Interface Applications Research Articles

VME-EFD : A novel framework to eliminate the Electrooculogram artifact from single-channel EEGs.

The diagnosis of neurological disorders often involves analyzing EEG data, which can be contaminated by artifacts from eye movements or blinking (EOG). To improve the accuracy of EEG-based analysis, we propose a novel framework, VME-EFD, which combines Variational Mode Extraction (VME) and Empirical Fourier Decomposition (EFD) for effective EOG artifact removal. In this approach, the EEG signal is first decomposed by VME into two segments: the desired EEG signal and the EOG artifact. The EOG component is further processed by EFD, where decomposition levels are analyzed based on energy and skewness. The level with the highest energy and skewness, corresponding to the artifact, is discarded, while the remaining levels are reintegrated with the desired EEG.
Simulations on both synthetic and real EEG datasets demonstrate that VME-EFD outperforms existing methods, with lower RRMSE (0.1358 vs. 0.1557, 0.1823, 0.2079, 0.2748), lower ΔPSD in the α band (0.10±0.01 and 0.17±0.04 vs. 0.89±0.91 and 0.22±0.19, 1.32±0.23 and 1.10±0.07, 2.86±1.30 and 1.19±0.07, 3.96±0.56 and 2.42±2.48), and higher correlation coefficient (CC: 0.9732 vs. 0.9695, 0.9514, 0.8994, 0.8730). The framework effectively removes EOG artifacts and preserves critical EEG features, particularly in the α band, making it highly suitable for brain-computer interface (BCI) applications.&#xD.

An Extreme Environment Capable Self‐Healing Single Active Layered Triboelectric Sensors as Fully Recyclable and Transparent Human‐Machine Interfaces

AbstractCurrent approaches to triboelectric devices require multiple layers. Here, a transparent ultra‐stretchable single active layered ionic‐based nanogenerator is shown that can self‐heal in both dry and wet environments. Despite being conductive, surprisingly the ionogel exhibits triboelectric properties and generates output voltages ranging from 1.7 to 70 V. By tuning the ionic liquid concentration, unique triboelectric devices with position‐detection and energy harvesting ability are developed. This design enables visualization of the mechanical force distribution through the mechano‐electric‐optical conversion process, which is achieved by changing it from an insulating dielectric material to a tribo‐negative ionic conducting material. The triboelectric device maintains its stability and self‐healing ability at extreme environments (−20 to 380 °C, pH 1.3 to 13), and the device can be fully closed‐loop recycled for reuse. Various human‐machine interface applications such as untethered self‐powered light‐emitting jellyfish‐like devices are demonstrated, a mechano‐thermal sensor array, and tactile recognition of object shapes with over 91% accuracy using convolution neural networks (CNN).

Brain-Computer Interface Application in Recognition and Regulation of Emotion

This paper investigates the application of Brain-Computer Interface (BCI) technology in the realms of emotion recognition and regulation. BCI technology facilitates direct communication between the brain and external devices, offering significant promise for improving human-environment interactions, particularly with regard to the identification and modulation of emotional states. By analyzing brain signals, such as electroencephalography, this study classifies emotions based on widely recognized models, including Ekmans model and the Russell circumplex model. To enhance the precision of emotion classification, machine learning algorithms, such as support vector machines and neural networks, are utilized. Moreover, this study explores BCIs potential in emotion regulation, focusing on neurofeedback and brain stimulation methods like transcranial direct current stimulation, which have shown therapeutic potential, particularly for disorders related to emotional dysregulation. Additionally, the paper delves into the integration of BCI with virtual reality to create immersive environments conducive to emotional therapy. Despite its considerable potential, BCI technology faces obstacles such as low data transmission rates and the complexities associated with user training. Nonetheless, the integration of BCI technology within Industry 4.0 frameworks holds promising opportunities for optimizing human-machine interactions and improving workplace safety.

