Brain tumor classification from MRI scans: a framework of hybrid deep learning model with Bayesian optimization and quantum theory-based marine predator algorithm

Frontiers in Oncology, Journal Year: 2024, Volume and Issue: 14

Published: Feb. 8, 2024

Brain tumor classification is one of the most difficult tasks for clinical diagnosis and treatment in medical image analysis. Any errors that occur throughout brain process may result a shorter human life span. Nevertheless, currently used techniques ignore certain features have particular significance relevance to problem favor extracting choosing deep features. One important area research learning-based categorization tumors using magnetic resonance imaging (MRI). This paper proposes an automated learning model optimal information fusion framework classifying from MRI images. The dataset this work was imbalanced, key challenge training selected networks. imbalance impacts performance models because it causes classifier become biased majority class. We designed sparse autoencoder network generate new images resolve imbalance. After that, two pretrained neural networks were modified hyperparameters initialized Bayesian optimization, which later utilized process. extracted global average pooling layer. contain few irrelevant information; therefore, we proposed improved Quantum Theory-based Marine Predator Optimization algorithm (QTbMPA). QTbMPA selects both networks’ best finally fuses serial-based approach. fused feature set passed classifiers final classification. tested on augmented Figshare accuracy 99.80%, sensitivity rate 99.83%, false negative 17%, precision 99.83% obtained. Comparison ablation study show improvement work.

Language: Английский

---

Hybrid deep learning approach for brain tumor classification using EfficientNetB0 and novel quantum genetic algorithm

PeerJ Computer Science, Journal Year: 2025, Volume and Issue: 11, P. e2556 - e2556

Published: Jan. 21, 2025

One of the most complex and life-threatening pathologies central nervous system is brain tumors. Correct diagnosis these tumors plays an important role in determining treatment plans patients. Traditional classification methods often rely on manual assessments, which can be prone to error. Therefore, multiple has gained significant interest recent years both medical computer science fields. The use artificial intelligence machine learning, especially automatic tumors, increasing significantly. Deep learning models achieve high accuracy when trained datasets classification. This study examined deep learning-based approaches for multi-class a new approach combining quantum genetic algorithms (QGA) was proposed. powerful feature extraction ability pre-trained EfficientNetB0 utilized combined with this algorithms, It aimed develop selection method. With hybrid method, reliability tumor achieved. proposed model achieved 98.36% 98.25%, respectively, different data sets significantly outperformed traditional methods. As result, method offers robust scalable solution that will help classify early accurate contribute field imaging patient outcomes.

Language: Английский

---

A multi-modality framework for precise brain tumor detection and multi-class classification using hybrid GAN approach

Biomedical Signal Processing and Control, Journal Year: 2025, Volume and Issue: 104, P. 107559 - 107559

Published: Feb. 11, 2025

Language: Английский

---

Brain tumor detection with integrating traditional and computational intelligence approaches across diverse imaging modalities - Challenges and future directions

Computers in Biology and Medicine, Journal Year: 2024, Volume and Issue: 175, P. 108412 - 108412

Published: April 16, 2024

Language: Английский

---

Artificial Intelligence-Based Algorithms in Medical Image Scan Segmentation and Intelligent Visual Content Generation—A Concise Overview

Electronics, Journal Year: 2024, Volume and Issue: 13(4), P. 746 - 746

Published: Feb. 13, 2024

Recently, artificial intelligence (AI)-based algorithms have revolutionized the medical image segmentation processes. Thus, precise of organs and their lesions may contribute to an efficient diagnostics process a more effective selection targeted therapies, as well increasing effectiveness training process. In this context, AI automatization scan increase quality resulting 3D objects, which lead generation realistic virtual objects. paper, we focus on AI-based solutions applied in intelligent visual content generation, i.e., computer-generated three-dimensional (3D) images context extended reality (XR). We consider different types neural networks used with special emphasis learning rules applied, taking into account algorithm accuracy performance, open data availability. This paper attempts summarize current development methods imaging that are XR. It concludes possible developments challenges applications reality-based solutions. Finally, future lines research directions applications, both solutions, discussed.

Language: Английский

---

Modified U-Net with attention gate for enhanced automated brain tumor segmentation

Neural Computing and Applications, Journal Year: 2025, Volume and Issue: unknown

Published: Jan. 2, 2025

Language: Английский

---

Coati optimization algorithm for brain tumor identification based on MRI with utilizing phase-aware composite deep neural network

Electromagnetic Biology and Medicine, Journal Year: 2025, Volume and Issue: unknown, P. 1 - 18

Published: Jan. 21, 2025

Brain tumors can cause difficulties in normal brain function and are capable of developing various regions the brain. Malignant tumours develop quickly, pass through neighboring tissues, extend to further or central nervous system. In contrast, healthy typically slowly do not invade surrounding tissues. Individuals frequently struggle with sensory abnormalities, motor deficiencies affecting coordination, cognitive impairments memory focus. this research, Utilizing Phase-aware Composite Deep Neural Network Optimized Coati Algorithm for Tumor Identification Based on Magnetic resonance imaging (PACDNN-COA-BTI-MRI) is proposed. First, input images taken from tumour Dataset. To execute this, image pre-processed using Multivariate Fast Iterative Filtering (MFIF) it reduces occurrence over-fitting collected dataset; then feature extraction Self-Supervised Nonlinear Transform (SSNT) extract essential features like model, shape, intensity. Then, proposed PACDNN-COA-BTI-MRI implemented Matlab performance metrics Recall, Accuracy, F1-Score, Precision Specificity ROC analysed. Performance approach attains 16.7%, 20.6% 30.5% higher accuracy; 19.9%, 22.2% 30.1% recall 21.9% 30.8% precision when analysed existing techniques tumor identification MRI-Based Learning Approach Efficient Classification (MRI-DLA-ECBT), Detection Convolutional Methods Chosen Machine Techniques (MRI-BTD-CDMLT) Image CNN-Based Method (MRI-BTID-CNN) methods, respectively.

Language: Английский

---

Enhancing brain tumor classification in MRI scans with a multi-layer customized convolutional neural network approach

Frontiers in Computational Neuroscience, Journal Year: 2024, Volume and Issue: 18

Published: June 12, 2024

The necessity of prompt and accurate brain tumor diagnosis is unquestionable for optimizing treatment strategies patient prognoses. Traditional reliance on Magnetic Resonance Imaging (MRI) analysis, contingent upon expert interpretation, grapples with challenges such as time-intensive processes susceptibility to human error.

Language: Английский

---

Brain Tumor Segmentation Using Ensemble CNN-Transfer Learning Models: DeepLabV3plus and ResNet50 Approach

Lecture notes in computer science, Journal Year: 2024, Volume and Issue: unknown, P. 340 - 354

Published: Jan. 1, 2024

Language: Английский

---

MRI intracranial Neoplasm classification using hybrid LOA-based deep learning classifier

Biomedical Signal Processing and Control, Journal Year: 2025, Volume and Issue: 104, P. 107560 - 107560

Published: Jan. 28, 2025

Language: Английский

---