Deep Learning Based Modulation Classification for Underwater Optical Wireless Communication System DOI
Dushyant Singh Chauhan, Gurjit Kaur, Dinesh Kumar

et al.

Published: Dec. 7, 2023

The Internet of Underwater Things (IoUT) is an emerging technology that facilitates communication and data sharing among underwater equipment. constrained transfer capacity frequent transmission failures in wireless channels provide significant issues the context IoUT. Modulation categorization crucial for optimizing spectrum allocation, guaranteeing dependable adaptable communication, mitigating interference, assuring network security, enabling various applications optical (UOWC). It has a key role enhancing efficiency user-friendliness UOWC systems. Deep learning (DL), effective classification method achieved success fields application. Nevertheless, its application systems not been thoroughly investigated. This work focuses on utilization DL systems, specifically purpose modulation categorization. A Convolutional Neural Network (CNN) employed to do task. We transform unprocessed modulated signals into constellation images with grid-like structure then input them CNN training network. simulation results demonstrate suggested strategy based CNN, provides comparable level accuracy without requiring manual selection features.

Language: Английский

Self-supervised multi-task learning for medical image analysis DOI
Huihui Yu, Qun Dai

Pattern Recognition, Journal Year: 2024, Volume and Issue: 150, P. 110327 - 110327

Published: Feb. 7, 2024

Language: Английский

Citations

11
PSFHSP-Net: an efficient lightweight network for identifying pubic symphysis-fetal head standard plane from intrapartum ultrasound images DOI Creative Commons
Ruiyu Qiu,

Mengqiang Zhou,

Jieyun Bai

et al.

Medical & Biological Engineering & Computing, Journal Year: 2024, Volume and Issue: 62(10), P. 2975 - 2986

Published: May 9, 2024

Abstract The accurate selection of the ultrasound plane for fetal head and pubic symphysis is critical precisely measuring angle progression. traditional method depends heavily on sonographers manually selecting imaging plane. This process not only time-intensive laborious but also prone to variability based clinicians’ expertise. Consequently, there a significant need an automated driven by artificial intelligence. To enhance efficiency accuracy identifying symphysis-fetal standard (PSFHSP), we proposed streamlined neural network, PSFHSP-Net, modified version ResNet-18. network comprises single convolutional layer three residual blocks designed mitigate noise interference bolster feature extraction capabilities. model’s adaptability was further refined expanding shared into task-specific layers. We assessed its performance against both heavyweight other lightweight models evaluating metrics such as F 1-score, (ACC), recall, precision, area under ROC curve (AUC), model parameter count, frames per second (FPS). PSFHSP-Net recorded ACC 0.8995, 1-score 0.9075, recall 0.9191, precision 0.9022. surpassed in these metrics. Notably, it featured smallest size (1.48 MB) highest processing speed (65.7909 FPS), meeting real-time criterion over 24 images second. While AUC our 0.930, slightly lower than that ResNet34 (0.935), showed marked improvement ResNet-18 testing, with increases 0.0435 0.0306, respectively. However, saw slight decrease from 0.9184 0.9022, reduction 0.0162. Despite trade-offs, compression significantly reduced 42.64 1.48 MB increased inference 4.4753 65.7909 FPS. results confirm capable swiftly effectively PSFHSP, thereby facilitating measurements development represents advancement automating analysis, promising enhanced consistency operator dependency clinical settings. Graphical abstract

Language: Английский

Citations

6
Detection of Masses in Mammogram Images Based on the Enhanced RetinaNet Network With INbreast Dataset DOI Creative Commons
Mingzhao Wang, Ran Liu, Joseph Luttrell

et al.

Journal of Multidisciplinary Healthcare, Journal Year: 2025, Volume and Issue: Volume 18, P. 675 - 695

Published: Feb. 1, 2025

Breast cancer is the most common major public health problems of women in world. Until now, analyzing mammogram images still main method used by doctors to diagnose and detect breast cancers. However, this process usually depends on experience radiologists always very time consuming. We propose introduce deep learning technology into for facilitation computer-aided diagnosis (CAD), address challenges class imbalance, enhance detection small masses multiple targets, reduce false positives negatives analysis. Therefore, we adopted enhanced RetinaNet images. Specifically, introduced a novel modification network structure, where feature map M5 processed ReLU function prior original convolution kernel. This strategic adjustment was designed prevent loss resolution mass features. Additionally, transfer techniques training through leveraging pre-trained weights from other applications, fine-tuned our improved model using INbreast dataset. The aforementioned innovations facilitate superior performance RetiaNet dataset INbreast, as evidenced mAP (mean average precision) 1.0000 TPR (true positive rate) 1.00 at 0.00 FPPI (false per image) experimental results demonstrate that defeats existing models having more generalization than published studies, it can also be applied types patients assist making proper diagnosis.

