EEG-Based Feature Classification Combining 3D-Convolutional Neural Networks with Generative Adversarial Networks for Motor Imagery.

BACKGROUND: The adoption of convolutional neural networks (CNNs) for decoding electroencephalogram (EEG)-based motor imagery (MI) in brain-computer interfaces has significantly increased recently. The effective extraction of motor imagery features is vital due to the variability among individuals and temporal states. METHODS: This study introduces a novel network architecture, 3D-convolutional neural network-generative adversarial network (3D-CNN-GAN), for decoding both within-session and cross-session motor imagery. Initially, EEG signals were extracted over various time intervals using a sliding window technique, capturing temporal, frequency, and phase features to construct a temporal-frequency-phase feature (TFPF) three-dimensional feature map. Generative adversarial networks (GANs) were then employed to synthesize artificial data, which, when combined with the original datasets, expanded the data capacity and enhanced functional connectivity. Moreover, GANs proved capable of learning and amplifying the brain connectivity patterns present in the existing data, generating more distinctive brain network features. A compact, two-layer 3D-CNN model was subsequently developed to efficiently decode these TFPF features. RESULTS: Taking into account session and individual differences in EEG data, tests were conducted on both the public GigaDB dataset and the SHU laboratory dataset. On the GigaDB dataset, our 3D-CNN and 3D-CNN-GAN models achieved two-class within-session motor imagery accuracies of 76.49% and 77.03%, respectively, demonstrating the algorithm's effectiveness and the improvement provided by data augmentation. Furthermore, on the SHU dataset, the 3D-CNN and 3D-CNN-GAN models yielded two-class within-session motor imagery accuracies of 67.64% and 71.63%, and cross-session motor imagery accuracies of 58.06% and 63.04%, respectively. CONCLUSIONS: The 3D-CNN-GAN algorithm significantly enhances the generalizability of EEG-based motor imagery brain-computer interfaces (BCIs). Additionally, this research offers valuable insights into the potential applications of motor imagery BCIs.

How liquids charge the superhydrophobic surfaces.

Liquid-solid contact electrification (CE) is essential to diverse applications. Exploiting its full implementation requires an in-depth understanding and fine-grained control of charge carriers (electrons and/or ions) during CE. Here, we decouple the electrons and ions during liquid-solid CE by designing binary superhydrophobic surfaces that eliminate liquid and ion residues on the surfaces and simultaneously enable us to regulate surface properties, namely work function, to control electron transfers. We find the existence of a linear relationship between the work function of superhydrophobic surfaces and the as-generated charges in liquids, implying that liquid-solid CE arises from electron transfer due to the work function difference between two contacting surfaces. We also rule out the possibility of ion transfer during CE occurring on superhydrophobic surfaces by proving the absence of ions on superhydrophobic surfaces after contact with ion-enriched acidic, alkaline, and salt liquids. Our findings stand in contrast to existing liquid-solid CE studies, and the new insights learned offer the potential to explore more applications.

Assessment of rTMS treatment effects for methamphetamine use disorder based on EEG microstates.

Microstates have been proposed as topographical maps representing large-scale resting-state networks and have recently been suggested as markers for methamphetamine use disorder (MUD). However, it is unknown whether and how they change after repetitive transcranial magnetic stimulation (rTMS) intervention. This study included a comprehensive subject population to investigate the effect of rTMS on MUD microstates. 34 patients with MUD underwent a 4-week randomized, double-blind rTMS intervention (active=17, sham=17). Two resting-state EEG recordings and VAS evaluations were conducted before and after the intervention period. Additionally, 17 healthy individuals were included as baseline controls. The modified k-means clustering method was used to calculate four microstates (MS-A∼MS-D) of EEG, and the FC network was also analyzed. The differences in microstate indicators between groups and within groups were compared. The durations of MS-A and MS-B microstates in patients with MUD were significantly lower than that in HC but showed significant improvements after rTMS intervention. Changes in microstate indicators were found to be significantly correlated with changes in craving level. Furthermore, selective modulation of the resting-state network by rTMS was observed in the FC network. The findings indicate that changes in microstates in patients with MUD are associated with craving level improvement following rTMS, suggesting they may serve as valuable evaluation markers.

