Review
Multidisciplinary Sciences
Dongmei Zhu, Dongbo Wang
Summary: This paper mainly focuses on the application of transformers in medical image processing, including image segmentation, reconstruction, and classification. The paper summarizes the improvement mechanisms of transformers and discusses the future development prospects and challenges.
JOURNAL OF RADIATION RESEARCH AND APPLIED SCIENCES
(2023)
Article
Computer Science, Information Systems
Wenjing Shuai, Jianzhao Li
Summary: This study discusses the issues existing in deep learning models for medical image classification and proposes a new algorithm solution to improve the generalization ability and classification accuracy of the model by explicitly utilizing location information and designing global spatial attention.
Review
Computer Science, Information Systems
Irena Galic, Marija Habijan, Hrvoje Leventic, Kresimir Romic
Summary: This work provides an overview of fundamental concepts, state-of-the-art models, and publicly available datasets in the field of medical imaging, with a focus on the application of deep learning methods. It also discusses current research conducted in various medical imaging areas, challenges faced, and future research directions.
Article
Automation & Control Systems
Yanyi Liu, Chen Wang, Yingyou Wen, Yixiang Huo, Jun Liu
Summary: The cellular image analysis system is crucial for disease diagnosis and pharmaceutical research. However, due to the differences in cellular image distribution, the analysis requires customized algorithms and parameter tuning, leading to low automation levels. This study proposes an efficient end-to-end cell segmentation algorithm, ECS-Net, which introduces the proposal focus module (PFM) and enhance mask feature head (EMFH) to improve segmentation accuracy. The algorithm achieves better detection and segmentation accuracy with fewer parameters and computational cost, enhancing cellular image analysis systems. Moreover, considering the medical IoT scenario, the scaled-down model with only 5.8M parameters has a minor decrease in accuracy, but significant application value.
IET CONTROL THEORY AND APPLICATIONS
(2023)
Article
Computer Science, Artificial Intelligence
Dwarikanath Mahapatra, Alexander Poellinger, Mauricio Reyes
Summary: In this paper, we propose an interpretabilit-guided inductive bias approach that enhances deep learning models in medical image analysis by enforcing the extraction of clinically relevant and spatially consistent features. Through experiments on medical image classification and segmentation tasks, we demonstrate that our approach outperforms conventional methods and generates saliency maps in higher agreement with clinical experts.
MEDICAL IMAGE ANALYSIS
(2022)
Article
Radiology, Nuclear Medicine & Medical Imaging
Bingxuan Wu, Fan Zhang, Liang Xu, Shuwei Shen, Pengfei Shao, Mingzhai Sun, Peng Liu, Peng Yao, Ronald X. Xu, Ronald X. Xu
Summary: In this study, a deep neural network called Modality Preserving U-Net (MPU-Net) is proposed for modality preserving analysis and segmentation in multimodal medical images. The experimental results demonstrate the superior performance of MPU-Net in the segmentation tasks for multimodal medical images, indicating its potential application in other segmentation tasks.
QUANTITATIVE IMAGING IN MEDICINE AND SURGERY
(2023)
Article
Computer Science, Artificial Intelligence
Krishna Gopal Dhal, Arunita Das, Swarnajit Ray, Jorge Galvez
Summary: The paper proposes a Histogram Based Fuzzy Clustering technique using an improved version of Firefly Algorithm, outperforming traditional clustering methods in terms of precision, robustness, and quality of the segmented outputs.
KNOWLEDGE-BASED SYSTEMS
(2021)
Article
Computer Science, Artificial Intelligence
Pengyi Hao, Kangjian Shi, Shuyuan Tian, Fuli Wu
Summary: Medical image segmentation from noisy labels is a challenging task. In this study, a novel uncertainty-aware iterative learning (UaIL) approach is proposed to address issues related to noisy annotations. UaIL trains two deep networks iteratively using original and augmented images, with a joint loss function that includes softened label loss, hard label loss, and consistency loss. Experimental results on two public datasets show that UaIL outperforms competing approaches, especially when dealing with serious label noises. UaIL is also validated on a private dataset, demonstrating its applicability in real-world scenarios with noisy labels.
