Brain Tumor Detection and Classification Using Deep Learning and Sine-Cosine Fitness Grey Wolf Optimization

Bioengineering, Journal Year: 2022, Volume and Issue: 10(1), P. 18 - 18

Published: Dec. 22, 2022

Diagnosing a brain tumor takes long time and relies heavily on the radiologist's abilities experience. The amount of data that must be handled has increased dramatically as number patients increased, making old procedures both costly ineffective. Many researchers investigated variety algorithms for detecting classifying tumors were accurate fast. Deep Learning (DL) approaches have recently been popular in developing automated systems capable accurately diagnosing or segmenting less time. DL enables pre-trained Convolutional Neural Network (CNN) model medical images, specifically cancers. proposed Brain Tumor Classification Model based CNN (BCM-CNN) is hyperparameters optimization using an adaptive dynamic sine-cosine fitness grey wolf optimizer (ADSCFGWO) algorithm. There followed by training built with Inception-ResnetV2. employs commonly used models (Inception-ResnetV2) to improve diagnosis, its output binary 0 1 (0: Normal, 1: Tumor). are primarily two types hyperparameters: (i) determine underlying network structure; (ii) hyperparameter responsible network. ADSCFGWO algorithm draws from sine cosine adaptable framework uses algorithms' strengths. experimental results show BCM-CNN classifier achieved best due enhancement CNN's performance optimization's hyperparameters. 99.98% accuracy BRaTS 2021 Task dataset.

Language: Английский

---

Classification of Monkeypox Images Based on Transfer Learning and the Al-Biruni Earth Radius Optimization Algorithm

Mathematics, Journal Year: 2022, Volume and Issue: 10(19), P. 3614 - 3614

Published: Oct. 2, 2022

The world is still trying to recover from the devastation caused by wide spread of COVID-19, and now monkeypox virus threatens becoming a worldwide pandemic. Although not as lethal or infectious numerous countries report new cases daily. Thus, it surprising that necessary precautions have been taken, will be if another pandemic occurs. Machine learning has recently shown tremendous promise in image-based diagnosis, including cancer detection, tumor cell identification, COVID-19 patient detection. Therefore, similar application may implemented diagnose invades human skin. An image can acquired utilized further condition. In this paper, two algorithms are proposed for improving classification accuracy images. based on transfer feature extraction meta-heuristic optimization selection parameters multi-layer neural network. GoogleNet deep network adopted extraction, Al-Biruni Earth radius algorithm, sine cosine particle swarm algorithm. Based these algorithms, binary hybrid algorithm selection, along with optimizing To evaluate publicly available dataset employed. assessment was performed terms ten evaluation criteria. addition, set statistical tests conducted measure effectiveness, significance, robustness algorithms. results achieved confirm superiority effectiveness methods compared other methods. average 98.8%.

Language: Английский

---

Waterwheel Plant Algorithm: A Novel Metaheuristic Optimization Method

Processes, Journal Year: 2023, Volume and Issue: 11(5), P. 1502 - 1502

Published: May 15, 2023

Attempting to address optimization problems in various scientific disciplines is a fundamental and significant difficulty requiring optimization. This study presents the waterwheel plant technique (WWPA), novel stochastic motivated by natural systems. The proposed WWPA’s basic concept based on modeling plant’s behavior while hunting expedition. To find prey, WWPA uses plants as search agents. We present mathematical model for use addressing problems. Twenty-three objective functions of varying unimodal multimodal types were used assess performance. results optimizing demonstrate strong exploitation ability get close optimal solution, show exploration zero major region space. Three engineering design also gauge potential improving practical programs. effectiveness was evaluated comparing its with those seven widely metaheuristic algorithms. When compared eight competing algorithms, simulation analyses that outperformed them finding more proportionate balance between exploitation.

