Intelligent Deep Learning Enabled Oral Squamous Cell Carcinoma Detection and Classification Using Biomedical Images

Computational Intelligence and Neuroscience, Journal Year: 2022, Volume and Issue: 2022, P. 1 - 11

Published: June 30, 2022

Oral cancer is one of the lethal diseases among available malignant tumors globally, and it has become a challenging health issue in developing low-to-middle income countries. The prognosis oral remains poor because over 50% patients are recognized at advanced stages. Earlier detection screening models for mainly based on experts' knowledge, necessitates an automated tool detection. recent developments computational intelligence (CI) computer vision-based approaches help to accomplish enhanced performance medical-image-related tasks. This article develops intelligent deep learning enabled squamous cell carcinoma classification (IDL-OSCDC) technique using biomedical images. presented IDL-OSCDC model involves recognition proposed employs Gabor filtering (GF) as preprocessing step eliminate noise content. In addition, NasNet exploited generation high-level features from input Moreover, grasshopper optimization algorithm (EGOA)-based belief network (DBN) employed classification. hyperparameter tuning DBN performed EGOA which turn boosts outcomes. experimentation outcomes benchmark imaging dataset highlighted its promising other methods with maximum accu y , prec n reca l F score 95%, 96.15%, 93.75%, 94.67% correspondingly.

Language: Английский

---

MLP-Like Model With Convolution Complex Transformation for Auxiliary Diagnosis Through Medical Images

IEEE Journal of Biomedical and Health Informatics, Journal Year: 2023, Volume and Issue: 27(9), P. 4385 - 4396

Published: July 19, 2023

Medical images such as facial and tongue have been widely used for intelligence-assisted diagnosis, which can be regarded the multi-label classification task disease location (DL) nature (DN) of biomedical images. Compared with complicated convolutional neural networks Transformers this task, recent MLP-like architectures are not only simple less computationally expensive, but also stronger generalization capabilities. However, models require better input features from image. Thus, study proposes a novel convolution complex transformation (CCT-MLP) model DL DN recognition Notably, Tokenizer multiple layers first to extract shallow make up loss spatial information obtained by MLP structure. Subsequently, Channel-MLP architecture transformations is deep-level contextual features. In way, multi-channel extracted mixed perform Experimental results on our constructed image datasets demonstrate that method outperforms existing methods in terms both accuracy (Acc) mean average precision (mAP).

Language: Английский

---

Energy Efficient Path Planning Scheme for Unmanned Aerial Vehicle Using Hybrid Generic Algorithm-Based Q-Learning Optimization

IEEE Access, Journal Year: 2023, Volume and Issue: 12, P. 13400 - 13417

Published: Dec. 19, 2023

Efficient path planning optimization strategies are required to maximize flying time while consuming the least energy. This research offers a novel approach for energy-efficient Unmanned Aerial Vehicles (UAVs) that combines hybrid evolutionary algorithm and Q-learning accounting UAV's velocity distance from obstacles. To overcome constraints of traditional approaches, methodology genetic algorithms Q-learning. The suggested optimizes path-planning decisions based on real-time information by considering Genetic Algorithm (GA) creates wide collection candidate pathways. In contrast, uses reinforcement learning make educated selections present proximity static integration allows UAV modify its dynamically energy requirements environmental constraints. main goal is develop scheme capable dealing with obstacle-filled environments improve efficiency collision avoidance during flight missions. Our experimental results show technique outperforms classical GA method in terms significantly reducing consumption maintaining suitable rate best cost desired locations. analysis performance GA/QL more than 57.14% compared GA.