IET IMAGE PROCESSING
(2023)
Article
Multidisciplinary Sciences
Qiong Chen, Lirong Zeng, Cong Lin
Summary: In this paper, we proposed a deep network embedded with rough fuzzy discretization (RFDDN) for OCT fundus image segmentation. By using fuzzy c-means clustering and genetic algorithm, we achieved pixel segmentation. Then, we introduced the deep supervised attention mechanism to obtain important multi-scale information. Experimental results showed that RFDDN outperformed other methods on evaluation indicators, with improvements of 3.3%, 2.6%, and 7.1% in DSC, sensitivity, and specificity compared to ISCLNet, respectively, as well as reductions of 6.6% and 19.7% in HD95 and ASD, respectively.
SCIENTIFIC REPORTS
(2023)
Article
Computer Science, Artificial Intelligence
Muhammad Irfan Ali, Mostafa K. El-Bably, El-Sayed A. Abo-Tabl
Summary: Approximation space is crucial for the accuracy of approximations on a subset of the universal set. This paper aims to develop new soft rough sets models using near open sets, enhancing the accuracy of approximations significantly. The concepts of near soft rough approximations and their properties are proposed, and comparisons with previous methods are made. An algorithm is provided for decision-making problems, and tested on hypothetical data for comparison with existing methods.
Article
Computer Science, Artificial Intelligence
Qian Jiang, Xin Jin, Xiaohui Cui, Shaowen Yao, Keqin Li, Wei Zhou
Summary: Medical image fusion combines multiple features of human tissue from different source images, which is beneficial for clinical diagnosis. This study introduces a similarity measure of fuzzy set theory to abstract and measure fuzzy features in medical image fusion, and proposes a lightweight medical image fusion technique based on this new measure. Experimental results show that the similarity measure of fuzzy set theory achieves excellent performance in medical image fusion.
IEEE TRANSACTIONS ON EMERGING TOPICS IN COMPUTATIONAL INTELLIGENCE
(2023)
Article
Engineering, Biomedical
Xin Wei, Fanghua Ye, Huan Wan, Jianfeng Xu, Weidong Min
Summary: In recent years, deep learning methods have made significant progress in medical image processing. However, the current medical segmentation methods still lack accuracy in segmenting small-scale and variable-scale objects. To address this issue, we propose Triple Attention Network (TANet) which includes a novel Triple Attention Module (TAM). Experimental results demonstrate that TANet outperforms previous models on multiple evaluation metrics and improves the Dice score by up to 7.1%.
BIOMEDICAL SIGNAL PROCESSING AND CONTROL
(2023)
Article
Computer Science, Artificial Intelligence
Juan Wang, Zetao Zhang, Minghu Wu, Yonggang Ye, Sheng Wang, Ye Cao, Hao Yang
Summary: In this work, an improved BlendMask nuclei instance segmentation framework is proposed, which incorporates dilated convolution aggregation module and context information aggregation module to enhance the performance of detecting and segmenting dense small objects and adhering nuclei. A distributional ranking loss function is also introduced to alleviate the imbalance between the target and the background. The proposed method outperforms several recent classic open-source nuclei instance segmentation methods on the DSB2018 dataset, achieving a 3.6% improvement on AP segmentation metric compared to BlendMask.
IET IMAGE PROCESSING
(2023)
Article
Computer Science, Information Systems
Abhishek Bal, Minakshi Banerjee, Amlan Chakrabarti, Punit Sharma
Summary: Automated brain tumor segmentation of MR image is a challenging task. The proposed method using rough-fuzzy C-means and shape based topological properties achieved better performance in terms of statistical volume metrics compared to previous algorithms.
JOURNAL OF KING SAUD UNIVERSITY-COMPUTER AND INFORMATION SCIENCES
(2022)
Article
Engineering, Electrical & Electronic
Carlos M. J. M. Dourado Jr, Suane Pires P. da Silva, Raul Victor M. da Nobrega, Pedro P. Reboucas Filho, Khan Muhammad, Victor Hugo C. de Albuquerque
Summary: The research proposes a new online approach based on deep learning tools and transfer learning concept to generate a computational intelligence framework for use with Internet of Health Things devices. The efficiency and reliability of the framework is validated using three medical databases, showing high classification accuracy.
