Review
Biochemical Research Methods
Karin Elimelech-Zohar, Yaron Orenstein
Summary: Nucleic-acid G-quadruplexes (G4s) are crucial in cellular processes, and experimental assays have been developed to measure them in high throughput. This has enabled the development of machine-learning-based methods, particularly deep neural networks, to predict G4s in any nucleic-acid sequence and species.
BRIEFINGS IN BIOINFORMATICS
(2023)
Review
Computer Science, Artificial Intelligence
Venkata Anuhya Ardeti, Venkata Ratnam Kolluru, George Tom Varghese, Rajesh Kumar Patjoshi
Summary: Cardiovascular diseases have become the leading cause of death globally in the past decade. Early prediction of these diseases using efficient tools like the electrocardiogram (ECG) can help in reducing complications for high-risk patients. Signal processing techniques, including machine learning approaches, are used to analyze and classify ECG signals for the early detection and diagnosis of cardiac conditions. The integration of IoT in healthcare industry enables remote diagnosis and monitoring of patient status in a smart and efficient manner.
EXPERT SYSTEMS WITH APPLICATIONS
(2023)
Article
Engineering, Industrial
Eduardo M. Coraca, Janito Ferreira, Euripedes G. O. Nobrega
Summary: Vibration-based structural health monitoring requires multiple sensors for reliable monitoring, which necessitates the development of machine learning methods. Recent advancements in deep learning techniques applied to vibration have shown promise in pattern identification from high-dimensional data. However, the lack of expert annotated labels related to damage conditions in real structures has hindered the use of supervised techniques, leading to the development of unsupervised methods. A proposed unsupervised framework combines Variational Autoencoders and a Hidden Markov Model to learn a degradation model and classify the state evolution from measured vibration signals, showing promising results for structural health monitoring.
RELIABILITY ENGINEERING & SYSTEM SAFETY
(2023)
Article
Engineering, Multidisciplinary
Jack Poole, Paul Gardner, Nikolaos Dervilis, Lawrence Bull, Keith Worden
Summary: This article discusses the limitation of labelled data in the practical application of structural health monitoring, and introduces transfer learning methods, specifically domain adaptation. By using statistic alignment, the performance degradation issue under class imbalance in traditional methods is addressed, and the effectiveness of statistic alignment is demonstrated in numerical and real-world scenarios.
STRUCTURAL HEALTH MONITORING-AN INTERNATIONAL JOURNAL
(2023)
Article
Engineering, Multidisciplinary
Bin Zhang, Xiaobin Hong, Yuan Liu
Summary: This article proposes a cross-structure ultrasonic guided wave structural health monitoring method based on distribution adaptation deep transfer learning to solve the feature generalization problem in different monitoring structures. The experimental results show that the method can utilize single-sensor monitoring data in one structure to achieve multi-sensor damage monitoring in another structure and achieve damage imaging visualization, with significantly superior imaging performance compared to existing methods.
STRUCTURAL HEALTH MONITORING-AN INTERNATIONAL JOURNAL
(2022)
Article
Engineering, Electrical & Electronic
Yekta Said Can, Bhargavi Mahesh, Elisabeth Andre
Summary: An automatic emotion recognition system is essential for various daily life applications, from monitoring emotional well-being to improving quality of life through better emotion regulation. Recognizing emotions based on physiological signals provides a reliable means, which can be captured by wearable devices like smart watches. However, the shift from laboratory to daily life research presents challenges such as low data quality, subjective self-reports, movement-related changes, and artifacts in physiological signals.
PROCEEDINGS OF THE IEEE
(2023)
Review
Computer Science, Artificial Intelligence
Giovanna Castellano, Gennaro Vessio
Summary: This paper provides an overview of deep learning approaches in visual arts, highlighting opportunities for computer science researchers to assist the art community with automatic tools for analyzing and understanding visual arts. Deeper understanding of visual arts has the potential to make them more accessible to a wider population, ultimately supporting the spread of culture.