Language: Английский

Citations

0
Navigating the Challenges and Opportunities of Tiny Deep Learning and Tiny Machine Learning in Lung Cancer Identification DOI Creative Commons

Yasir Salam Abdulghafoor,

Auns Qusai Al-Neami, Ahmed Faeq Hussein

et al.

Al-Nahrain Journal for Engineering Sciences, Journal Year: 2025, Volume and Issue: 28(1), P. 97 - 120

Published: April 7, 2025

Lung cancer is the most common dangerous disease that, if treated late, can lead to death. It more likely be successfully discovered at an early stage before it worsens. Distinguishing size, shape, and location of lymphatic nodes identify spread around these nodes. Thus, identifying lung remarkably helpful for doctors. diagnosed by expert doctors; however, their limited experience may misdiagnosis cause medical issues in patients. In line computer-assisted systems, many methods strategies used predict malignancy level that plays a significant role provide precise abnormality detection. this paper, use modern learning machine-based approaches was explored. More than 70 state-of-the-art articles (from 2019 2024) were extensively explored highlight different machine deep (DL) techniques models detection, classification, prediction cancerous tumors. The efficient model Tiny DL must built assist physicians who are working rural centers swift rapid diagnosis cancer. combination lightweight Convolutional Neural Networks resources could produce portable with low computational cost has ability substitute skill doctors needed urgent cases.

Language: Английский

Citations

0
Improving breast cancer prediction via progressive ensemble and image enhancement DOI
Huong Hoang Luong,

Minh Dat Vo,

Hong Phuc Phan

et al.

Multimedia Tools and Applications, Journal Year: 2024, Volume and Issue: unknown

Published: May 3, 2024

Language: Английский

Citations

3
A hybrid detection model for acute lymphocytic leukemia using support vector machine and particle swarm optimization (SVM-PSO) DOI Creative Commons

Lama K. Alsaykhan,

Mashael Maashi

Scientific Reports, Journal Year: 2024, Volume and Issue: 14(1)

Published: Oct. 8, 2024

Leukemia, a hematological disease affecting the bone marrow and white blood cells (WBCs), ranks among top ten causes of mortality worldwide. Delays in decision-making often hinder timely application suitable medical treatments. Acute lymphoblastic leukemia (ALL) is one primary forms, constituting approximately 25% childhood cancer cases. However, automated ALL diagnosis challenging. Recently, machine learning (ML) has emerged as an important tool for building detection models. In this study, we present hybrid model that improves accuracy process by combining support vector (SVM) particle swarm optimization (PSO) approaches to automatically identify ALL. We use SVM represent two-dimensional image complete classification process. PSO employed enhance performance model, reducing error rates enhancing result accuracy. The input images are obtained from two public datasets (ALL-IDB1 ALL-IDB2), online utilized training testing proposed model. results indicate our SVM-PSO high accuracy, outperforming stand-alone algorithms demonstrating superior performance, enhanced confusion matrix, higher rate. This advancement holds promise quality technical software field using learning.

Language: Английский

Citations

3
In-depth Analysis of Artificial Intelligence for Climate Change Mitigation DOI Open Access

Liyu Lu

Published: Feb. 1, 2024

Due to the major impact of climate change on world's environment, political and economic systems, mitigation has become a pressing priority for international community requires rapid action from whole society. With continuous advancement artificial intelligence research, integration AI other technologies makes it more used in It promising innovative avenue field mitigation. This paper comprehensively considers key role technology mitigation, such as modeling, optimization renewable energy development intelligent solutions sustainable practices CSS technology, affirms its future prospects. also describes challenges As researchers, policymakers, industries collaborate refine methodologies integrate them into practical applications, concerted effort is required establish ethical guidelines, transparency standards, inclusive governance frameworks.

Language: Английский

Citations

2
TFCNet: A texture-aware and fine-grained feature compensated polyp detection network DOI

Xiaoying Pan,

Yaya Mu,

Chenyang Ma

et al.

Computers in Biology and Medicine, Journal Year: 2024, Volume and Issue: 171, P. 108144 - 108144

Published: Feb. 14, 2024

Language: Английский

Citations

1
A Fusion Framework of Pre-trained Deep Learning Models for Oral Squamous Cell Carcinoma Classification DOI
Muhammad Attique Khan,

Momina Mir,

Muhammad Sami Ullah

et al.

Lecture notes in networks and systems, Journal Year: 2024, Volume and Issue: unknown, P. 769 - 782

Published: Jan. 1, 2024

Language: Английский

Citations

1
Wafer defect identification with optimal hyper-parameter tuning of support vector machine using the deep feature of ResNet 101 DOI
Santi Kumari Behera,

Shishir Dash,

Rajat Amat

et al.

International Journal of Systems Assurance Engineering and Management, Journal Year: 2023, Volume and Issue: 15(3), P. 1294 - 1304

Published: Dec. 26, 2023

Language: Английский

Citations

2