Language: Английский

---

Task Reverse Offloading with Deep Reinforcement Learning in Multi-Access Edge Computing

Published: Aug. 15, 2023

The Multi-access Edge Computing (MEC) technology's quick development greatly benefits the Collaborative Mobile Infrastructure System (CMIS). To combine data and produce tasks, crowd-sensing will be transferred to MEC server in CMIS. Nevertheless, if there are too many devices, it becomes extremely difficult for decide appropriately based on from devices infrastructure. This study builds a framework reverse offloading that carefully balances relationship between task completion time user mobile energy consumption. Moreover, decrease system use generally, an adaptive optimal method Deep Q-Network is created (DQN). results of simulations demonstrate suggested approach may successfully minimize consumption work latency when compared full local fixed techniques.

Language: Английский

---

Enhancing Medical Services Through Machine Learning and UAV Technology

Advances in computational intelligence and robotics book series, Journal Year: 2024, Volume and Issue: unknown, P. 307 - 343

Published: Jan. 17, 2024

This chapter focuses on the enhancement of medical services through integration unmanned aerial vehicle (UAV) technology and machine learning algorithms. It explores broad spectrum applications benefits that arise from combining these two technologies. By employing UAVs for automated delivery, supplies can be efficiently transported to remote or inaccessible regions, thereby improving access vital items. Remote patient monitoring, facilitated learning, enables real-time data collection analysis, enabling early identification health issues. equipped with equipment capabilities enhance emergency response by providing immediate assistance during critical situations. Disease surveillance outbreak management benefit use machine-learning algorithms identify disease hotspots predict spread illnesses.

Language: Английский

---

Green Machine Learning Approach for QoS Improvement in Cellular Communications

2022 IEEE 2nd International Maghreb Meeting of the Conference on Sciences and Techniques of Automatic Control and Computer Engineering (MI-STA), Journal Year: 2022, Volume and Issue: unknown, P. 523 - 528

Published: May 23, 2022

Green cellular communications are becoming an important approach due to large-scale and complex radio networks. Due the dynamic network behaviors related interference distribution, traffic bottlenecks, congestion points, hotspots, there is a need evaluate processes in systems addition ensuring spectrum availability. The delay, loss rate, SNR most issues that may affect communication performance. Artificial intelligent algorithms such as machine learning (ML) enable detection of dynamics networks by analyzing evaluating links qualities. It enables extraction knowledge from autonomously. extracted information helps know about every change wireless parameters, frequency, modulation, route selection, etc. This paper provides details use ML green efficiently upgrade enhances different approaches including quality services (QoS), signal load, energy efficiency, which critical paradigms. also presents technical concept solve significant problems communications, future aspects considerations for consumption minimization using communications.

Language: Английский

---

Deep Learning Approaches for Object Detection in Autonomous Driving: Smart Cities Perspective

Published: Jan. 1, 2025

Language: Английский

---

Automatic colorectal cancer detection using machine learning and deep learning based on feature selection in histopathological images

Biomedical Signal Processing and Control, Journal Year: 2025, Volume and Issue: 107, P. 107866 - 107866

Published: March 27, 2025

Language: Английский

---

EPDTNet + -EM: Advanced Transfer Learning and SubNet Architecture for Medical Image Diagnosis

Cognitive Computation, Journal Year: 2025, Volume and Issue: 17(2)

Published: April 1, 2025

Language: Английский

---

Attacks Detection in 6G Wireless Networks using Machine Learning

Published: Aug. 15, 2023

Unlike the fifth generation (5G), which is well recognized for network cloudification with micro-service-oriented design, sixth (6G) of networks directly tied to intelligent orchestration and management. The Attacks Detection in 6G (AD6Gs) wireless created by this research uses a Machine Learning (ML) algorithm. pre-processing stage ML-AD6Gs process initial step. second involves feature selection approach. Correlation Feature Selection algorithm (CFS) used implement suggested hybrid strategy. It selects best subset reduces dimensionality each independent analyses dataset CICDDOS2019. voting average method as an aggregation step, two classifiers—Random Forest (RF) Support Vector (SVM)—are modified be ML Algorithms. proposed shown outperformed existing classification method. accuracy was 99.9%% CICDDOS2019 false alarm rate 0.00102

Language: Английский

---