Superhydrophobic Designs for Durable Radiative Cooling.

Passive radiative cooling technology is an eco-friendly and energy-free alternative to conventional cooling systems. However, a major challenge in implementing radiative cooling in an outdoor environment is the presence of contamination, which significantly degrades the cooling effectiveness. In response to this challenge, researchers have explored superhydrophobic radiative coolers with self-cleaning abilities as a potential solution. In this Perspective, we summarize the latest progress and highlight certain design principles and strategies for integrating superhydrophobicity into radiative cooling structures. These strategies can be classified into three distinct categories: spraying particles, constructing pores, and creating patterns. Finally, we identify future challenges and opportunities in superhydrophobic radiative coolers, intending to push the technology toward practical applications.

Effects of Background Music on Mental Fatigue in Steady-State Visually Evoked Potential-Based BCIs.

As a widely used brain-computer interface (BCI) paradigm, steady-state visually evoked potential (SSVEP)-based BCIs have the advantages of high information transfer rates, high tolerance for artifacts, and robust performance across diverse users. However, the incidence of mental fatigue from prolonged, repetitive stimulation is a critical issue for SSVEP-based BCIs. Music is often used as a convenient, non-invasive means of relieving mental fatigue. This study investigates the compensatory effect of music on mental fatigue through the introduction of different modes of background music in long-duration, SSVEP-BCI tasks. Changes in electroencephalography power index, SSVEP amplitude, and signal-to-noise ratio were used to assess participants' mental fatigue. The study's results show that the introduction of exciting background music to the SSVEP-BCI task was effective in relieving participants' mental fatigue. In addition, for continuous SSVEP-BCI tasks, a combination of musical modes that used soothing background music during the rest interval phase proved more effective in reducing users' mental fatigue. This suggests that background music can provide a practical solution for long-duration SSVEP-based BCI implementation.

Ultrathin, soft, radiative cooling interfaces for advanced thermal management in skin electronics.

Thermal management plays a notable role in electronics, especially for the emerging wearable and skin electronics, as the level of integration, multifunction, and miniaturization of such electronics is determined by thermal management. Here, we report a generic thermal management strategy by using an ultrathin, soft, radiative-cooling interface (USRI), which allows cooling down the temperature in skin electronics through both radiative and nonradiative heat transfer, achieving temperature reduction greater than 56°C. The light and intrinsically flexible nature of the USRI enables its use as a conformable sealing layer and hence can be readily integrated with skin electronics. Demonstrations include passive cooling down of Joule heat for flexible circuits, improving working efficiency for epidermal electronics, and stabling performance outputs for skin-interfaced wireless photoplethysmography sensors. These results offer an alternative pathway toward achieving effective thermal management in advanced skin-interfaced electronics for multifunctionally and wirelessly operated health care monitoring.

An online hybrid BCI combining SSVEP and EOG-based eye movements.

Hybrid brain-computer interface (hBCI) refers to a system composed of a single-modality BCI and another system. In this paper, we propose an online hybrid BCI combining steady-state visual evoked potential (SSVEP) and eye movements to improve the performance of BCI systems. Twenty buttons corresponding to 20 characters are evenly distributed in the five regions of the GUI and flash at the same time to arouse SSVEP. At the end of the flash, the buttons in the four regions move in different directions, and the subject continues to stare at the target with eyes to generate the corresponding eye movements. The CCA method and FBCCA method were used to detect SSVEP, and the electrooculography (EOG) waveform was used to detect eye movements. Based on the EOG features, this paper proposes a decision-making method based on SSVEP and EOG, which can further improve the performance of the hybrid BCI system. Ten healthy students took part in our experiment, and the average accuracy and information transfer rate of the system were 94.75% and 108.63 bits/min, respectively.

