency masking are used for data augmentation. • Several deep learning models are evaluated on two different tasks; The GRU model performed the best. [ABSTRACT FROM AUTHOR]

• The MIF-AFNet can fuse the deep features of RRIs and ECG for AF binary detection. • The data augmentation method of vertical flipping can eliminate the morphological imbalance, thus effectively alleviating overfitting. • The MIF-AFNet is five-fold cross-validated on the CPSC2021 dataset and independently tested on four external databases. • The performance results verify excellent generalization performance for the method. The reliable detection of atrial fibrillation (AF) is important for diagnosing the disease, tracking its progression, and developing individualized care strategies. However, models based on limited data are prone to data dependency due to differences in data feature distribution, which generally degrades their performance on unseen external datasets. In this work, we propose a multi-input fusion AF detection network (MIF-AFNet), which cascades residual convolutional neural networks and bidirectional long short-term memory networks to capture the deep features of electrocardiogram (ECG) and RR intervals (RRIs), respectively. Additionally, the ECG signals use a data augmentation method to alleviate the morphological imbalance. MIF-AFNet learns a robust feature representation for accurate AF detection by fusing the available information from RRIs and ECG. The proposed method was developed and evaluated using 5 long-term ECG datasets (CPSC2021, AFDB, LTAF, MITDB, and NSRDB) from PhysioNet. The subject-wise five-fold cross-validation was performed on CPSC2021, and the proposed method achieved an AF detection accuracy of 98.63%. The generalization performance is further evaluated on four external independent datasets (AFDB, LTAF, MITDB, and NSRDB), achieving accuracies of 98.63%, 97.04%, 98.07%, and 100%, respectively. The results show that the proposed method can accurately detect AF from long-term ECG recordings. In addition, the low complexity of the model makes it less demanding on computing resources. Therefore, it has the potential to improve the automatic diagnosis and management of AF in wearable device-based long-term home monitoring. [ABSTRACT FROM AUTHOR]

Heartbeat is a crucial vital sign, and radio-frequency technology can measure the micro-vibration of the body surface from a distance and extract the heartbeat signal, which can solve the compliance problem caused by wearable sensors. However, most existing works verify short-term controlled experiments and seldom consider the challenges posed by random body movements and posture changes during long-term monitoring in daily life scenarios. In this article, we propose a systematic solution called Radar-Beat, which enables accurate heartbeat monitoring using an mmWave FMCW radar device. Our goal is to promote the integration of contactless heartbeat monitoring into life scenarios. We proposes a sensitive body motion detection algorithm and an optimal Range-bin selection algorithm, which can automatically identify body motion and update the best heartbeat signal channel. Then, we construct a personalized heartbeat template for each signal segment and propose a global optimization model to improve the accuracy of heartbeat length estimation. We use the synchronized ECG signal as the ground truth, results show strong agreement between Radar-Beat and synchrony ECG devices for heart rate and inter-beat interval (IBI) measurements in healthy subjects. In the heartbeat monitoring experiment for a total of 72 h and 56 min involving 11 participants, under a time coverage of 91.2%, the median error of IBI estimation was 12 ms. Radar-Beat has strong robustness to different individuals in different postures and positions. It can be deployed in a hospital bed or home to enable continuous heartbeat monitoring in an unobtrusive way. • Non-contact FMCW radar enables beat-by-beat heartbeat monitoring in real-life scenarios. • Accurate automatic detection algorithm for body motion proposed. • Adaptive algorithm selects range bins for high-quality heartbeat in varied positions. • A template matching-based global optimization model achieves precise IBI estimation. [ABSTRACT FROM AUTHOR]