IEEE JOURNAL ON SELECTED AREAS IN COMMUNICATIONS
(2021)
Article
Management
Peiman Ghasemi, Fariba Goodarzian, Angappa Gunasekaran, Ajith Abraham
Summary: This paper proposes a bi-level mathematical model to address the location, routing and allocation of medical centers to distribution depots during the COVID-19 pandemic outbreak. The results show that the objective functions of the interdictor and fortifier models increase with the increasing demand.
INTERNATIONAL JOURNAL OF LOGISTICS MANAGEMENT
(2023)
Article
Automation & Control Systems
Rizk M. Rizk-Allah, Aboul Ella Hassanien
Summary: This paper proposes a novel algorithm named EO-PS, based on the hybridization of equilibrium optimizer and pattern search techniques, for accurate and reliable wind farm layout optimization design. The algorithm operates in two phases, utilizing equilibrium optimizer in the first phase to explore the search space and pattern search in the second phase to guide the searching towards better solutions. The algorithm is implemented and tested on irregular land space in Egypt, achieving optimal layout configuration for practical planning trends. The comprehensive results and analyses confirm the competitive performance of EO-PS in terms of solution quality and reliability.
Article
Computer Science, Artificial Intelligence
Meera Ramadas, Ajith Abraham
Summary: Air pollution is a global issue that can cause major health hazards. Satellite remote sensing is an effective way to monitor the atmosphere and improve understanding of complex images through clustering and segmentation techniques. The novel DiDE algorithm showed superior outcomes compared to traditional approaches, and its application in multi-level thresholding significantly reduced computational delay and improved image quality.
NEURAL COMPUTING & APPLICATIONS
(2023)
Article
Computer Science, Information Systems
E. Rajalakshmi, R. Elakkiya, V. Subramaniyaswamy, L. Prikhodko Alexey, Grif Mikhail, Maxim Bakaev, Ketan Kotecha, Lubna Abdelkareim Gabralla, Ajith Abraham
Summary: A novel vison-based hybrid deep neural net methodology is proposed in this study for recognizing Indian and Russian sign gestures. The proposed framework aims to establish a single framework for tracking and extracting multi-semantic properties, such as non-manual components and manual co-articulations. By using a 3D deep neural net with atrous convolutions for spatial feature extraction, attention-based Bi-LSTM for temporal and sequential feature extraction, modified autoencoders for abstract feature extraction, and a hybrid attention module for discriminative feature extraction, the proposed sign language recognition framework yields better results than other state-of-the-art frameworks.
Article
Computer Science, Information Systems
S. U. Aswathy, P. P. Fathimathul Rajeena, Ajith Abraham, Divya Stephen
Summary: Lung malignancy, one of the most common types of cancer worldwide, was studied in this research. The study focused on the multifaceted nature of lung cancer diagnosis and proposed a method using nanotechnology for precise segmentation of lesions in nano-CT images. The results showed high accuracy and precision in tumor classification and segmentation.
Article
Computer Science, Artificial Intelligence
Benkuan Cui, Kun Ma, Leping Li, Weijuan Zhang, Ke Ji, Zhenxiang Chen, Ajith Abraham
Summary: Despite the benefits provided by the Internet and social media, the proliferation of fake news has had negative effects on society and individuals. This paper proposes a Chinese fake news detection model using a Third-order Text Graph Tensor and Information Propagation Network. Data augmentation and a novel text graph tensor representation are employed to address the challenges of feature sparsity and capturing context information. The model outperforms existing methods in fake news detection according to experimental results on four public datasets.
APPLIED INTELLIGENCE
(2023)
Article
Multidisciplinary Sciences
Mincheol Shin, Mucheol Kim, Geunchul Park, Ajith Abraham
Summary: High-performance computing supports advancements in various scientific disciplines by providing computing power and insights. This paper proposes an adaptive variable sampling model for performance analysis in high-performance computing environments. The model automatically selects optimal variables for performance prediction without requiring expert knowledge. Experiments show that the model improves speed by at least 24.25% and up to 58.75% without sacrificing accuracy.