NEURAL COMPUTING & APPLICATIONS
(2021)
Review
Energy & Fuels
Jiangang Lu, Ruifeng Zhao, Bo Liu, Zhiwen Yu, Jinjiang Zhang, Zhanqiang Xu
Summary: Non-intrusive load monitoring (NILM) analyzes voltage and current data at a single point on the bus to obtain detailed electricity consumption information of each appliance, which is important for improving power grid efficiency and user energy efficiency. Signature extraction, a critical step in NILM, allows algorithms to achieve accurate state detection and energy disaggregation. The development and application of voltage-current (V-I) trajectory signatures for appliance identification present an intermediate domain between computer vision and NILM. Identifying V-I trajectories enables the detection of appliance operating states and can further promote the field's development.
Article
Agriculture, Multidisciplinary
Suharjito, Gregorius Natanael Elwirehardja, Jonathan Sebastian Prayoga
Summary: This research focuses on creating a mobile application using lightweight CNN models, ImageNet transfer learning, 9-angle crop data augmentation, and post-training quantization. The best model achieved an overall test accuracy of 0.893 on TensorFlow Lite with 96 ms classification time per image, outperforming other models.
COMPUTERS AND ELECTRONICS IN AGRICULTURE
(2021)
Review
Environmental Sciences
Weifeng Chen, Guangtao Shang, Aihong Ji, Chengjun Zhou, Xiyang Wang, Chonghui Xu, Zhenxiong Li, Kai Hu
Summary: This paper introduces the development of VSLAM technology and semantic VSLAM based on deep learning. It emphasizes the importance of semantic information for robots to understand the environment and provides some classic VSLAM open-source algorithms.
Article
Engineering, Electrical & Electronic
Yichu Xu, Lefei Zhang, Bo Du, Liangpei Zhang
Summary: Hyperspectral anomaly detection (HAD) is an important image application that can identify pixels with anomalous spectral signatures without prior information. This review focuses on HAD based on machine learning methods and discusses traditional machine learning and deep-learning-based approaches. Several representative methods are evaluated on real datasets, and conclusions and future directions are summarized.
IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING
(2022)
Article
Green & Sustainable Science & Technology
Yongchao Zhu, Caichao Zhu, Jianjun Tan, Yong Tan, Lei Rao
Summary: This study proposes a novel method for operational state prediction of wind turbine generators (WTGs) using limited monitoring data and fault information. The proposed method combines long short-term memory, fuzzy synthesis, and feature-based transfer learning to address the discrepancy in data distribution among the WTGs. Experimental results demonstrate that the proposed method can sensitively detect potential faults in advance and achieve high accuracy.
Article
Engineering, Multidisciplinary
Ruihan Wang, Hui Chen, Cong Guan
Summary: This study proposed an innovative RCNN structure for intelligent health monitoring of diesel engines, utilizing deep learning and ensemble learning. The framework automatically extracts discriminative features of vibration signals, accelerates network training with Adabound optimizer, and combines diagnostic results from multiple CNNs. Experimental results show the efficiency and superiority of the proposed RCNN.
Article
Computer Science, Artificial Intelligence
Sunil Kumar Prabhakar, Seong-Whan Lee
Summary: With the advancement of high-tech multimedia technologies, music resources are now available online, prompting the need for classifying different music genres. A robust music classifier is required to easily tag unlabeled music and enhance user experience in media players. Existing approaches using manual feature extraction and traditional machine learning techniques have limitations in accuracy, multiclass classification, and handling large datasets. This study proposes five innovative approaches for music genre classification, achieving a classification accuracy of 93.51% with the deep learning BAG model.
EXPERT SYSTEMS WITH APPLICATIONS
(2023)
Article
Chemistry, Multidisciplinary
Laiqa Rukhsar, Waqas Haider Bangyal, Muhammad Sadiq Ali Khan, Ag Asri Ag Ibrahim, Kashif Nisar, Danda B. Rawat
Summary: RNA-Seq analysis is a useful tool for finding cancer specific biomarkers. This study analyzed RNA-Seq data from five cancer types using deep learning models, and found that Convolutional Neural Network (CNN) achieved the best results for classification.
APPLIED SCIENCES-BASEL
(2022)
Article
Automation & Control Systems
Muhammad Shafiq, Kalpana Thakre, Kalluri Rama Krishna, Noel Jeygar Robert, Ashok Kuruppath, Devendra Kumar
Summary: Smart industries utilize modern technologies like machine learning and big data for supply chain management and increased productivity. However, quality control remains a major challenge, particularly in its impact on production rate. These industries rely on supervised learning to improve inspection and effectively control production parameters, selecting mechanisms that improve production and ensure high quality.