Eeg based smart emotion recognition using meta heuristic optimization and hybrid deep learning techniques

In the domain of passive brain-computer interface applications, the identification of emotions is both essential and formidable. Significant research has recently been undertaken on emotion identification with electroencephalogram (EEG) data. The aim of this project is to develop a system that can analyse an individual’s EEG and differentiate among positive, neutral, and negative emotional states. The suggested methodology use Independent Component Analysis (ICA) to remove artefacts from Electromyogram (EMG) and Electrooculogram (EOG) in EEG channel recordings. Filtering techniques are employed to improve the quality of EEG data by segmenting it into alpha, beta, gamma, and theta frequency bands. Feature extraction is performed with a hybrid meta-heuristic optimisation technique, such as ABC-GWO. The Hybrid Artificial Bee Colony and Grey Wolf Optimiser are employed to extract optimised features from the selected dataset. Finally, comprehensive evaluations are conducted utilising DEAP and SEED, two publically accessible datasets. The CNN model attains an accuracy of approximately 97% on the SEED dataset and 98% on the DEAP dataset. The hybrid CNN-ABC-GWO model achieves an accuracy of approximately 99% on both datasets, with ABC-GWO employed for hyperparameter tuning and classification. The proposed model demonstrates an accuracy of around 99% on the SEED dataset and 100% on the DEAP dataset. The experimental findings are contrasted utilising a singular technique, a widely employed hybrid learning method, or the cutting-edge method; the proposed method enhances recognition performance.

Synthesis and characterization of paraffin microcapsules with magnetized graphene oxide core/shell interface for thermal management applications

Synthesis and characterization of paraffin microcapsules with magnetized graphene oxide core/shell interface for thermal management applications

The Implementation of Gamification Methods in the Edusting Application as an Educational Medium to Enhance Mothers Knowledge of Stunting

Interventions for stunting are an urgent public health priority. Because mothers' limited knowledge about stunting has a direct impact on stunting prevention and management. One of the extraordinary outreach programs is the stunting socialization program for mothers of toddlers carried out at the integrated health post. However, the socialization action is considered too limited. Gamification is one potential approach to increase the effectiveness of these efforts by introducing game design and game elements into educational content in three areas: childcare practices, nutritional intake, and stunting prevention. This is expected to explore mothers' involvement and motivation to prevent stunting. Mothers certainly prefer a more attractive interactive interface and gamification applications will stimulate cognitive and emotional involvement so that mothers feel more comfortable in the learning process about stunting management. In this study, we present the design of an education that aims to help mothers understand and overcome stunting. This application is expected to make mothers better understand parenting patterns, nutrition, and stunting management so that it can help mothers' learning to increase the speed of mothers' understanding of stunting prevention.

TMSA-Net:A novel attention mechanism for improved motor imagery EEG signal processing

Electroencephalography (EEG) is a non-invasive method used to record the brain’s electrical activity, widely employed in brain-computer interface (BCI) applications for decoding motor imagery (MI) signals. Traditional models combining Convolutional Neural Networks (CNNs) and Transformers for decoding Motor Imagery Electroencephalography (MI-EEG) signals often struggle to capture the crucial interrelationships between local and global features effectively, resulting in suboptimal performance. To address this issue, we propose a novel computationally efficient model that integrates CNNs with a novel attention mechanism within the Transformer architecture. This approach effectively captures both local and global dependencies, thereby enhancing feature extraction and decoding accuracy. Evaluations against state-of-the-art models (EEGNet, Deep ConvNet, IFNet, Conformer) on three public datasets (BCIC-IV-2a, BCIC-IV-2b, HGD) reveal substantial performance enhancements. Specifically, our model achieves improvements in accuracy of 9.78%, 11.05%, 4.81%, and 3.75% over these models on the BCIC-IV-2a dataset. Ablation studies and visualization techniques (t-SNE, Grad-CAM) further corroborate the effectiveness and interpretability of our model, highlighting its superior decoding capabilities. The code is available at https://github.com/Whit3Zhao/TMSA-Net.git.

Recent Advances in Nanomaterial-Based Biosignal Sensors.