Language: Английский

---

Meta-Heuristic Optimization of LSTM-Based Deep Network for Boosting the Prediction of Monkeypox Cases

Mathematics, Journal Year: 2022, Volume and Issue: 10(20), P. 3845 - 3845

Published: Oct. 17, 2022

Recent technologies such as artificial intelligence, machine learning, and big data are essential for supporting healthcare monitoring systems, particularly Monkeypox confirmed cases. Infected uninfected cases around the world have contributed to a growing dataset, which is publicly available can be used by intelligence learning predict of at an early stage. Motivated this, we propose in this paper new approach accurate prediction based on optimized Long Short-Term Memory (LSTM) deep network. To fine-tune hyper-parameters LSTM-based network, employed Al-Biruni Earth Radius (BER) optimization algorithm; thus, proposed denoted BER-LSTM. Experimental results show effectiveness when assessed using various evaluation criteria, Mean Bias Error, recorded (0.06) prove superiority approach, six different models included conducted experiments. In addition, four algorithms considered comparison purposes. The approach. On other hand, several statistical tests applied analyze stability significance These include one-way Analysis Variance (ANOVA), Wilcoxon, regression tests. these emphasize robustness, significance, efficiency

Language: Английский

---

An Al-Biruni Earth Radius Optimization-Based Deep Convolutional Neural Network for Classifying Monkeypox Disease

Diagnostics, Journal Year: 2022, Volume and Issue: 12(11), P. 2892 - 2892

Published: Nov. 21, 2022

Human skin diseases have become increasingly prevalent in recent decades, with millions of individuals developed countries experiencing monkeypox. Such conditions often carry less obvious but no devastating risks, including increased vulnerability to monkeypox, cancer, and low self-esteem. Due the visual resolution monkeypox disease images, medical specialists high-level tools are typically required for a proper diagnosis. The manual diagnosis is subjective, time-consuming, labor-intensive. Therefore, it necessary create computer-aided approach automated disease. Most research articles on relied convolutional neural networks (CNNs) using classical loss functions, allowing them pick up discriminative elements images. To enhance this, novel framework Al-Biruni Earth radius (BER) optimization-based stochastic fractal search (BERSFS) proposed fine-tune deep CNN layers classifying from As first step approach, we use CNN-based models learn embedding input images Euclidean space. In second step, an optimized classification model based triplet function calculate distance between pairs space features that may be used distinguish different cases, cases. uses human obtained African hospital. experimental results study demonstrate framework’s efficacy, as outperforms numerous examples prior problems. On other hand, statistical experiments Wilcoxon analysis variance (ANOVA) tests conducted evaluate terms effectiveness stability. recorded confirm superiority method when compared optimization algorithms machine learning models.

Language: Английский

---

Advanced Meta-Heuristic Algorithm Based on Particle Swarm and Al-Biruni Earth Radius Optimization Methods for Oral Cancer Detection

IEEE Access, Journal Year: 2023, Volume and Issue: 11, P. 23681 - 23700

Published: Jan. 1, 2023

Oral cancer is a deadly form of cancerous tumor that widely spread in low and middle-income countries. An early affordable oral diagnosis might be achieved by automating the detection precancerous malignant lesions mouth. There are many research attempts to develop robust machine-learning model can detect from images. However, these still lacking high precision detection. Therefore, this work aims propose new approach capable detecting medical images with higher accuracy. In work, novel based on convolutional neural network (CNN) optimized deep belief (DBN). The design parameters CNN DBN using optimization algorithm, which developed as hybrid Particle Swarm Optimization (PSO) Al-Biruni Earth Radius (BER) algorithms denoted (PSOBER). Using standard biomedical dataset available Kaggle repository, proposed shows promising results outperforming various competing approaches an accuracy 97.35%. addition, set statistical tests, such One-way analysis-of-variance (ANOVA) Wilcoxon signed-rank conducted prove significance stability approach. methodology solid efficient, specialists adopt it. additional larger scale required confirm findings highlight other features utilized for

Language: Английский

---

Innovative Feature Selection Method Based on Hybrid Sine Cosine and Dipper Throated Optimization Algorithms

IEEE Access, Journal Year: 2023, Volume and Issue: 11, P. 79750 - 79776

Published: Jan. 1, 2023

Introduction: In pattern recognition and data mining, feature selection is one of the most crucial tasks. To increase efficacy classification algorithms, it necessary to identify relevant subset features in a given domain. This means that challenge can be seen as an optimization problem, thus meta-heuristic techniques utilized find solution. Methodology: this work, we propose novel hybrid binary algorithm solve problem by combining two algorithms: Dipper Throated Optimization (DTO) Sine Cosine (SC) algorithm. The new referred bSCWDTO. We employed sine cosine improve exploration process ensure converges quickly accurately. Thirty datasets from University California Irvine (UCI) machine learning repository are used evaluate robustness stability proposed bSCWDTO addition, K-Nearest Neighbor (KNN) classifier measure selected features' effectiveness problems. Results: achieved results demonstrate algorithm's superiority over ten state-of-the-art methods, including original DTO SC, Particle Swarm (PSO), Whale Algorithm (WOA), Grey Wolf (GWO), Multiverse (MVO), Satin Bowerbird Optimizer (SBO), Genetic (GA), GWO GA, Firefly (FA). Moreover, Wilcoxon's rank-sum test was performed at 0.05 significance level study statistical difference between method alternative methods. Conclusion: These emphasized method's significance, superiority, difference.

