Article
Engineering, Electrical & Electronic
Praveen Dwivedi, Rohit Singh, Brajendra Singh Sengar, Amitesh Kumar, Vivek Garg
Summary: In this study, a new simulation approach for transient analysis of a single cavity dielectric-modulated p-type tunnel field-effect transistor (TFET) for biosensing applications was investigated. The device performance was examined using a 2D device simulator, with results calibrated against experimental data. The study focused on DC transfer characteristics, transient response of drain current, sensitivity, and selectivity, showing significant improvement in results for biosensing applications.
IEEE SENSORS JOURNAL
(2021)
Article
Computer Science, Artificial Intelligence
Xi Chen, Xiao Shao, Xin Pan, Gaochao Luo, Maoqiang Bi, Tianyan Jiang, Kang Wei
Summary: A method combining variational mode decomposition, multi-scale entropy, and image feature is proposed in this paper to extract feature parameters of partial discharge (PD) signals. Experimental results show that the proposed method has good robustness in processing noisy signals.
CAAI TRANSACTIONS ON INTELLIGENCE TECHNOLOGY
(2022)
Article
Chemistry, Analytical
Wenbo Wu, Yun Pan
Summary: This paper proposes a design method for a modular convolutional neural network model that solves the problems of overfitting and large model parameters in image recognition tasks. By fusing the features extracted from multiple submodules, the model achieves accelerated convergence. An attention-based gate unit is added to optimize the model structure and reduce the floating-point operations per second (FLOPs) dynamically. The proposed model outperforms VggNet, ResNet, and GoogLeNet in terms of simplicity and parameter size, achieving good results in various Kaggle datasets.
Article
Computer Science, Hardware & Architecture
D. Venugopal, V Mohan, S. Ramesh, S. Janupriya, Sangsoon Lim, Seifedine Kadry
Summary: In recent times, pattern recognition of communication modulation signals has gained significant attention in various application areas. The proposed CSM-FFDNN model utilizes fractal features and deep neural networks for communication signal modulation pattern recognition, achieving superior performance compared to existing methods.
COMPUTER SYSTEMS SCIENCE AND ENGINEERING
(2022)
Article
Computer Science, Artificial Intelligence
Bhawna Ahuja, Virendra P. Vishwakarma
Summary: This paper presents a deterministic extreme learning machine for neural network with feedforward architecture formulated with multiple kernel learning, and further enhances the approach by incorporating Gray level co-occurrence matrix (GLCM) for multi-modal feature extraction.
EXPERT SYSTEMS WITH APPLICATIONS
(2021)
Article
Computer Science, Artificial Intelligence
Rami N. Khushaba, Erik Scheme, Ali H. Al-Timemy, Angkoon Phinyomark, Ahmed Al-Taee, Adel Al-Jumaily
Summary: This paper introduces a novel approach using Fusion of Time Domain Descriptors and Range Spatial Filtering for processing EMG signals, achieving superior performance compared to traditional methods and other state-of-the-art models.
EXPERT SYSTEMS WITH APPLICATIONS
(2021)
Article
Computer Science, Artificial Intelligence
Liming Han, Chi Yang, Qing Li, Bin Yao, Zixian Jiao, Qianyang Xie
Summary: This study introduces a coordinate-based regression method called Dynamic Deformable Transformer (DDT) for face alignment, which outperforms existing methods in terms of simplicity and efficiency.
IET COMPUTER VISION
(2023)
Article
Computer Science, Artificial Intelligence
Tao Zhang, Chen-Feng Long, Yang-Jun Deng, Wei-Ye Wang, Si-Qiao Tan, Heng-Chao Li
Summary: This paper proposes a novel low-rank preserving embedding regression (LRPER) method by integrating LRR, linear regression, and projection learning into a unified framework to enhance the discriminability of low-dimensional features extracted by LRR-based subspace learning methods. LRPER utilizes LRR to reveal underlying structure information and employs a robust metric L-2, L-1-norm to measure reconstruction error and regression loss. An alternative iteration algorithm is designed to optimize the proposed model, and the experimental results show that LRPER outperforms some state-of-the-art feature extraction methods.
