SAlexNet: Superimposed AlexNet using Residual Attention Mechanism for Accurate and Efficient Automatic Primary Brain Tumor Detection and Classification

Results in Engineering, Journal Year: 2025, Volume and Issue: unknown, P. 104025 - 104025

Published: Jan. 1, 2025

Language: Английский

---

A comprehensive systematic literature review of ML in nanotechnology for sustainable development

Nanotechnology Reviews, Journal Year: 2024, Volume and Issue: 13(1)

Published: Jan. 1, 2024

Abstract The rapid expansion of nanotechnology has transformed numerous sectors, with nanoproducts now ubiquitous in everyday life, electronics, healthcare, and pharmaceuticals. Despite their widespread adoption, concerns persist regarding potential adverse effects, necessitating vigilant risk management. This systematic literature review advocates for leveraging artificial intelligence (AI) machine learning (ML) methodologies to enhance simulations refine safety assessments nanomaterials (NMs). Through a comprehensive examination the existing literature, this study seeks explain pivotal role AI boosting NMs sustainability efforts across six key research themes. It explores significance advancing sustainability, hazard identification, diverse applications field. In addition, it evaluates past strategies while proposing innovative avenues future exploration. By conducting analysis, aims illuminate current landscape, identify challenges, outline pathways integrating ML promote sustainable practices within nanotechnology. Furthermore, extending these technologies monitor real-world behaviour delivery. its thorough investigation, endeavours address obstacles pave way safe utilization nanotechnology, thereby minimizing associated risks.

Language: Английский

---

Convolutional Neural Network–Machine Learning Model: Hybrid Model for Meningioma Tumour and Healthy Brain Classification

Journal of Imaging, Journal Year: 2024, Volume and Issue: 10(9), P. 235 - 235

Published: Sept. 20, 2024

This paper presents a hybrid study of convolutional neural networks (CNNs), machine learning (ML), and transfer (TL) in the context brain magnetic resonance imaging (MRI). The anatomy is very complex; inside skull, tumour can form any part. With MRI technology, cross-sectional images are generated, radiologists detect abnormalities. When size small, it undetectable to human visual system, necessitating alternative analysis using AI tools. As widely known, CNNs explore structure an image provide features on SoftMax fully connected (SFC) layer, classification items that belong input classes established. Two comparison studies for meningioma tumours healthy brains presented this paper: (i) classifying original CNN two pre-trained CNNs, DenseNet169 EfficientNetV2B0; (ii) determining which ML combination yields most accurate when replaced with three models; context, Random Forest (RF), K-Nearest Neighbors (KNN), Support Vector Machine (SVM) were proposed. In binary brains, EfficientNetB0-SVM shows accuracy 99.5% test dataset. A generalisation results was performed, overfitting prevented by bagging ensemble method.

Language: Английский

---

Innovative deep learning and quantum entropy techniques for brain tumor MRI image edge detection and classification model

Alexandria Engineering Journal, Journal Year: 2025, Volume and Issue: 122, P. 588 - 604

Published: March 20, 2025

Language: Английский

---

Improving healthcare sustainability using advanced brain simulations using a multi-modal deep learning strategy with VGG19 and bidirectional LSTM

Frontiers in Medicine, Journal Year: 2025, Volume and Issue: 12

Published: April 10, 2025

Brain tumor categorization on MRI is a challenging but crucial task in medical imaging, requiring high resilience and accuracy for effective diagnostic applications. This study describe unique multimodal scheme combining the capabilities of deep learning with ensemble approaches to overcome these issues. The system integrates three new modalities, spatial feature extraction using pre-trained VGG19 network, sequential dependency Bidirectional LSTM, classification efficiency through LightGBM classifier. combination both methods leverages complementary strengths convolutional neural networks recurrent networks, thus enabling model achieve state-of-the-art performance scores. outcomes confirm efficacy this approach, which achieves total 97%, an F1-score 0.97, ROC AUC score 0.997. With synergistic harnessing features, enhances rates effectively deals high-dimensional data, compared traditional single-modal methods. scalable methodology has possibility greatly augmenting brain diagnosis planning treatment imaging studies.

Language: Английский

---

An XAI-Enhanced EfficientNetB0 Framework for Precision Brain Tumor Detection in MRI Imaging

Journal of Neuroscience Methods, Journal Year: 2024, Volume and Issue: 410, P. 110227 - 110227

Published: July 20, 2024

Language: Английский

---