Language: Английский

---

A Binary Waterwheel Plant Optimization Algorithm for Feature Selection

IEEE Access, Journal Year: 2023, Volume and Issue: 11, P. 94227 - 94251

Published: Jan. 1, 2023

The vast majority of today's data is collected and stored in enormous databases with a wide range characteristics that have little to do the overarching goal concept. Feature selection process choosing best features for classification problem, which improves classification's accuracy. considered multi-objective optimization problem two objectives: boosting accuracy while decreasing feature count. To efficiently handle process, we propose this paper novel algorithm inspired by behavior waterwheel plants when hunting their prey how they update locations throughout exploration exploitation processes. proposed referred as binary plant (bWWPA). In particular approach, search space well technique's mapping from continuous discrete spaces are both represented new model. Specifically, fitness cost functions factored into algorithm's evaluation modeled mathematically. assess performance algorithm, set extensive experiments were conducted evaluated terms 30 benchmark datasets include low, medium, high dimensional features. comparison other recent algorithms, experimental findings demonstrate bWWPAperforms better than competing algorithms. addition, statistical analysis performed one-way analysis-of-variance (ANOVA) Wilcoxon signed-rank tests examine differences between compared These experiments' results confirmed superiority effectiveness handling process.

Language: Английский

---

A Novel Bio-Inspired Optimization Algorithm Design for Wind Power Engineering Applications Time-Series Forecasting

Biomimetics, Journal Year: 2023, Volume and Issue: 8(3), P. 321 - 321

Published: July 20, 2023

Wind patterns can change due to climate change, causing more storms, hurricanes, and quiet spells. These changes dramatically affect wind power system performance predictability. Researchers practitioners are creating advanced forecasting algorithms that combine parameters data sources. Advanced numerical weather prediction models, machine learning techniques, real-time meteorological sensor satellite used. This paper proposes a Recurrent Neural Network (RNN) model incorporating Dynamic Fitness Al-Biruni Earth Radius (DFBER) algorithm predict patterns. The of this is compared with several other popular including BER, Jaya Algorithm (JAYA), Fire Hawk Optimizer (FHO), Whale Optimization (WOA), Grey Wolf (GWO), Particle Swarm (PSO)-based models. evaluation done using various metrics such as relative root mean squared error (RRMSE), Nash Sutcliffe Efficiency (NSE), absolute (MAE), bias (MBE), Pearson’s correlation coefficient (r), determination (R2), agreement (WI). According the analysis presented in study, proposed RNN-DFBER-based outperforms models considered. suggests RNN model, combined DFBER algorithm, predicts effectively than alternative To support findings, visualizations provided demonstrate effectiveness RNN-DFBER model. Additionally, statistical analyses, ANOVA test Wilcoxon Signed-Rank test, conducted assess significance reliability results.

Language: Английский

---

Brain Tumor Segmentation Using Deep Learning on MRI Images

Diagnostics, Journal Year: 2023, Volume and Issue: 13(9), P. 1562 - 1562

Published: April 27, 2023

Brain tumor (BT) diagnosis is a lengthy process, and great skill expertise are required from radiologists. As the number of patients has expanded, so amount data to be processed, making previous techniques both costly ineffective. Many academics have examined range reliable quick for identifying categorizing BTs. Recently, deep learning (DL) methods gained popularity creating computer algorithms that can quickly reliably diagnose or segment To identify BTs in medical images, DL permits pre-trained convolutional neural network (CNN) model. The suggested magnetic resonance imaging (MRI) images included BT segmentation dataset, which was created as benchmark developing evaluating diagnosis. There 335 annotated MRI collection. For purpose testing algorithms, brain (BraTS) dataset produced. A CNN also utilized model-building process segmenting using BraTS dataset. train model, categorical cross-entropy loss function an optimizer, such Adam, were employed. Finally, model's output successfully identified segmented attaining validation accuracy 98%.

Language: Английский

---