IET COMPUTER VISION
(2023)
Article
Computer Science, Information Systems
Ebrahim Al-wajih, Rozaida Ghazali
Summary: In this paper, the accuracy of LBP technique and its variations were enhanced using a sliding window approach, and the results showed that in some cases, LBP technique still remains the most effective.
MULTIMEDIA TOOLS AND APPLICATIONS
(2021)
Article
Telecommunications
Shahin Shafei, Hamid Vahdati, Tohid Sedghi, Asghar Charmin
Summary: In this research, a remarkable feature combination based on modulation methods for content-based image retrieval was presented. The features generated from improved contourlet transform and spectral correlation coefficient functions were re-composed as frequency statistic specification, with spatial signals dominion proposed. The proposed features were employed for efficient image retrieval high-level features in a large-scale database of 10,000 images, resulting in a robust feature matrix for database image retrieval. The system achieved a higher retrieval percentage with over 90% accuracy in mentioned database, demonstrating the precision and efficiency of the proposed system.
WIRELESS PERSONAL COMMUNICATIONS
(2022)
Article
Computer Science, Information Systems
Sheela Ramachandra, Suchithra Ramachandran
Summary: This paper proposes a Periocular recognition algorithm that utilizes region-specific and sub-image-based neighbor gradient feature extraction to achieve better recognition results. The proposed method segments the periocular region into sub-regions and extracts features using different algorithms. Experimental results demonstrate that the proposed method outperforms traditional algorithms on multiple datasets.
JOURNAL OF KING SAUD UNIVERSITY-COMPUTER AND INFORMATION SCIENCES
(2022)
Article
Chemistry, Analytical
Rajamanickam Yuvaraj, Prasanth Thagavel, John Thomas, Jack Fogarty, Farhan Ali
Summary: Advances in signal processing and machine learning have accelerated EEG-based emotion recognition research. This study compared the classification accuracy of various sets of EEG features to identify emotional states. By evaluating the performance on five independent datasets, it was found that the FD-CART feature-classification method achieved the highest accuracy for valence and arousal. These findings suggest the reliability of the FD features derived from EEG data for emotion recognition, and may contribute to the development of a real-time EEG-based emotion recognition system.
Article
Multidisciplinary Sciences
Sofien Gannouni, Arwa Aledaily, Kais Belwafi, Hatim Aboalsamh
Summary: Recognizing emotions using biological brain signals requires accurate signal processing and feature extraction methods. This study proposes a novel and adaptive channel selection method, along with the identification of epoch instants during emotions, to enhance the accuracy of the system. Experimental results show that the proposed method outperforms existing studies in multi-class emotion recognition with an average accuracy rate exceeding 89%. The method also shows improvement in accuracy rate when compared to existing algorithms dealing with 9 emotions, by 8%.
SCIENTIFIC REPORTS
(2021)
Article
Computer Science, Information Systems
Erdal Tasci, Aybars Ugur
Summary: With the increasing number of digital images, computer-aided classification of image types is widely used. The feature extraction and selection stages play a crucial role in improving classification performance. In this study, a novel pattern recognition framework combining diverse and large-scale handcrafted feature extraction methods and the selection stage is developed. Genetic algorithms are used for feature selection. Experimental results show high accuracy rates on different datasets, making the proposed method competitive with existing state-of-the-art methods.
MULTIMEDIA TOOLS AND APPLICATIONS
(2022)
Article
Computer Science, Artificial Intelligence
Dongmei Mo, Zhihui Lai, Jie Zhou, Hu Qinghua
Summary: In this paper, a new method called Jointly Sparse Orthogonal Linear Discriminant Analysis (JSOLDA) is proposed to improve the performance of linear discriminant analysis in the field of computer vision and pattern recognition. The proposed method obtains the sparse orthogonal projections for feature extraction through constrained scatter matrix decomposition. Experimental results demonstrate that JSOLDA outperforms several well-known methods based on linear discriminant analysis and L2,1-norm.