• Rapid and accurate determination of the volume set point V0 through PVM. • Continuous tracking of V0 changes using UVTM. • Development of a non-invasive continuous blood pressure monitor. • Experimental results demonstrate the high accuracy and reliability of PVM and UVTM. The volume-clamp method (VCM) is a widely used non-invasive continuous blood pressure (BP) monitoring technique, but determining the volume setpoint (V 0) in the open-loop phase can be time-consuming and may cause finger tissue displacement. Additionally, frequent interruptions in BP measurement are required during the closed-loop phase to reset V 0. We proposed two new approaches: the proportional value method (PVM) for rapid V 0 determination in the open-loop phase and the uninterrupted V 0 tracking method (UVTM) in the closed-loop phase. PVM adaptively searches for V 0 by analyzing the shape index (Prop) of the plethysmographic (PG) signal under constant cuff pressure, while UVTM introduces a new vascular compliance calculation method in closed-loop phase to track changes in V 0 without interrupting BP measurement. We describe the non-invasive continuous BP monitor (NC-BPM) based on these methods. Experimental results show that PVM can accurately determine V 0 , with a mean arterial pressure error of −1.72 ± 2.80 mmHg when compared to oscillometric method (OSCM). UVTM in NC-BPM produced a mean absolute error (MAE) of 2.44 ± 2.91 mmHg for mean arterial pressure (MAP) in induced BP fluctuation test compared to the Nexfin monitor. In the clinical study with invasive reference, the MAE in measuring MAP by NC-BPM was 4.69 ± 5.39 mmHg. Our results show that PVM and UVTM are highly accurate and reliable, with the potential to integrate effectively into VCM-based BP monitoring devices that could benefit patients, healthcare providers, and researchers. [ABSTRACT FROM AUTHOR]

Heartbeat is a crucial vital sign, and radio-frequency technology can measure the micro-vibration of the body surface from a distance and extract the heartbeat signal, which can solve the compliance problem caused by wearable sensors. However, most existing works verify short-term controlled experiments and seldom consider the challenges posed by random body movements and posture changes during long-term monitoring in daily life scenarios. In this article, we propose a systematic solution called Radar-Beat, which enables accurate heartbeat monitoring using an mmWave FMCW radar device. Our goal is to promote the integration of contactless heartbeat monitoring into life scenarios. We proposes a sensitive body motion detection algorithm and an optimal Range-bin selection algorithm, which can automatically identify body motion and update the best heartbeat signal channel. Then, we construct a personalized heartbeat template for each signal segment and propose a global optimization model to improve the accuracy of heartbeat length estimation. We use the synchronized ECG signal as the ground truth, results show strong agreement between Radar-Beat and synchrony ECG devices for heart rate and inter-beat interval (IBI) measurements in healthy subjects. In the heartbeat monitoring experiment for a total of 72 h and 56 min involving 11 participants, under a time coverage of 91.2%, the median error of IBI estimation was 12 ms. Radar-Beat has strong robustness to different individuals in different postures and positions. It can be deployed in a hospital bed or home to enable continuous heartbeat monitoring in an unobtrusive way. • Non-contact FMCW radar enables beat-by-beat heartbeat monitoring in real-life scenarios. • Accurate automatic detection algorithm for body motion proposed. • Adaptive algorithm selects range bins for high-quality heartbeat in varied positions. • A template matching-based global optimization model achieves precise IBI estimation. [ABSTRACT FROM AUTHOR]