Recent research for medical fields, robotics, and wearable electronics aims to utilize biosignal sensors to gather bio-originated information and generate new values such as evaluating user well-being, predicting behavioral patterns, and supporting disease diagnosis and prevention. Notably, most biosignal sensors are designed for body placement to directly acquire signals, and the incorporation of nanomaterials such as metal-based nanoparticles or nanowires, carbon-based or polymer-based nanomaterials-offering stretchability, high surface-to-volume ratio, and tunability for various properties-enhances their adaptability for such applications. This review categorizes nanomaterial-based biosignal sensors into three types and analyzes them: 1) biophysical sensors that detect deformation such as folding, stretching, and even pulse, 2) bioelectric sensors that capture electric signal originating from human body such as heart and nerves, and 3) biochemical sensors that catch signals from bio-originated fluids such as sweat, saliva and blood. Then, limitations and improvements to nanomaterial-based biosignal sensors is depicted. Lastly, it is highlighted on deep learning-based signal processing and human-machine interface applications, which can enhance the potential of biosignal sensors. Through thispaper, it is aim to provide an understanding of nanomaterial-based biosignal sensors, outline the current state of the technology, discuss the challenges that be addressed, and suggest directions for development.

Web Application for Diabetes Prediction using Machine Learning Techniques

The objective of this project is to predict a person's risk of having diabetes by utilizing Support Vector Machine (SVM) algorithms in an intuitive web application interface. This application attempts to provide accurate and reasonable predictions by using input health parameters (number of pregnancies, blood pressure, glucose level, insulin level, age, skin thickness, diabetes pedigree function, etc.) that users provide via a graphical user interface (GUI). By combining the power of SVM with user-friendly web technology, the project endeavors to enhance accessibility to predictive healthcare tools. The seamless integration of Machine Learning into a web application facilitates a simple and effective method for diabetes prediction, which could aid people in making accurate choices regarding their health. By promoting preventive measures and giving people early awareness, this initiative hopes to support proactive healthcare.

Bioinspired Magnetized String with Tension-Dependent Eigenfrequencies for Wearable Human-Machine Interactions.

Flexible and wearable devices have exhibited potential for applications in the fields of human-machine interactions (HMIs) and Internet of Things. However, challenges remain in the improvement of the communication storage capacity with a simplified architecture. Inspired by tension regulation in natural tendons, a single-channel wearable HMI strategy is proposed using the eigenfrequency of magnetized strings as a sensing solution. Based on electromagnetic induction, mechanical vibration of the magnetized string can electrically induce periodical damping signals in the coil that are associated with the intrinsic eigenfrequency property of the string. Using a theoretical vibration model, nonoverlapping eigenfrequencies are precisely customized by designing the dimension/modulus or tension status of the string to broaden the eigenfrequency library. By integrating strings with different eigenfrequencies, multiple commands can be realized with a single communication channel. Moreover, identifiable commands can be flexibly tuned with only one magnetized string by customizing the tensile length (string tension) for eigenfrequency regulation. Demonstrations such as tactile addressing, authentication systems, and robotic control indicate the potential of the interface for multifunctional HMI applications. We expect that this strategy will provide a valuable reference for the future design of wearable HMI interfaces with high storage capacity and controllability in an accessible architecture.

Deep Learning for Weed Detection and Segmentation in Agricultural Crops Using Images Captured by an Unmanned Aerial Vehicle