PATTERN RECOGNITION
(2023)
Article
Chemistry, Analytical
Christopher Schnur, Payman Goodarzi, Yevgeniya Lugovtsova, Jannis Bulling, Jens Prager, Kilian Tschoeke, Jochen Moll, Andreas Schuetze, Tizian Schneider
Summary: This paper proposes a machine learning-based automated damage detection method that harnesses data-driven analysis for structural health monitoring systems. Experimental results demonstrate that the method is capable of reliably detecting structural damage automatically, and it can identify damaged structures at different damage locations and temperatures even without prior training.
Article
Instruments & Instrumentation
Sebastian Schorr, Dirk Baehre, Andreas Schuetze
Summary: The integration of process data and machine learning methods in manufacturing processes opens up new possibilities for quality control and condition monitoring. Predicting workpiece quality in early machining stages can lead to efficient utilization of resources. However, the implementation of machine learning methods in real manufacturing processes is limited due to the need for validation and proving their capability for quality prediction.
TM-TECHNISCHES MESSEN
(2022)
Article
Instruments & Instrumentation
Payman Goodarzi, Andreas Schuetze, Tizian Schneider
Summary: This paper explores the issue of performance degradation in machine learning methods and data-driven models when facing real-life situations. It compares the prediction results of traditional machine learning and deep neural networks in industrial condition monitoring. The study finds that domain shift can be visualized using feature extraction and principal component analysis, and the cross-domain validated results of FESC/R are comparable to state-of-the-art methods.
TM-TECHNISCHES MESSEN
(2022)
Editorial Material
Instruments & Instrumentation
Klaus-Dieter Sommer, Michael Heizmann, Andreas Schuetze
TM-TECHNISCHES MESSEN
(2022)
Article
Instruments & Instrumentation
Tanja Dorst, Tizian Schneider, Sascha Eichstaedt, Andreas Schuetze
Summary: This paper proposes the application of measurement uncertainty propagation in machine learning (ML) and extends the automated ML toolbox (AMLT) to consider uncertainty. By applying the principles described in the Guide to the Expression of Uncertainty in Measurement (GUM), uncertainty propagation is carried out for each step in the AMLT. The results show that considering measurement uncertainty in machine learning can effectively assess the reliability of results and the basis of decisions.
TM-TECHNISCHES MESSEN
(2023)
Article
Environmental Sciences
Yannick Robin, Johannes Amann, Payman Goodarzi, Tizian Schneider, Andreas Schuetze, Christian Bur
Summary: In this study, deep learning methods are used to calibrate MOS gas sensors in a complex environment and mitigate the problem of long calibration times and transferring calibrations between sensors. The results show that transfer learning significantly reduces calibration time while maintaining good prediction accuracy.
Editorial Material
Instruments & Instrumentation
Andreas Schuetze
TM-TECHNISCHES MESSEN
(2022)
Editorial Material
Instruments & Instrumentation
Klaus-Dieter Sommer, Michael Heizmann, Andreas Schutze
TM-TECHNISCHES MESSEN
(2023)
Article
Instruments & Instrumentation
Dennis Arendes, Johannes Amann, Cyril Tessier, Oliver Brieger, Andreas Schuetze, Christian Bur
Summary: This article presents a novel gas mixing apparatus (GMA) that can provide welldefined gas mixtures for calibrating gas sensors. The GMA has been optimized for settling speed and self-monitoring, and can provide up to 14 individual test gases with a wide concentration range. The system is controlled by Python software and can exchange gas mixtures quickly. The article also demonstrates the analytical quantification of the system and the use of photoionization detectors for internal leakage detection.
TM-TECHNISCHES MESSEN
(2023)
Article
Instruments & Instrumentation
Caroline Schultealbert, Tobias Baur, Tilman Sauerwald, Andreas Schuetze
Summary: This study investigates the poisoning effect of cyclic siloxane octamethylcyclotetrasiloxane on a commercially available semiconductor gas sensor in temperature cycled operation. The data is assessed using the Sauerwald-Baur model and the DSR method, and compared with a sensor in constant temperature operation mode. The research identifies a feature, the differential surface oxidation (DSO), derived from the Sauerwald-Baur model, which quantitatively expresses the sensor condition in terms of siloxane poisoning. The study also demonstrates the sensor's self-compensation ability using this feature.