INTERNATIONAL JOURNAL OF ADVANCED MANUFACTURING TECHNOLOGY
(2023)
Article
Chemistry, Analytical
Habib Ullah Khan, Shah Nazir
Summary: In today's advanced world, artificial intelligence and machine learning play a crucial role in performing complex tasks efficiently. However, these technologies require a large amount of data, which can be collected through the internet of things. Smart sensors and intelligent hardware components can be developed to provide effective services, especially in smart cities.
Article
Environmental Sciences
Binyue Deng, Denghui Zhang, Fashan Dong, Junjian Zhang, Muhammad Shafiq, Zhaoquan Gu
Summary: Deep neural networks improve remote sensing technology by extracting valuable information from images. However, they are vulnerable to adversarial attacks, which can affect the normal recognition and detection of remote sensing systems. To address this, we propose a rust-style adversarial patch generation framework based on style transfer to reduce detection and improve attack success rate in the physical domain.
Article
Computer Science, Information Systems
Anwar Hussain, Shah Nazir, Fazlullah Khan, Lewis Nkenyereye, Ayaz Ullah, Sulaiman Khan, Sahil Verma, Kavita
Summary: Future communication technologies like 6G can provide higher mobility and better quality-of-service requirements to the Internet of Things (IoT). To handle the demands of large-scale heterogeneous IoT networks, reliable and scalable solutions are needed, such as the proxy mobile IPv6 (PMIPv6) protocol. In this article, a demand-based resource-efficient location-aware PMIPv6 extension is proposed, which utilizes location information and received signal strength (RSS) to enhance the performance of the PMIPv6 protocol in terms of signaling cost and load distribution. Comparison with existing RSS-based PMIPv6 extension protocols shows that the proposed scheme improves performance and is resource-friendly for next-generation large-scale IoT networks.
IEEE INTERNET OF THINGS JOURNAL
(2023)
Article
Computer Science, Information Systems
Ankur Gupta, Habib Ullah Khan, Shah Nazir, Muhammad Shafiq, Mohammad Shabaz
Summary: The metaverse promises immersive experiences, but privacy, security, and control issues must be resolved. This paper focuses on the security issues and enabling technologies/platforms in the metaverse. It also addresses the challenges faced by developers, service providers, and stakeholders, which, if ignored, could hinder adoption and appeal. Finally, ideas for building a viable Zero-Trust Architecture model for the metaverse are presented.
Article
Engineering, Electrical & Electronic
Muhammad Shafiq, Changqing Du, Nasir Jamal, Junaid Hussain Abro, Tahir Kamal, Salman Afsar, Md. Solaiman Mia
Summary: According to the World Health Organization, heart disease is the leading cause of death globally. Detecting cardiovascular disease in its early stages may help reduce the overall mortality rate. Recent advancements in technologies such as the Internet of Things, cloud storage, and machine learning offer hope for a global paradigm shift. Using sensors to capture vital signs at the bedside has become increasingly common, and there is a need for automated and intelligent identification of heart disorders.
JOURNAL OF SENSORS
(2023)
Article
Computer Science, Information Systems
Wenjun Yao, Muhammad Shafiq, Xiaoxin Lin, Xiang Yu
Summary: As software systems become larger and more complex, there is a need to effectively judge the presence of defects in programs. Current software defect prediction methods only extract semantic information at the syntactic level, lacking features to mine defect manifestations at the semantic level of code. This paper proposes a software defect prediction method based on program semantics feature mining (PSFM) method, which extracts semantic information from code and mines defect features to predict software defects. Experimental results show that the PSFM method outperforms existing software defect prediction methods in terms of F-measure value.
Article
Computer Science, Information Systems
Habib Ullah Khan, Farhad Ali, Yazeed Yasin Ghadi, Shah Nazir, Inam Ullah, Heba G. Mohamed
Summary: Improvements in communication and networking technologies have transformed people's lives and organizations' activities. Web 2.0 innovation has provided a variety of hybridized applications and tools that have changed enterprises' functional and communication processes.