Triboelectric wetting for continuous droplet transport.

Manipulating liquid is of great significance in fields from life sciences to industrial applications. Owing to its advantages in manipulating liquids with high precision and flexibility, electrowetting on dielectric (EWOD) has been widely used in various applications. Despite this, its efficient operation generally needs electrode arrays and sophisticated circuit control. Here, we develop a largely unexplored triboelectric wetting (TEW) phenomenon that can directly exploit the triboelectric charges to achieve the programmed and precise water droplet control. This key feature lies in the rational design of a chemical molecular layer that can generate and store triboelectric charges through agile triboelectrification. The TEW eliminates the requirement of the electric circuit design and additional source input and allows for manipulating liquids of various compositions, volumes, and arrays on various substrates in a controllable manner. This previously unexplored wetting mechanism and control strategy will find diverse applications ranging from controllable chemical reactions to surface defogging.

A large EEG dataset for studying cross-session variability in motor imagery brain-computer interface.

In building a practical and robust brain-computer interface (BCI), the classification of motor imagery (MI) from electroencephalography (EEG) across multiple days is a long-standing challenge due to the large variability of the EEG signals. We collected a large dataset of MI from 5 different days with 25 subjects, the first open-access dataset to address BCI issues across 5 different days with a large number of subjects. The dataset includes 5 session data from 5 different days (2-3 days apart) for each subject. Each session contains 100 trials of left-hand and right-hand MI. In this report, we provide the benchmarking classification accuracy for three conditions, namely, within-session classification (WS), cross-session classification (CS), and cross-session adaptation (CSA), with subject-specific models. WS achieves an average classification accuracy of up to 68.8%, while CS degrades the accuracy to 53.7% due to the cross-session variability. However, by adaptation, CSA improves the accuracy to 78.9%. We anticipate this new dataset will significantly push further progress in MI BCI research in addressing the cross-session and cross-subject challenge.

Achieving efficient power generation by designing bioinspired and multi-layered interfacial evaporator.

Water evaporation is a natural phase change phenomenon occurring any time and everywhere. Enormous efforts have been made to harvest energy from this ubiquitous process by leveraging on the interaction between water and materials with tailored structural, chemical and thermal properties. Here, we develop a multi-layered interfacial evaporation-driven nanogenerator (IENG) that further amplifies the interaction by introducing additional bionic light-trapping structure for efficient light to heat and electric generation on the top and middle of the device. Notable, we also rationally design the bottom layer for sufficient water transport and storage. We demonstrate the IENG performs a spectacular continuous power output as high as 11.8 µW cm-2 under optimal conditions, more than 6.8 times higher than the currently reported average value. We hope this work can provide a new bionic strategy using multiple natural energy sources for effective power generation.

The Paradigm Design of a Novel 2-class Unilateral Upper Limb Motor Imagery Tasks and its EEG Signal Classification.

Multitasking motor imagery (MI) of the unilateral upper limb is potentially more valuable in stroke rehabilitation than the current conventional MI in both hands. In this paper, a novel experimental paradigm was designed to imagine two motions of unilateral upper limb, which is hand gripping and releasing, and elbow reciprocating left and right. During this experiment, the electroencephalogram (EEG) signals were collected from 10 subjects. The time and frequency domains of the EEG signals were analyzed and visualized, indicating the presence of different Event-Related Desynchronization (ERD) or Event-Related Synchronization (ERS) for the two tasks. Then the two tasks were classified through three different EEG decoding methods, in which the optimized convolutional neural network (CNN) based on FBCNet achieved an average accuracy of 67.8%, obtaining a good recognition result. This work not only can advance the studies of MI decoding of unilateral upper limb, but also can provide a basis for better upper limb stroke rehabilitation in MI-BCI.

Prefrontal fNIRS-based clinical data analysis of brain functions in individuals abusing different types of drugs.