TM-TECHNISCHES MESSEN
(2023)
Proceedings Paper
Chemistry, Analytical
Oliver Brieger, Julian Joppich, Caroline Schultealbert, Tobias Baur, Christian Bur, Andreas Schuetze
Summary: In this study, a setup for simultaneous measurement of substances separated by gas chromatography, mass spectrometry, and metal oxide semiconductor gas sensor is proposed. The behavior of the sensor under different conditions is investigated, enabling the evaluation of gas sensors and measuring chambers in terms of effective GC detector parts.
2022 IEEE INTERNATIONAL SYMPOSIUM ON OLFACTION AND ELECTRONIC NOSE (ISOEN 2022)
(2022)
Proceedings Paper
Chemistry, Analytical
Yannick Robin, Johannes Amann, Payman Goodarzi, Andreas Schuetze, Christian Bur
Summary: With the help of deep learning techniques, particularly convolutional neural networks and neural architecture search, the noise of gas sensor calibration models can be significantly reduced. By applying the concept of transfer learning, the calibration time of sensors can be greatly shortened while maintaining high accuracy.
2022 IEEE INTERNATIONAL SYMPOSIUM ON OLFACTION AND ELECTRONIC NOSE (ISOEN 2022)
(2022)
Proceedings Paper
Chemistry, Analytical
Julian Joppich, Wolfhard Reimringer, Thorsten Conrad, Bettina Mannebeck, Christoph Mannebeck, Christian Bur, Andreas Schuetze
Summary: This article presents an approach for calibrating and validating gas sensor systems for odor monitoring based on a draft of a standard, and identifies the limits and restrictions during the conducted measurements. Sensor systems were set up near or at odor intense plants and trained with diluted samples of the target odor. Three calibration campaigns were conducted, and the models built using the data from at least the first campaign were investigated for their ability to interpret data from other campaigns and field inspection data. The results highlight starting points for further investigations.
2022 IEEE INTERNATIONAL SYMPOSIUM ON OLFACTION AND ELECTRONIC NOSE (ISOEN 2022)
(2022)
Proceedings Paper
Chemistry, Analytical
Julian Joppich, Oliver Brieger, Ksenia Karst, Daniel Becher, Christian Bur, Andreas Schutze
Summary: An approach for training temperature cycled metal oxide semiconductor gas sensors to monitor food quality is proposed in this study. A setup consisting of various gas sensors and a GC-MS is utilized to measure the headspace of strawberries in food containers. The food quality is assessed based on appearance, smell, and overall edibility by several individuals. The connection between human perception, GC-MS component peaks, and gas sensor data is evaluated, and significant features are extracted from the temperature cycle to develop a model that can correlate the sensor data with the assessed edibility of the food under investigation.
2022 IEEE INTERNATIONAL SYMPOSIUM ON OLFACTION AND ELECTRONIC NOSE (ISOEN 2022)
(2022)
Article
Chemistry, Analytical
Mengmeng Guo, Na Luo, Yueling Bai, Zhenggang Xue, Qingmin Hu, Jiaqiang Xu
Summary: A porous heterostructure WO3-C/In2O3 was designed and prepared for a miniature H2 sensor, which showed higher response value, lower operating temperature, fast response-recovery speed, and low limit of detection.
SENSORS AND ACTUATORS B-CHEMICAL
(2024)
Article
Chemistry, Analytical
Feng Hu, Hui Hu, Yuting Li, Xiaohui Wang, Xiaowen Shi
Summary: Arsenic contamination in water bodies is a significant health risk. This study developed a chitosan-catechol modified electrode for rapid and accurate detection of trace amounts of arsenic. The modified electrode demonstrated good detection capability and resistance to ionic interference, making it suitable for in situ detection.
SENSORS AND ACTUATORS B-CHEMICAL
(2024)
Article
Chemistry, Analytical
Yantao Zhang, Qian Liu, Tao Tian, Chunhua Xu, Pengli Yang, Lianju Ma, Yi Hou, Hui Zhou, Yongjun Gan
Summary: In this study, a lysosome-targeting buffering fluorogenic probe (Lyso-BFP) was designed and synthesized, demonstrating excellent photostability, pH specificity, and responsiveness to lysosomal acidification in living cells. The performance of Lyso-BFP in pH sensing was attributed to the inhibition of the photo-induced electron transfer process. Lyso-BFP allowed for wash-free imaging and long-term real-time monitoring of lysosome pH changes based on its off-on fluorescence behavior and buffer strategy.