Artificial Intelligence (AI) has changed how processes are developed, and decisions are made in the agricultural area replacing manual and repetitive processes with automated and more efficient ones. This study presents the application of deep learning techniques to detect and segment weeds in agricultural crops by applying models with different architectures in the analysis of images captured by an Unmanned Aerial Vehicle (UAV). This study contributes to the computer vision field by comparing the performance of the You Only Look Once (YOLOv8n, YOLOv8s, YOLOv8m, and YOLOv8l), Mask R-CNN (with framework Detectron2), and U-Net models, making public the dataset with aerial images of soybeans and beans. The models were trained using a dataset consisting of 3021 images, randomly divided into test, validation, and training sets, which were annotated, resized, and increased using the Roboflow application interface. Evaluation metrics were used, which included training efficiency (mAP50 and mAP50-90), precision, accuracy, and recall in the model’s evaluation and comparison. The YOLOv8s variant achieved higher performance with an mAP50 of 97%, precision of 99.7%, and recall of 99% when compared to the other models. The data from this manuscript show that deep learning models can generate efficient results for automatic weed detection when trained with a well-labeled and large set. Furthermore, this study demonstrated the great potential of using advanced object segmentation algorithms in detecting weeds in soybean and bean crops.

Deployment of intelligent irrigation monitoring system with Android app for machine learning prediction.

Water is a fundamental necessity for humans and a critical resource in agriculture. However, water scarcity poses a significant challenge, especially considering that agriculture accounts for a substantial portion of freshwater usage. The inadequate monitoring resources in agriculture lead to unnecessary wastage of water, affecting crop growth and the water supply. This challenge has spurred us to focus on the agricultural domain, aiming to conserve water by imparting intelligence into irrigation systems by converging IoT and machine learning technologies. Our study has proposed a Smart Irrigation System (SIS), designed to operate remotely, leveraging open-source IoT platforms. Utilizing IoT and machine learning methodologies can effectively optimize water usage in irrigation practices. The designed system comprises hardware and software tools, where comprehensive data is monitored through an application interface. The sensor's interface with a 32-bit microcontroller is used to gather data such as environmental temperature, humidity, and soil moisture and temperature. The user (farmer) continuously receives real-time updates about the condition of the farmland through the Blynk app. The app triggers a notification advising the user to start or stop watering based on pre-set threshold values of soil moisture. Additionally, users can control the water pump remotely through the app. The data collected from the deployed smart irrigation system was used to train a machine learning algorithm incorporating weather parameters and soil temperature to predict soil moisture content. This trained model aims to offer guidance for future assessments of unknown soil samples based on the weather conditions.

Wearable EEG Neurofeedback Based-on Machine Learning Algorithms for Children with Autism: A Randomized, Placebo-controlled Study.

Behavioral interventions have been shown to ameliorate the electroencephalogram (EEG) dynamics underlying the behavioral symptoms of autism spectrum disorder (ASD), while studies have also demonstrated that mirror neuron mu rhythm-based EEG neurofeedback training improves the behavioral functioning of individuals with ASD. This study aimed to test the effects of a wearable mu rhythm neurofeedback training system based on machine learning algorithms for children with autism. A randomized, placebo-controlled study was carried out on 60 participants aged 3 to 6 years who were diagnosed with autism, at two center-based intervention sites. The neurofeedback group received active mu rhythm neurofeedback training, while the control group received a sham neurofeedback training. Other behavioral intervention programs were similar between the two groups. After 60 sessions of treatment, both groups showed significant improvements in several domains including language, social and problem behavior. The neurofeedback group showed significantly greater improvements in expressive language (P=0.013) and cognitive awareness (including joint attention, P=0.003) than did the placebo-controlled group. Artificial intelligence-powered wearable EEG neurofeedback, as a type of brain-computer interface application, is a promising assistive technology that can provide targeted intervention for the core brain mechanisms underlying ASD symptoms.

Wear-free sliding electrical contacts with ultralow electrical resistivity

Sliding electrical contacts are commonly applied in electrical connectors, such as conductive slip rings, pantographs, switches, and commutators. However, they suffer from several unavoidable problems caused by friction and wear, including high energy consumption, intermittent failures, limited life, and even failure disasters. In this study, we realized an ultralow-friction and long-distance wear-free state, defined as structural superlubricity (SSL), between sliding electrical interfaces under ambient conditions. A conductive SSL can be implemented in experiments with single-crystal graphite flakes on flattened metals, such as Au and Ni films. Furthermore, we found that depositing a 2 to 3-nm-thick diamond-like carbon (DLC) film on a nickel alloy can lead to an even lower resistivity than that of metals alone. In addition, we revealed the mechanism by which DLC films can improve the conductivity between graphite and metals through density functional-theory simulations. In addition, we prepared a prototype of the SSL slip ring and proved that it possessed ultralow friction, was wear-free, and had no intermittent failures. Consequently, our results demonstrate a unique type of electrical-contact interface for applications requiring conduction while sliding. Thus, we opened the door for SSL electromechanical coupling.