• Rapid and accurate determination of the volume set point V0 through PVM. • Continuous tracking of V0 changes using UVTM. • Development of a non-invasive continuous blood pressure monitor. • Experimental results demonstrate the high accuracy and reliability of PVM and UVTM. The volume-clamp method (VCM) is a widely used non-invasive continuous blood pressure (BP) monitoring technique, but determining the volume setpoint (V 0) in the open-loop phase can be time-consuming and may cause finger tissue displacement. Additionally, frequent interruptions in BP measurement are required during the closed-loop phase to reset V 0. We proposed two new approaches: the proportional value method (PVM) for rapid V 0 determination in the open-loop phase and the uninterrupted V 0 tracking method (UVTM) in the closed-loop phase. PVM adaptively searches for V 0 by analyzing the shape index (Prop) of the plethysmographic (PG) signal under constant cuff pressure, while UVTM introduces a new vascular compliance calculation method in closed-loop phase to track changes in V 0 without interrupting BP measurement. We describe the non-invasive continuous BP monitor (NC-BPM) based on these methods. Experimental results show that PVM can accurately determine V 0 , with a mean arterial pressure error of −1.72 ± 2.80 mmHg when compared to oscillometric method (OSCM). UVTM in NC-BPM produced a mean absolute error (MAE) of 2.44 ± 2.91 mmHg for mean arterial pressure (MAP) in induced BP fluctuation test compared to the Nexfin monitor. In the clinical study with invasive reference, the MAE in measuring MAP by NC-BPM was 4.69 ± 5.39 mmHg. Our results show that PVM and UVTM are highly accurate and reliable, with the potential to integrate effectively into VCM-based BP monitoring devices that could benefit patients, healthcare providers, and researchers. [ABSTRACT FROM AUTHOR]

• A novel end-to-end deep learning network for cross-scenario automatic sleep staging. • A joint distribution loss and a stage transition loss to solve the shift data distribution among domains. • The performance of sleep staging has been increased after using the proposed method. Sleep staging via manual inspection is time-consuming and inefficient, and as such, automatic sleep stage classification methods have been proposed and successfully implemented. However, these conventional supervised learning models are not suitable for cross-scenario sleep staging tasks due to the shift data distribution problem. This study aims to explore the use of transfer learning to transfer knowledge from a labeled source dataset to a new target domain where labels are difficult to obtain, so as to help solve the classification problem in a new domain in practical sleep staging scenarios. A novel end-to-end deep learning network that includes a feature extractor, a sleep stage classifier, and a domain adaptation network was proposed. The proposed domain adaptation network can accomplish distribution alignment between source and target domains, through the application of both joint distribution loss and stage transition loss. Cross-channel, cross-subject and channel, and cross-dataset experiments were designed to verify the feasibility of the proposed network using three publicly available datasets. The results indicate an average accuracy improvement ranging from 0.1% to 6.1% (average of 2.9%), a macro-averaging F1-score improvement ranging from 0.1% to 5.1% (average of 2.6%), and a Cohen's kappa coefficient improvement ranging from 0.016 to 0.083 (average of 0.045) after using transfer learning, when compared with a non-transfer model. Relative to similar existing transfer learning methods for sleep staging, the proposed method was implemented in an unsupervised manner and achieved success in cross-scenario sleep staging tasks, which is more suitable for practical applications in daily life. [ABSTRACT FROM AUTHOR]

This paper reports a Film Bulk Acoustic Resonator (FBAR) with micro through-hole array in the top electrode. When the FBAR device is used as a sensor based on adsorption and its piezoelectric actuation layer under the top electrode acts as sensitive layer, the through-hole array provide paths for gas to reach the sensitive layer directly and improves sensitivity of the sensor. In addition, thermal modules, a temperature sensor and a heater, are added to monitor and control working temperature of the resonator, and high temperature is available to enhance thermal desorption. We applied this FBAR system to a humidity sensor. The results show that the micro through-hole array design made sensitivity rise to 18.6 ppm/1%, namely 21.2 kHz/1%, which was 18 times higher than samples without. Meanwhile, the thermal modules kept working temperature stable and made it vary with the power supplied. [ABSTRACT FROM AUTHOR]

This paper reports a Film Bulk Acoustic Resonator (FBAR) with micro through-hole array in the top electrode. When the FBAR device is used as a sensor based on adsorption and its piezoelectric actuation layer under the top electrode acts as sensitive layer, the through-hole array provide paths for gas to reach the sensitive layer directly and improves sensitivity of the sensor. In addition, thermal modules, a temperature sensor and a heater, are added to monitor and control working temperature of the resonator, and high temperature is available to enhance thermal desorption. We applied this FBAR system to a humidity sensor. The results show that the micro through-hole array design made sensitivity rise to 18.6 ppm/1%, namely 21.2 kHz/1%, which was 18 times higher than samples without. Meanwhile, the thermal modules kept working temperature stable and made it vary with the power supplied. [ABSTRACT FROM AUTHOR]