SENSORS AND ACTUATORS B-CHEMICAL
(2024)
Article
Chemistry, Analytical
Wei Cai, Wenbo Sun, Jiayue Wang, Xiaokui Huo, Xudong Cao, Xiangge Tian, Xiaochi Ma, Lei Feng
Summary: In this study, a near-infrared fluorescent probe HCBG was developed for imaging of alpha-GLC. HCBG exhibited excellent selectivity and sensitivity towards alpha-GLC in complex bio-samples, and showed good cell permeability for in situ real-time imaging. Through the high-throughput screening system established by HCBG, a natural alpha-GLC inhibitor was successfully isolated and identified. This study provides a novel fluorescence visualization tool for discovering and exploring the biological functions of diabetes-related gut microbiota, and a high-throughput screening approach for alpha-GLC inhibitor.
SENSORS AND ACTUATORS B-CHEMICAL
(2024)
Article
Chemistry, Analytical
Trey W. Pittman, Xi Zhang, Chamindie Punyadeera, Charles S. Henry
Summary: Heart failure is a growing epidemic and a significant clinical and public health problem. Researchers have developed a portable and affordable diagnostic device for heart failure that can be used at the point-of-care, providing a valid alternative to current diagnostics approaches.
SENSORS AND ACTUATORS B-CHEMICAL
(2024)
Article
Chemistry, Analytical
Anders O. Tjell, Barbara Jud, Roland Schaller-Ammann, Torsten Mayr
Summary: An optical hydrogen peroxide sensor based on catalytic degradation and the detection of produced oxygen is presented. The sensor offers higher resolution and better sensitivity at lower H2O2 concentrations. By removing O2 from the sample solution, a more sensitive O2 sensor can be used for measurement. The sensor has been successfully applied in a flow-through cell to measure H2O2 concentration in different flow rates.
SENSORS AND ACTUATORS B-CHEMICAL
(2024)
Article
Chemistry, Analytical
Seong Jae Kim, Ji-hun Jeong, Gaabhin Ryu, Yoon Sick Eom, Sanha Kim
Summary: Surface-enhanced Raman spectroscopy (SERS) is a high-sensitivity, label-free detection method with various analytical applications. Researchers have developed a hydrophobic SERS substrate based on engineered carbon nanotube arrays (CNT-SERS) and studied the role of structural design at both micro and nanoscales. The substrate demonstrated controlled self-enrichment capability and enhanced sensitivity, with a significant increase in the SERS signal. The study also proposed a theoretical model and a concentration strategy inspired by plants for analyte deposition on microarrays.
SENSORS AND ACTUATORS B-CHEMICAL
(2024)
Article
Chemistry, Analytical
Dan Zhao, Renjun Jiang, Xiaoqiang Liu, Subbiah Alwarappan
Summary: In this study, a novel ternary composite material was constructed by assembling cerium vanadate nanorods on reduced graphene oxide-microcrystalline cellulose nanosheets, and it was used for real-time monitoring of the concentration of superoxide anions in vivo. The ternary composite showed excellent conductivity, large surface area, and abundant active sites, leading to a wider linear range, high sensitivity, low detection limit, and fast response time for superoxide anion detection.
SENSORS AND ACTUATORS B-CHEMICAL
(2024)
Article
Chemistry, Analytical
Tengfei Wang, Liwen Wang, Guang Wu, Dating Tian
Summary: In this study, a covalent organic framework material TaTp-COF with porous and uniform spheres was successfully prepared via hydrothermal reaction, and it was found to significantly enhance the aggregation-induced emission (AIE) of berberine. The unique emission properties of berberine on TaTp-COF were studied and utilized for the sensitive detection of berberine.