Intelligent Control System for the Hard X-Ray Nanoprobe Beamline Beam Optimization Based on Automatic Evolution Algorithm and Expert System.

A synchrotron radiation beamline automatic optimization system has been used in the Shanghai Synchrotron Radiation Facility, improving the optimization efficiency, but it does not store and use the beamline adjusting experience, and cannot quickly optimize and store the experienced improvement. The expert system combined with an automatic evolutionary algorithm is used for intelligent beamline optimization; the algorithm initialization is optimized by invoking database experience, the convergence is quickly completed near the optimal solution, and the system's learning is improved by storing experience results. The software was designed on the EPICS (Version 3.15) platform, which was used to implement the algorithm in Python language, the expert database was developed with MongoDB tool (Version 4.0.27), and the upper application interface was designed with CSS software (Phoebus Version 4.7.2). The system was successfully tested on the BL13U hard X-ray nanoprobe beamline of Shanghai Synchrotron Radiation Facility. The results show that the maximum convergence time of a single objective with four-axis degrees of freedom is about 2 min, and the speed is increased by 15 times. The solution set obtained by using multi-objective two and four-axis degrees of freedom is better overall. The system can effectively improve the optimization efficiency and effect, and its universality can be extended to other synchrotron radiation devices and beamlines to promote the development of intelligent beamline modulation technology.

Mobility-as-a-service and travel behaviour change: How multimodal bundles reshape our travel choices

Mobility-as-a-service (MaaS) is an innovative framework aiming to promote sustainable mobility via altering users travel behaviour. Despite its potential, empirical evidence of MaaS effectiveness in achieving sustainable goals is scarce due to limited real-world trials and commercial offers of MaaS. This study leverages high-quality data from the Sydney MaaS trial and the Application of Programming Interface (API) to develop an original Error Components-Random Parameter Logit Model that describe the choices of transport modes under the “MaaS era”. Analysis results offer the first insight into how multimodal bundles reshape mode choices and the extent to which subscription bundles could be used as a powerful tool to nudge users towards more sustainable choices. The results show that multimodal bundles present an appealing alternative for the users and help them reduce their private car use. Simulations were conducted to provide guidance for designing multimodal bundles that are both attractive to users and friendly to the environment.

Brain-computer interface applications in the aging population with Alzheimers and Parkinsons disease

Abstract. The rapidly aging population faces increasing prevalence of neurodegenerative conditions like Alzheimer's Disease (AD) and Parkinson's Disease (PD), which significantly impair cognitive and motor functions. There is an urgent need for increased research, awareness, and development of effective treatments to mitigate their impact on the affected individuals and their families. This paper delves into the history, mechanisms, and applications of brain-computer interfaces (BCIs) as innovative solutions to mitigate these impacts. BCIs enable direct brain-to-device communication, circumventing conventional neuromuscular routes. We highlight how neurofeedback and neuroplasticity facilitated by BCIs not only restore motor skills and cognitive functions but also enhance the overall quality of life for sufferers. Through applications in cognitive training, motor assistance, and advanced communication aids, BCIs present a significant promise despite the ethical and practical challenges they pose. The continual advancement of this technology promises to a future when its incorporation into treatment plans will be widely used and provide significant advantages for those with PD and AD.