• An end-to-end domain adaptation network is proposed for subject-specific sleep staging. • A single-lead Electrocardiograph (ECG) signal is utilized as the input. • Multi-class focal loss and a domain aligning layer are combined to solve data imbalances and domain shifts. • State-of-the-art performance of sleep staging based on single-lead ECG is obtained on the public SHHS2, SHHS1, and MESA datasets, respectively. • The method has excellent performance for accuracy and Cohen's Kappa in terms of cross datasets. Sleep plays a vital role in human physical and mental health. To accurately identify sleep structure under comfortable and convenient conditions, many machine-learning methods have been applied in the classification of sleep staging based on an Electrocardiogram (ECG); however, few works have solved the problem of generalization in the subject-specific sleep staging. The main reason for the problem is the difference of domain distribution across subjects. Most works are also classified based on ECG-derived signals and manual features. In this paper, we describe an end-to-end deep adaptation framework that classifies sleep stages into four classes based on a single-lead ECG to overcome the above problems. In particular, multi-class focal loss and a domain aligning layer based on the maximum mean discrepancy have been combined to solve data imbalances and domain shifts during three-step processing. We evaluate the method based on the three public datasets contained in SHHS2, SHHS1, and MESA. Compared to the performance of the model without domain aligning, the accuracy of the model for the public datasets has been improved by more than 20%, and the Kappa coefficient has also been improved by close to 0.4, which achieves state-of-the-art solutions compared to the baseline. In addition, the method has excellent performance for accuracy and Cohen's Kappa in terms of cross datasets. The proposed method, which confuses the domain-variant features, makes important contributions to the prediction of different subjects' sleep structures. The domain-adaptation setup, which varies across subjects and across environments, may provide a new approach to health monitoring. [ABSTRACT FROM AUTHOR]

• We explore radar-based heartbeat identification under open-set condition initially. • We propose a novel DDLM model without heartbeat segmentation and feature engineering. • The method shows great effectiveness in both open-set and closed-set environment. Non-contact continuous biometric identification system based on the heartbeat signals has attracted more attention due to its privacy-friendly properties. Current methods, however, mostly focused on the traditional machine learning methods under the close-set condition. This paper aims to investigate the feasibility of using the cardiac radar heartbeat signals and deep learning techniques to identify person in the open-set environment. A novel dipole deep learning model (DDLM) was proposed for the open-set person identification problem without heartbeat segmentation and feature engineering. The normalized heartbeat samples with time duration of 5 s were used as input and encoded into the feature space, where the encoded features of the same person cluster closely around the corresponding negative pole and repel far from positive pole, and those of different persons separate loosely from each other. Finally, threshold on the distance from the features to the dipoles in the feature space was set for each known identity. Extensive experiments conducted on a public dataset of clinically recorded vital signs indicate that:(1) The proposed model shows high stability under close-set condition with an accuracy higher than 99 % with 30 subjects.(2) The accuracy and the F1-score attain 93.42 % and 93.57 % under the open-set condition with the maximum openness of 29.3 %, respectively. The proposed model shows high effectiveness in person identification using heartbeat signals. The DDLM outperforms most of the existing methods under the close-set condition. And the DDLM shows a promising future for person identification in open-set environment. [ABSTRACT FROM AUTHOR]

