Emerging Intelligent Computing Techniques and their Applications Using Machine Learning
Closed ---
Purchase IssueJournal: Recent Advances in Electrical & Electronic Engineering
Guest editor(s):Dr. Vikash Yadav
Co-Guest Editor(s): Dr. Dimitrios A. Karras,Dr. Chuan-Ming Liu
Introduction
This special issue will aim to discuss computational intelligence approaches, initiatives and applications in engineering and science fields (including Machine Intelligence, Mining Engineering, Modeling and Simulation, Computer, Communication, Networking and Information Engineering, Systems Engineering, Innovative Computing Systems, Adaptive Technologies for Sustainable Growth, and Theoretical and Applied Sciences). This collection should inspire various scholars to contribute research on intelligence principles and approaches in their respective research communities, while enriching the body of research on computational intelligence.
Keywords
Machine Learning, Image Processing, Computational Intelligence, Emerging Technology, Cloud Computing, Fuzzy Logic
HSLE: A Hybrid Ensemble Classifier for Prediction of Heart Disease
Author(s): Pradeep Kumar Kushwaha*, Arvind Dagur and Dhirendra Shukla
DOI: 10.2174/0123520965291887240422051823
Abstract: Background: Detecting heart disease in a timely manner is vital for preventing its progression, as... Read more.
An Efficient Approach for Diabetes Classification Using Feature Selection and Hyperparameter Tuning
Author(s): Bhanu Prakash Lohani*, Arvind Dagur. and Dhirendra Shukla
DOI: 10.2174/0123520965291885240315051751
Abstract: Background: Diabetes mellitus, stemming from insulin deficiency or resistance, poses acute and chronic health issues... Read more.
A Comparative Analysis of Machine Learning Models for the Classification of Heart Failure Patients in the Intensive Care Unit
Author(s): Mateo Gaudin, Swapandeep Kaur, Preeti Sharma* and Rajeev Kumar
DOI: 10.2174/0123520965312805240506113451
Abstract: Background: Heart failure is the leading cause of death globally over the last several decades.... Read more.
Integrating Machine Learning and Feature Extraction for Islanding Detection in Grid-Connected Photovoltaic Systems: A Hybrid Intelligent Approach
Author(s): INDU BHUSHAN*, Subhash Chandra and Arvind Yadav
Abstract
In recent years, the integration of renewable energy sources, particularly photovoltaic (PV) systems, into the grid has garnered considerable attention. However, the distributed nature of these grid-integrated PV systems has introduced challenges concerning grid faults and maintenance. This paper introduces a pioneering approach to augment the monitoring of grid-integrated PV systems by integrating intelligent methods, specifically machine learning and feature extraction techniques.The primary focus of this approach is on islanding detection, which involves promptly identifying grid faults or maintenance challenges and initiating the grid's transition to an isolated mode of operation. To accomplish this, an intelligent signalling method is employed, capitalizing on the capabilities of distributed generation (DG) networks. By recording critical signals such as voltage, current, and frequency at common coupling points, fault conditions can be accurately detected. The process of feature extraction plays a pivotal role in this approach. Signals obtained from the grid are subjected to wavelet transformation to extract pertinent information that characterizes fault conditions. These extracted features are then utilized as inputs to machine learning methods, facilitating the proposal of intelligent islanding scenarios. To assess the efficacy of the proposed approach, simulations are conducted on a grid-connected PV system. The recorded signals are meticulously analyzed, and the extracted features are employed to train machine learning models. The performance of these models is evaluated based on their ability to accurately detect fault conditions and initiate appropriate islanding scenarios.The results obtained demonstrate the immense potential of the proposed approach in bolstering the monitoring of grid-integrated PV systems. By synergizing machine learning techniques with feature extraction and intelligent signaling methods, the detection of grid faults and maintenance challenges can be substantially improved, thereby fostering more efficient and dependable operation of these systems.
Utilizing Deep Learning Techniques for Detecting Image Spam on Social Media Platforms
Author(s): Himani Jain* and Amit Dixit
Abstract
Image spam involves the practice of concealing text within an image. Various machine learning techniques are used to categories image spam, utilizing a wide range of features extracted from the images. Convolutional neural networks (CNNs) are commonly used for image classification and feature extraction tasks because of their outstanding performance. In this study, our main focus is to analyses image spam using a CNN model that incorporates deep learning techniques. This model has been meticulously fine-tuned and optimized to deliver exceptional performance in both feature extraction and classification tasks. In addition, we performed comparative evaluations of our model on different image spam datasets that were specifically created to make the classification task more challenging. The results we obtained show a significant improvement in classification accuracy compared to other methods used on the same datasets.