BACKGROUND: The activation degree of the orbitofrontal cortex (OFC) functional area in drug abusers is directly related to the craving for drugs and the tolerance to punishment. Currently, among the clinical research on drug rehabilitation, there has been little analysis of the OFC activation in individuals abusing different types of drugs, including heroin, methamphetamine, and mixed drugs. Therefore, it becomes urgently necessary to clinically investigate the abuse of different drugs, so as to explore the effects of different types of drugs on the human brain. METHODS: Based on prefrontal high-density functional near-infrared spectroscopy (fNIRS), this research designs an experiment that includes resting and drug addiction induction. Hemoglobin concentrations of 30 drug users (10 on methamphetamine, 10 on heroin, and 10 on mixed drugs) were collected using fNIRS and analyzed by combining algorithm and statistics. RESULTS: Linear discriminant analysis (LDA), Support vector machine (SVM) and Machine-learning algorithm was implemented to classify different drug abusers. Oxygenated hemoglobin (HbO2) activations in the OFC of different drug abusers were statistically analyzed, and the differences were confirmed. Innovative findings: in both the Right-OFC and Left-OFC areas, methamphetamine abusers had the highest degree of OFC activation, followed by those abusing mixed drugs, and heroin abusers had the lowest. The same result was obtained when OFC activation was investigated without distinguishing the left and right hemispheres. CONCLUSIONS: The findings confirmed the significant differences among different drug abusers and the patterns of OFC activations, providing a theoretical basis for personalized clinical treatment of drug rehabilitation in the future.

Pressure-Sensitive Adhesive with Enhanced and Phototunable Underwater Adhesion.

Pressure-sensitive adhesives (PSAs) are extensively used in diverse applications such as semiconductor manufacturing, labeling, and healthcare because of their quick and viscoelasticity-driven physical adhesion to dry surfaces. However, most of the existing PSAs normally fail to maintain or even establish adhesion under harsh conditions, particularly underwater, due to the lack of robust chemical functionalities for chemistry-based adhesion. Meanwhile, these PSAs are incapable of altering the adhesion in response to external stimuli, limiting their employment in applications requiring dynamic adhesion. Here, we develop a chemically functionalized PSA (f-PSA) with enhanced and phototunable underwater adhesion by incorporating an underwater adhesion enhancer (i.e., mussel-inspired catechol) and a photoresponsive functionality (i.e., anthracene) into a standard acrylic PSA matrix. The synergistic coupling of viscoelasticity-driven physical adhesion originating from the matrix with catechol-enabled chemical adhesion secures sufficient interfacial molecular interactions and leads to enhanced underwater adhesion. The efficient dimerization of anthracene via light-triggered cycloaddition facilely mediates the viscoelastic property of f-PSA, rendering the phototunable adhesion. We envision that this f-PSA can open up more opportunities for applications such as underwater manipulation, transfer printing, and medical adhesives.

Classification accuracy and functional difference prediction in different brain regions of drug abuser prefrontal lobe basing on machine-learning.

Taking different types of addictive drugs such as methamphetamine, heroin, and mixed drugs causes brain functional Changes. Based on the prefrontal functional near-infrared spectroscopy, this study was designed with an experimental paradigm that included the induction of resting and drug addiction cravings. Hemoglobin concentrations of 30 drug users (10 on methamphetamine, 10 on heroin, and 10 on mixed type) were collected. For these three types of individuals, the convolutional neural networks (CNN) was designed to classify eight Brodmann areas and the entire prefrontal area, and the average accuracy of the three classifications on each functional area was obtained. As a result, the classification accuracy was lower on the left side than on the right in the dorsolateral prefrontal cortex (DLPFC) of the drug users, while it was higher on the left than on the right in the ventrolateral prefrontal cortex (VLPFC), frontopolar prefrontal cortex (FPC), and orbitofrontal cortex (OFC). Then the differences in eight functional areas between the three types of individuals were statistically analyzed, and results showed significant differences in the right VLPFC and right OFC.