Review
Automation & Control Systems
Shreyas Gawde, Shruti Patil, Satish Kumar, Pooja Kamat, Ketan Kotecha, Ajith Abraham
Summary: Industry 4.0 is the era of smart manufacturing, which relies heavily on machinery. Maintaining critical rotating machines is the top priority for engineers to minimize unplanned shutdowns and increase their useful life. This paper aims to provide a systematic literature review on the data-driven approach for multi-fault diagnosis of industrial rotating machines, highlighting the foundational work, comparative study, major challenges, and research gaps in this field.
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE
(2023)
Article
Automation & Control Systems
Alicia Passah, Samarendra Nath Sur, Ajith Abraham, Debdatta Kandar
Summary: Research in artificial intelligence focuses on teaching machines to understand and interpret visual data in the field of computer vision. Through the use of digital images, deep learning models, and synthetic aperture radar (SAR), machines can properly recognize and classify items and respond accordingly. This paper presents a survey on different techniques and architectures proposed for SAR image applications, covering target detection and recognition models, analyzing their techniques and performances. It provides novel discussions, comparisons, and observations, highlighting the advantages and disadvantages of different approaches to inspire future research and suggest potential directions for hybrid models.
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE
(2023)
Article
Computer Science, Information Systems
Amir El-Ghamry, Ashraf Darwish, Aboul Ella Hassanien
Summary: Smart farming is an advanced approach to managing a farm, which involves monitoring crop health and productivity using technology and information. The Internet of Things enables smart farming by collecting and storing data, but also exposes it to cyber-attacks. Therefore, an intrusion detection system that can adapt to the challenges of IoT networks in agriculture is crucial.
INTERNET OF THINGS
(2023)
Article
Computer Science, Information Systems
Mayur Wankhade, Chandra Sekhara Rao Annavarapu, Ajith Abraham
Summary: Sentiment classification is a crucial task in natural language processing. This research investigates the impact of text preprocessing techniques on sentiment classification and proposes a novel framework called CBMAFM that leverages the synergistic power of CNN and BiLSTM through a multi-attention fusion mechanism. The framework preserves both local and global context dependencies, resulting in improved performance compared to other state-of-the-art methods.
MULTIMEDIA TOOLS AND APPLICATIONS
(2023)
Article
Computer Science, Information Systems
Behzad Saemi, Ali Asghar Rahmani Hosseinabadi, Azadeh Khodadadi, Seyedsaeid Mirkamali, Ajith Abraham
Summary: The task scheduling problem in Mobile Cloud Computing (MCC) is a difficult problem to solve, and this study proposes a non-dominated multi-objective strategy based on the Harris Hawks Optimization (HHO) technique to address this issue. By comparing with other algorithms, it is found that the proposed method performs better in terms of job completion time and energy savings.
Article
Computer Science, Information Systems
Deepali Arun Bhanage, Ambika Vishal Pawar, Ketan Kotecha, Ajith Abraham
Summary: This paper proposes a semantic log analysis model that utilizes three log features to capture the essence of the log message. By employing the BERT pre-trained model and an attention-based OLSTM classifier, the proposed model is able to detect failures in different infrastructures. The evaluation results demonstrate that the system delivers improved and stable results across various IT infrastructures.
Article
Computer Science, Information Systems
Poria Pirozmand, Ali Asghar Rahmani Hosseinabadi, Maedeh Jabbari Chari, Faezeh Pahlavan, Seyedsaeid Mirkamali, Gerhard-Wilhelm Weber, Summera Nosheen, Ajith Abraham
Summary: In this study, a discrete metaheuristic method called D-PFA is proposed to efficiently solve the Traveling Salesman Problem (TSP). By comparing and validating with other algorithms, the proposed method has shown to be more competitive and resilient in solving TSP.
Article
Education & Educational Research
Rana Saeed Al-Maroof, Said A. Salloum, Aboul Ella Hassanien, Khaled Shaalan
Summary: This study examines the impact of fear emotion on the adoption of Google Meet by students and teachers during the COVID-19 pandemic. The findings demonstrate that fear, specifically related to family lockdown, education failure, and loss of social relationships, has a significant effect on the adoption of this educational platform. The study also compares different data analysis techniques and finds that the J48 classifier is the most effective in predicting the dependent variable in most cases.
INTERACTIVE LEARNING ENVIRONMENTS
(2023)