Multimodal Medical Image Fusion Method based on the Swin Transformer and Self-supervised Contrast Learning
Author(s): Yuwei Wang, Lei Wang*, Zizhen Huang, Yukun Zhang and Yaolong Han
Abstract
Medical image fusion is of great significance for diagnosis and treatment because it provides a strong tool to use the more comprehensive and accurate medical information for doctors. Though great progress has been made by the deep learning-based fusion methods in recent years, there still are some troubling challenges, such as the low contrast, weak preservation ability of the details and edges, the loss of the global information, and poor color fidelity. To deal with these problems, a novel multi-modal medical image fusion method based on the Swin Transformer network and the self-supervised contrastive learning scheme is proposed. As an encoder-decoder architecture, the Swin Transformer can well utilize the hierarchical attention mechanisms to model the feature dependencies at different scales, and effectively capture both of the global and local information. With the help of the four types of defined loss functions, the self-supervised contrastive learning can maximize the similarity between positive samples and minimize the similarity between positive and negative samples to make the fused images closer to the source images in the projected latent space, guaranteeing the fusion quality. By compared with seven state-of-the-art fusion methods, experimental results fully demonstrate the superiority of the proposed method in both subjective and objective evaluations.
Waiting Time Distribution of High and Low Priority Traffic on Blockchain Network under Zero Knowledge Proof
Author(s): Dhaneshwar Kumar*, Rajesh Kumar Shrivastava, Umesh Gupta and Indrajeet Gupta
Abstract
The delay sensitivity of high-priority traffic is a significant concern in blockchain-based networks, particularly in the context of heterogeneous traffic and emergency responses facilitated by zero-knowledge proof (ZKP). In this study, the Pollaczek-Khintchine equation is employed to examine the distribution of waiting times for high and low-priority traffic. The empirical evidence demonstrates that the allocation of high-priority traffic remains unaffected by low- priority traffic factors. Conversely, the allocation of low-priority traffic is contingent upon the high priority.
Active region contour model with pre-fitting contour initialization for fast MRI segmentation of brain tumors
Author(s): Amar Singh*, Rajesh Kumar Shrivastava and Ashutosh Srivastava
Abstract
One of the biggest challenges for active region contour models, which are well-known for their flexibility and accuracy in MRI segmentation, is generating the first mask. The task of creating an initial mask that precisely defines the boundaries of the objects is intrinsically difficult, especially when dealing with intricate architecture or minute fluctuations in intensity. Beginning masks that are too loose or accurate might result in less-than-ideal segmentation outcomes, as the contour may converge to the wrong limits or miss significant features. For MRI brain tumor segmentation, a more sophisticated method is an active region contour model with the pre-fitting mask. This technique enhances the accuracy of tumor border delineation by fusing the adaptability of dynamic contour models with the accuracy of a pre-fitting mask. The MRI can be divided into two classes using an entropy-based methodology, from which the connected region can be extracted. The initial mask for the active region contour model can be created using the appropriate statistics (such as area, perimeter, density, and centroid) of the linked regions. The active contour model is guided by the pre-fitting mask to improve segmentation inside the pre-defined zone. The suggested approach uses iterative optimization techniques, such as gradient descent, to bend the active contour towards the tumor borders to decrease the number of iterations in estimating the tumor region in the MRI. The suggested approach seeks to increase the precision and effectiveness of tumor border delineation by utilizing the flexibility of active contours and the accuracy of a pre-fitting mask. The suggested approach yields good results when compared to existing pre-fitting mask approaches and is verified using the BraTS dataset.
Comparative Analysis of Radiological and Machine Learning-Based Interpretations for Differentiating COVID-19 and Pneumonia
Author(s): Ratish Srivasatava*, Dev Ras Pandey and Ashutosh Kumar Singh
Abstract
In case of any respiratory condition, diagnosis is extremely important, so much so that it would even require an adept professional for such identification. The problem of pneumonia has been recognized as a global issue concerning health, its common occurrence, and devastating outcome. The following paper will look into the complex world in the diagnosis of pneumonia, particularly bacterial, viral, and COVID-19 pneumonia cases. The use of chest X-rays and CT scans has come in the nick of time in the diagnosis of pneumonia by the accumulation of pulmonary symptoms. Radiological interpretations are clearly observer dependent and, hence, subjective, which calls for complementary diagnostic techniques. Deep learning algorithms bring revolutionary value to the analysis of medical images. Machine learning, in every deep learning algorithm, has the power to revolutionize pneumonia diagnosis. This contrasts the radiological interpretation with that by machine learning-based. This study would improve the pneumonia diagnostic perspective by unveiling their strengths, weaknesses, and synergies. The emergence of the COVID-19 outbreak was making it hard for the global health community to single out COVID-19-associated pneumonia, as was the case with its other etiologies. The present study aims to assess subtle differences in the landscape of pneumonia that could better the precision of the diagnosis. Such comparisons between these two diseases may help to boost better diagnostic methods, better patient outcomes, and, in turn, public health measures for the control of respiratory diseases.