Review
Chemistry, Multidisciplinary
Waqas Latif, Claudia Ciniglia, Manuela Iovinella, Muhammad Shafiq, Stefania Papa
Summary: White Rot Fungi (WRF) have the ability to degrade a wide range of pollutants in industrial wastewater through their unique enzymatic system. This literature review focuses on WRF's potential applications, including bioremediation, biosorption, and co-culturing with bacteria. Despite challenges in scaling up WRF-based treatment facilities, they can play an important role in degrading complex organic and inorganic pollutants not treated by conventional wastewater treatment plants.
APPLIED SCIENCES-BASEL
(2023)
Article
Computer Science, Information Systems
Lin Teng, Yulong Qiao, Muhammad Shafiq, Gautam Srivastava, Abdul Rehman Javed, Thippa Reddy Gadekallu, Shoulin Yin
Summary: This paper proposes a federated learning approach based on prior knowledge and a bilateral segmentation network for image edge extraction. The method distributes training images through federated learning to obtain prior knowledge, and uses a local uniform sparsity method to enhance detail features. By introducing a dilated pyramid pooling layer and multi-scale feature fusion module, the final result is obtained by combining the result with prior knowledge and the result with the context path. Experimental results show that the proposed method greatly improves extraction accuracy compared with traditional and state-of-the-art methods.
IEEE TRANSACTIONS ON NETWORK AND SERVICE MANAGEMENT
(2023)
Article
Computer Science, Information Systems
Habib Ullah Khan, Anwar Hussain, Shah Nazir, Farhad Ali, Muhammad Zubair Khan, Inam Ullah
Summary: The future of communications in the Internet of Things (IoT) environment requires secure, scalable, and resource-efficient mobility solutions to support mobility requirements. The proposed location-based, resource-efficient PMIPv6 extension protocol provides a cost-effective and resource-friendly solution for enhanced mobility in the 6G-enabled IoT environment.
Article
Environmental Sciences
Anwar Shah, Bahar Ali, Fazal Wahab, Inam Ullah, Kassian T. T. Amesho, Muhammad Shafiq
Summary: This paper proposes an entropy-based grid approach (EGO) for outlier detection in clustered data. The proposed method, EGO, detects outliers by using entropy on the dataset or individual cluster. Experimental results show that the proposed approach detects outliers more precisely and improves the precision and compactness of clusters obtained from hard clustering algorithms.
ENVIRONMENTAL SCIENCE AND POLLUTION RESEARCH
(2023)
Review
Computer Science, Information Systems
Ghulam Amjad Hussain, Waqar Hassan, Farhan Mahmood, Muhammad Shafiq, Habibur Rehman, John A. Kay
Summary: In modern power systems, condition based monitoring and diagnosis are crucial for the effective and reliable operation of high voltage equipment (HVE). Among various monitoring techniques, partial discharges (PD) measurement is considered a key method to assess insulation health. Online PD monitoring, as a promising technique, reduces power failure incidents in power system components. This paper provides a comprehensive literature review of current online PD monitoring techniques for different high voltage electric components in power systems, and proposes a smart PD monitoring framework based on wireless sensor board to enhance the overall performance of power systems.
Article
Engineering, Electrical & Electronic
Shoulin Yin, Liguo Wang, Muhammad Shafiq, Lin Teng, Asif Ali Laghari, Muhammad Faizan Khan
Summary: In this study, we propose an object detection and interpretation model based on gradient-weighted class activation mapping and reinforcement learning. By extracting remote sensing image (RSI) features and learning the mapping relationship between image features and text semantic features, we achieve the interpretation and description of RSI content. Experimental results show that the proposed method has high detection accuracy and good description performance for RSIs in complex environments.
IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING
(2023)
Article
Computer Science, Artificial Intelligence
Denghui Zhang, Muhammad Shafiq, Liguo Wang, Gautam Srivastava, Shoulin Yin
Summary: This study proposes a security scheme suitable for computation-limited devices in IoT, achieving secure and efficient transmission of high-resolution remote sensing images using visual cryptography. The recognition performance of small encryption datasets for remote sensing images is improved by fine-tuning the pre-trained model from large-scale datasets.
CAAI TRANSACTIONS ON INTELLIGENCE TECHNOLOGY
(2023)