• Multiple physiological parameters analysis during Head-Up Tilt Test (HUTT). • Machine learning to achieve early prediction of HUTT outcome. • Genetic algorithm for feature selection and a derived index for results interpretation. Head-up Tilt Testing (HUTT) is a widely used medical tool for the diagnosis of unexplained syncope. Current HUTT protocols, however, are time-consuming. This study aims to investigate the feasibility of using hemodynamic monitoring and machine learning techniques to achieve early prediction of HUTT outcome for syncope patients. A total of 209 subjects participated in this study from June 2016 to November 2019, and hemodynamic signals were collected via a Finometer device (Finapres Medical Systems BV, The Netherlands). We extracted features with a total dimension of 4,313 from the early 18 min (5 min of supine position and 13 min of tilting position). A genetic algorithm (GA) was introduced for feature selection, and an index called the selection ratio (SR) was proposed to further analyze the GA selection result. Four machine learning models were established for this classification task, and their performance results were compared. The maximum tilting duration was shortened from 35 min to 13 min, and a best area under receiver operating characteristic curve of 0.94 via 5-fold cross-validation was obtained by the SVR model, with a sensitivity of 0.86 and a specificity of 0.82. The performance of all algorithms improved after feature selection by GA. The proposed approach is a promising method to shorten the diagnosis time compared to the existing diagnosis process. The GA introduced in this study is an effective feature selection tool to improve model performance. The proposed SR index effectively contributes to the usability and interpretability of the model. [ABSTRACT FROM AUTHOR]

Listeria monocytogenes widely exists in the natural environment and does great harm, which can cause worldwide public safety problem. Infection with L. monocytogenes can cause rapid death of Kupffer cell (KCs) in liver tissue and liver damage. American ginseng saponins is a natural compound in plants, which has great potential in inhibiting L. monocytogenes infection. Therefore, American ginseng stem-leaf saponins (AGS) and American ginseng heat-transformed saponins (HTS) were used as raw materials to study their bacteriostatic experiments in vivo and in vitro. In this experiment, female Kunming mice were randomly divided into five groups: control group, negative group, AGS group, HTS group (10 mg/kg/day in an equal volume via gastric administration) and penicillin group, each group containing six mice. Profiles AGS and HTS components were evaluated by high-performance liquid chromatography (HPLC) analysis. The bacteriostatic effect of AGS and HTS on L. monocytogenes was evaluated by inhibition zone test, minimum inhibitory concentration (MIC) and minimum bactericidal concentration (MBC). The bacteriostatic effect of AGS and HTS pretreatment on mice infected with L. monocytogenes were studies by animal experimental. The results showed that the content of polar saponins in AGS was 0.81 ± 0.003 mg/mg, less polar saponins was 0.08 ± 0.02 mg/mg, the content of polar saponins in HTS was 0.10 ± 0.01 mg/mg, less polar saponins was 0.76 ± 0.02 mg/mg. The in vitro bacteriostatic diameter of HTS (16.6 ± 0.8 mm) is large than that of AGS (10.2 ± 1.2 mm). AGS and HTS pretreatment could reduce the colony numbers in the livers of mice infected with Listeria monocytogenes. The levels of alanine aminotransferase (ALT), IL-1β, IL-6, TNF-α and IFN-γ in the livers of mice in the pretreatment group were significantly lower than those in the negative group. There were obvious leukoplakia, calcification and other liver damage on the liver surface in the negative control group, and obvious inflammatory cell infiltration in HE sections. AGS and HTS pretreatment can reduce liver injury caused by L. monocytogenes and protect the liver. Compared with AGS, HTS has higher content of less polar saponins and better bacteriostatic effect in vitro. The count of bacterial in liver tissue of HTS group was significantly lower, the survival rate was significantly higher than that of AGS group. Less polar saponins had better bacteriostatic effect. Collectively, less polar saponins pretreatment has a protective effect on mice infected with L. monocytogenes , to which alleviated liver damage, improved anti-inflammatory ability and immunity of the body, protected liver may contribute. [Display omitted] • HTS and AGS pretreatment can reduce the mortality of mice infected with Listeria monocytogenes. • The anti-inflammatory and protective effect of HTS is better than that of AGS. • Improve the utilization rate of stems and leaves of American ginseng and reduce the problem of environmental pollution. [ABSTRACT FROM AUTHOR]