Application of bi-modal signal in the classification and recognition of drug addiction degree based on machine learning.

Most studies on drug addiction degree are made based on statistical scales, addicts' account, and subjective judgement of rehabilitation doctors. No objective, quantified evaluation has been made. This paper uses devises the synchronous bimodal signal collection and experimentation paradigm with electroencephalogram (EEG) and forehead high-density near-infrared spectroscopy (NIRS) device. The drug addicts are classified into mild, moderate and severe groups with reference to the suggestions of researchers and medical experts. Data of 45 drug addicts (mild: 15; moderate: 15; and severe: 15) is collected, and then used to design an addiction degree testing algorithm based on decision fusion. The algorithm is used to classify mild, moderate and severe addiction. This paper pioneers to use two types of Convolutional Neural Network (CNN) to abstract the EEG and NIR data of drug addicts, and introduces batch normalization to CNN, thus accelerating training process, reducing parameter sensitivity, and enhancing system robustness. The characteristics output by two CNNs are transformed into dimensions. Two new characteristics are assigned with a weight of 50% each. The data is used for decision fusion. In the networks, 27 subjects are used as training sets, 9 as validation sets, and 9 as testing sets. The 3-class accuracy remains to be 63.15%, preliminarily justifying this method as an effective approach to measure drug addiction degree. And the method is ready to use, objective, and offers results in real time.

Hierarchically Hollow Microfibers as a Scalable and Effective Thermal Insulating Cooler for Buildings.

Daytime passive radiative cooling is a promising electricity-free pathway for cooling terrestrial buildings. Current research interest in this cooling strategy mainly lies in tailoring the optical spectra of materials for strong thermal emission and high solar reflection. However, environmental heat gain poses a crucial challenge to building cooling at subambient temperatures. Herein, we devise a scalable thermal insulating cooler (TIC) consisting of hierarchically hollow microfibers as the building envelope that simultaneously achieves passive daytime radiative cooling and thermal insulation to reduce environmental heat gain. The TIC demonstrates efficient solar reflection (94%) and long-wave infrared emission (94%), yielding a temperature drop of about 9 °C under sunlight of 900 W/m2. Notably, the thermal conductivity of the TIC is lower than that of air, thus preventing heat flow from external environments to indoor space in the summer, an additional benefit that does not sacrifice the radiative cooling performance. A building energy simulation shows that 48.5% of cooling energy could be saved if the TIC is widely deployed in China.

An Automatic Glioma Segmentation System Using a Multilevel Attention Pyramid Scene Parsing Network.

BACKGROUND: Due to the significant variances in their shape and size, it is a challenging task to automatically segment gliomas. To improve the performance of glioma segmentation tasks, this paper proposed a multilevel attention pyramid scene parsing network (MLAPSPNet) that aggregates the multiscale context and multilevel features. METHODS: First, T1 pre-contrast, T2-weighted fluid-attenuated inversion recovery (FLAIR) and T1 post-contrast sequences of each slice are combined to form the input. Afterwards, image normalization and augmentation techniques are applied to accelerate the training process and avoid overfitting, respectively. Furthermore, the proposed MLAPSPNet that introduces multilevel pyramid pooling modules (PPMs) and attention gates is constructed. Eventually, the proposed network is compared with some existing networks. RESULTS: The dice similarity coefficient (DSC), sensitivity and Jaccard score of the proposed system can reach 0.885, 0.933 and 0.8, respectively. The introduction of multilevel pyramid pooling modules and attention gates can improve the DSC by 0.029 and 0.022, respectively. Moreover, compared with Res-UNet, Dense-UNet, residual channel attention UNet (RCA-UNet), DeepLab V3+ and UNet++, the DSC is improved by 0.032, 0.026, 0.014, 0.041 and 0.011, respectively. CONCLUSION: The proposed multilevel attention pyramid scene parsing network can achieve stateof- the-art performance, and the introduction of multilevel pyramid pooling modules and attention gates can improve the performance of glioma segmentation tasks.