SENSORS AND ACTUATORS B-CHEMICAL
(2024)
Article
Chemistry, Analytical
Lin Li, Yilei Ding, Lei Xu, Shuoran Chen, Guoliang Dai, Pengju Han, Lixin Lu, Changqing Ye, Yanlin Song
Summary: In this study, a novel TTI based on a ratiometric fluorescent nanosensor is designed, which has the advantages of high accuracy and low cost. Experimental and theoretical investigations confirm its pH responsiveness and demonstrate its good sensitivity and reliability. By monitoring the total volatile basic nitrogen, this TTI can accurately predict food spoilage and can be adaptively modified for different types of food. The TTI based on this nanosensor enables visual monitoring of food quality.
SENSORS AND ACTUATORS B-CHEMICAL
(2024)
Article
Chemistry, Analytical
Fangju Chen, Xueting Wang, Wei Chen, Chenwen Shao, Yong Qian
Summary: Lung cancer is the second most common malignant tumor worldwide. Drug resistance in lung cancer leads to treatment failure and recurrence in majority of patients. This study developed a fluorescent prodrug that can be activated in cancer cells to release drugs, and its signal can be tracked by imaging. It shows a unique autophagy-driven ferroptosis effect, indicating its potential for targeting drug-resistant cancer cells.
SENSORS AND ACTUATORS B-CHEMICAL
(2024)
Article
Chemistry, Analytical
Weichao Li, Qiming Yuan, Zhangcheng Xia, Xiaoxue Ma, Lifang He, Ling Jin, Xiangfeng Chu, Kui Zhang
Summary: This study developed a high-performance gas sensor for formaldehyde detection by modifying ZnSnO3 with ZnO QDs and SnO2 QDs. The modified sensor showed improved sensing response and lower working temperature. The presence of ZnO QDs formed rich heterojunctions, increased surface area, and provided oxygen deficiency for formaldehyde sensing reaction, thus enhancing the sensor performance. This research provides an alternative method to enhance the sensing properties of MOS by QDs modification.
SENSORS AND ACTUATORS B-CHEMICAL
(2024)
Article
Chemistry, Analytical
Joung-Il Moon, Eun Jung Choi, Younju Joung, Jin-Woo Oh, Sang-Woo Joo, Jaebum Choo
Summary: A novel nanoplasmonic substrate was developed for biomedical applications, which showed strong hot spots for detecting biomarkers at low concentrations. The substrate, called AuNPs@M13, was made by immobilizing 60 nm gold nanoparticles onto the surface of an M13 bacteriophage scaffold. It demonstrated higher sensitivity and lower limit of detection compared to commercially available assays.
SENSORS AND ACTUATORS B-CHEMICAL
(2024)
Article
Chemistry, Analytical
Ning Li, Ya Zhang, Ying Xu, Xiaofang Liu, Jian Chen, Mei Yang, Changjun Hou, Danqun Huo
Summary: The molecular subtype of breast cancer guides treatment and drug selection. Invasive tests can promote cancer cell metastasis, so the development of high-performance, low-cost diagnostic tools for cancer prognosis is crucial. Liquid biopsy techniques enable noninvasive, real-time, dynamic, multicomponent, quantitative, and long-term observations at the cellular, genetic, and molecular levels. A Cu-Zr metal-organic framework (MOF) nanoenzyme with monatomic Cu attachment has been synthesized and proven to have high catalytic performance. The sensor constructed using this nanoenzyme shows potential for accurate classification of breast cancer serum samples.
SENSORS AND ACTUATORS B-CHEMICAL
(2024)
Article
Chemistry, Analytical
Jeongmin Kim, Hyemin Kim, Seunghyun Park, Hyeonaug Hong, Yong Jae Kim, Jiyong Lee, Jaeho Kim, Seung-Woo Cho, Wonhyoung Ryu
Summary: This study presents a method to fabricate independently functioning microneedle (MN) electrodes with narrow intervals for high precision electrochemical sensing. The optimized mixture of photocurable polymer and single-wall carbon nanotubes was used to mold single composite MNs, which were then attached to pre-patterned electrodes. Plasma etching and electropolymerization were performed to enhance the electrochemical activity, and Prussian blue and glucose oxidase were electrodeposited on the MNs for glucose detection. The MN electrodes showed good sensitivity and linearity, and the feasibility of glucose detection was demonstrated in an in vivo mouse study.
SENSORS AND ACTUATORS B-CHEMICAL
(2024)