A reinforcement learning approach for reducing traffic congestion using deep Q learning

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

Published: Dec. 12, 2024

Language: Английский

---

Optimized Real-Time Decision Making with EfficientNet in Digital Twin-Based Vehicular Networks

Electronics, Journal Year: 2025, Volume and Issue: 14(6), P. 1084 - 1084

Published: March 9, 2025

Real-time decision-making is vital in vehicular ad hoc networks (VANETs). It essential to improve road safety and ensure traffic efficiency flow. Integrating digital twins within VANET (DT-VANET) creates virtual replicas of physical vehicles, allowing in-depth analysis effective decision-making. Many network applications now use convolutional neural (CNNs). However, the growing model size latency make implementing them real-time systems challenging, most previous studies focusing on using CNNs still face significant challenges. Some models with sustainable performances have recently been proposed. One advanced among EfficientNet. may consider it a family significantly fewer parameters computational costs. This paper proposes EfficientNet-based optimized DT-VANET architecture. investigates performance EfficientNet digital-based networks. Extensive experiments proved that outperforms CNN (ResNet50, VGG16) accuracy, latency, efficiency, convergence time, which proves its effectiveness DT-VANET.

Language: Английский

---

A novel network-level fused deep learning architecture with shallow neural network classifier for gastrointestinal cancer classification from wireless capsule endoscopy images

BMC Medical Informatics and Decision Making, Journal Year: 2025, Volume and Issue: 25(1)

Published: March 31, 2025

Deep learning has significantly contributed to medical imaging and computer-aided diagnosis (CAD), providing accurate disease classification diagnosis. However, challenges such as inter- intra-class similarities, class imbalance, computational inefficiencies due numerous hyperparameters persist. This study aims address these by presenting a novel deep-learning framework for classifying localizing gastrointestinal (GI) diseases from wireless capsule endoscopy (WCE) images. The proposed begins with dataset augmentation enhance training robustness. Two architectures, Sparse Convolutional DenseNet201 Self-Attention (SC-DSAN) CNN-GRU, are fused at the network level using depth concatenation layer, avoiding costs of feature-level fusion. Bayesian Optimization (BO) is employed dynamic hyperparameter tuning, an Entropy-controlled Marine Predators Algorithm (EMPA) selects optimal features. These features classified Shallow Wide Neural Network (SWNN) traditional classifiers. Experimental evaluations on Kvasir-V1 Kvasir-V2 datasets demonstrate superior performance, achieving accuracies 99.60% 95.10%, respectively. offers improved accuracy, precision, efficiency compared state-of-the-art models. addresses key in GI diagnosis, demonstrating its potential efficient clinical applications. Future work will explore adaptability additional optimize complexity broader deployment.

Language: Английский

---

Efficiency meets Accuracy: Benchmarking Object Detection Models for Pathology Detection in Wireless Capsule Endoscopy

IEEE Access, Journal Year: 2024, Volume and Issue: 12, P. 126793 - 126817

Published: Jan. 1, 2024

Language: Английский

---

King Abdulaziz University Hospital Capsule Dataset: A Novel Small-bowel Endoscopic Image Repository from Saudi Arabia

Data in Brief, Journal Year: 2024, Volume and Issue: 57, P. 111093 - 111093

Published: Nov. 8, 2024

Wireless Capsule Endoscopy (WCE) has fundamentally transformed diagnostic methodologies for small-bowel (SB) abnormalities, providing a comprehensive and non-invasive gastrointestinal assessment in contrast to conventional endoscopic procedures. The King Abdulaziz University Hospital (KAUHC) dataset comprises annotated WCE images specifically curated Saudi Arabian residents. Comprising 10.7 million frames derived from 157 studies, KAUHC been classified into Normal, Arteriovenous Malformations, Ulcer categories. Following the application of specific inclusion exclusion criteria, 3301 labeled 86 studies were identified. Upon admission patients, data collection phase was initiated, involving administration OMOM capsule use recording device video documentation. A thorough evaluation these recordings undertaken by multiple gastroenterologists identify any pathological abnormalities. identified observations are subsequently extracted, categorized, prepared validation using Machine Learning (ML) classifiers. aims not only address scarcity imaging resources Middle East but also advance development tools ML applications SB abnormalities exploratory research on diseases.

Language: Английский

---