Single-Beat Measurement of Left Ventricular Contractility in Normothermic Ex Situ Perfused Porcine Hearts.

OBJECTIVE: For heart transplantation, donor heart status needs to be evaluated during normothermic ex situ perfusion (ESHP). Left ventricular end-systolic elastance (E es) measures the left ventricular contractile function, but its estimation requires the occlusion of the left atrium line in the ESHP, which may cause unnecessary damage to the donor heart. We present a novel method to quantify E es based on hemodynamic parameters obtained from only one steady-state PV loop in ESHP. METHODS: E es was obtained by the end-systolic point (P es, V es) and the volume axis intercept point of E es (V 0). V 0 was estimated through the support vector machine regression (SVR) method using parameters derived from the measured steady-state PV loop. To achieve high V 0 estimation accuracy, a filter-based support vector machine recursive feature elimination method (SVM-RFE) algorithm selected the parameters for V 0 estimation. Hemodynamic parameter samples (n = 101) obtained from ESHP experiments with pig's hearts were used to train the E es calculation model. Early post-transplantation outcomes in six heart transplantation experiments were then estimated from the trained E es calculation model. RESULTS: E es calculated by the proposed method agreed well with conventional multi-beat estimates obtained by the occlusion process (r = 0.88, p < 0.001, n = 101) and was capable of predicting the early post-transplant cardiac index (r = 0.84, p < 0.05, n = 6). CONCLUSION: This method effectively assesses left ventricular contractility during ESHP and predicts early post-transplant outcomes in the porcine model. SIGNIFICANCE: Our approach is the first to quantify E es by estimating V 0 from steady-state beats in ESHP for accurately predicting early post-transplantation outcomes.

Bioinspired Soot-Deposited Janus Fabrics for Sustainable Solar Steam Generation with Salt-Rejection.

Inspired by lotus leaves, self-floating Janus cotton fabric is successfully fabricated for solar steam generation with salt-rejecting property. The layer-selective soot-deposited fabrics not only act as a solar absorber but also provide the required superhydrophobicity for floating on the water. With a polyester protector, the prepared Janus evaporator exhibits a sustainable evaporation rate of 1.375 kW m-2 h-1 and an efficiency of 86.3% under 1 sun (1 kW m-2) and also performs well under low intensity and inclined radiation. Furthermore, no special apparatus and/or tedious processes are needed for preparing this device. With a cost of less than $1 per m2, this flexible Janus absorber is a promising tool for portable solar vapor generator.

Analyte-resolved magnetoplasmonic nanocomposite to enhance SPR signals and dual recognition strategy for detection of BNP in serum samples.

B-type natriuretic peptide (BNP) is a short peptide that is considered to be an important heart failure (HF)-related biomarker. Due to its low concentration in the blood and short half-life, the sensitive detection of BNP is a bottleneck for diagnosing patients at early stages of HF. In this paper, we report a facile surface plasmon resonance (SPR) sensor to measure BNP; the sensor is based on aptamer-functionalized Au nanoparticles (GNPs-Apt) and antibody-modified magnetoplasmonic nanoparticles (MNPs-Ab) to enable dual screening of BNP in complex environments. During sensing, BNP forms MNP-Ab/BNP/GNP-Apt nanoconjugates that can be rapidly separated from the complex sample by a magnet to avoid degradation within the analyte's half-life. The developed SPR biosensor shows high selectivity, a wide dynamic response range of BNP concentrations from 100 fg/mL to 10 ng/mL, and a low detection limit of 28.2 fg/mL (S/N = 3). Using the proposed sensor, BNP was successfully detected in clinical samples. Thus, the designed SPR biosensor provides a novel and sensitive sensing platform for BNP detection with potential applications in clinical practice.