A comparative analysis of metaheuristic algorithms for optimizing curved roof structures

This article explores the optimal design of curved steel structures with a focus on minimizing their weight by determining the most suitable cross-sections. Customized optimization algorithms were developed to identify structural designs that meet safety and durability requirements, adhering to design constraints set by the ASD-AISC specification. To ensure reliable results, the study employs a variety of metaheuristic optimization methods: Biogeography-Based Optimization (BBO), Dynamic Harmony Search (DHS), Wolf Colony Algorithm (WCA), and Honey Badger Algorithm (HBA). The Honey Badger Algorithm, a relatively new addition to the field, was used for the first time in the context of structural optimization for curved roof systems. This unique approach aims to evaluate its performance in civil engineering problems and compare it with other established algorithms. The major challenges of this study lie in the inherent complexity of dome structures, which involve a large number of elements and nonlinear constraints. The subdivision of the structure into smaller groups was necessary to manage the computational load, although this introduced additional complexities. Despite these challenges, the metaheuristic methods demonstrated their robustness in addressing such intricate engineering problems. Additionally, data retrieval is facilitated through Open Application Interface (OAPI) functions, enabling seamless data transfer between SAP 2000 and Visual Basic. The study's final design example involves a dome model with 2556 elements, which was both modeled and optimized. The results demonstrate the effectiveness of these optimization algorithms in achieving structurally sound and efficient designs. The insights gained from this study contribute to our understanding of optimized cross-sections in curved steel structures, offering valuable guidance for improving structural performance and minimizing material usage.

Evaluation of consumer-grade wireless EEG systems for brain-computer interface applications.

With the growing popularity of consumer-grade electroencephalogram (EEG) devices for health, entertainment, and cognitive research, assessing their signal quality is essential. In this study, we evaluated four consumer-grade wireless and dry-electrode EEG systems widely used for brain-computer interface (BCI) research and applications, comparing them with a research-grade system. We designed an EEG phantom method that reproduced µV-level amplitude EEG signals and evaluated the five devices based on their spectral responses, temporal patterns of event-related potential (ERP), and spectral patterns of resting-state EEG. We discovered that the consumer-grade devices had limited bandwidth compared with the research-grade device. A late component (e.g., P300) was detectable in the consumer-grade devices, but the overall ERP temporal pattern was distorted. Only one device showed an ERP temporal pattern comparable to that of the research-grade device. On the other hand, we confirmed that the activation of the alpha rhythm was observable in all devices. The results provide valuable insights for researchers and developers when it comes to selecting suitable EEG devices for BCI research and applications.

Deep Autoencoder for Real-Time Single-Channel EEG Cleaning and Its Smartphone Implementation Using TensorFlow Lite With Hardware/Software Acceleration.

To remove signal contamination in electroencephalogram (EEG) traces coming from ocular, motion, and muscular artifacts which degrade signal quality. To do this in real-time, with low computational overhead, on a mobile platform in a channel count independent manner to enable portable Brain-Computer Interface (BCI) applications. We propose a Deep AutoEncoder (DAE) neural network for single-channel EEG artifact removal, and implement it on a smartphone via TensorFlow Lite. Delegate based acceleration is employed to allow real-time, low computational resource operation. Artifact removal performance is quantified by comparing corrupted and ground-truth clean EEG data from public datasets for a range of artifact types. The on-phone computational resources required are also measured when processing pre-saved data. DAE cleaned EEG shows high correlations with ground-truth clean EEG, with average Correlation Coefficients of 0.96, 0.85, 0.70 and 0.79 for clean EEG reconstruction, and EOG, motion, and EMG artifact removal respectively. On-smartphone tests show the model processes a 4 s EEG window within 5 ms, substantially outperforming a comparison FastICA artifact removal algorithm. The proposed DAE model shows effectiveness in single-channel EEG artifact removal. This is the first demonstration of a low-computational resource deep learning model for mobile EEG in smartphones with hardware/software acceleration. This work enables portable BCIs which require low latency real-time artifact removal, and potentially operation with a small number of EEG channels for wearability. It makes use of the artificial intelligence accelerators found in modern smartphones to improve computational performance compared to previous artifact removal approaches.