DualXBrNet: A Hybrid Dual-Path Cross-Level Fusion Framework for Multiclass Brain Tumor Classification Using MRI Images
- 1Department of Computer Engineering, College of Engineering, University of Basrah, Basrah 61004, Iraq
Abstract
Brain tumor classification from magnetic resonance imaging (MRI) can support computer-aided diagnosis by enabling automated differentiation of tumor and non-tumor categories. This study presents DualXBrNet, a hybrid dual-level fusion framework for multiclass brain tumor classification using the original BRISC2025 dataset. Three pretrained convolutional neural networks (CNNs), ResNet50V2, MobileNetV2, and EfficientNetB0, were optimized to learn complementary representations of glioma, meningioma, pituitary tumor, and no-tumor MRI images. Their deep features were integrated through a learnable Feature Fusion network, while the class probability outputs of the three CNNs and the Feature Fusion model were subsequently combined using a leakage-aware five-fold out-of-fold (OOF) decision-level stacking strategy. A low-complexity Logistic Regression meta-classifier was trained exclusively on the resulting 16-dimensional OOF probability representations to prevent information leakage during meta-learning. On the independent official test set, DualXBrNet achieved 96.80% accuracy and a macro-average ROC-AUC of 99.77%. Oracle, statistical, computational complexity, feature space, and Occlusion Sensitivity analyses were further conducted to assess model complementarity, computational efficiency, representation quality, and interpretability. The results demonstrate that feature-level fusion provides the primary performance improvement while OOF-based decision stacking offers complementary refinement within a leakage-aware evaluation protocol. Overall, the proposed framework emphasizes methodological rigor through leakage-aware meta-learning, independent testing, and comprehensive model analysis, rather than relying on classification accuracy alone.
Related
- ReViT-RF: A hybrid ResNet-50–Vision Transformer with a Radiomic Feature Fusion framework for lung tumor classification and TNM-based stage prediction using CT images
- Boundary-aware multi-scale residual learning with hybrid radiomic–deep fusion for lung tumor segmentation and classification
- HayabusaNet: Hybrid attention-based multiscale fusion CNN for accurate and efficient brain tumor classification in MRI scans
- A segmentation-guided CNN–Vision transformer feature fusion framework for multi-class breast ultrasound image classification
- Hybrid deep learning–driven explainable AI framework for fault detection and classification in smart power grids
BibTeX
@article{alazzawi2026dualxbrnet,
title = {{DualXBrNet: A Hybrid Dual-Path Cross-Level Fusion Framework for Multiclass Brain Tumor Classification Using MRI Images}},
author = {Sarah Al-Azzawi and Ghaida A. Al-Suhail},
journal = {BioMedInformatics},
year = {2026},
month = oct,
doi = {10.3390/biomedinformatics6050092},
url = {https://doi.org/10.3390/biomedinformatics6050092},
note = {CC BY},
}
RIS
TY - JOUR
TI - DualXBrNet: A Hybrid Dual-Path Cross-Level Fusion Framework for Multiclass Brain Tumor Classification Using MRI Images
AU - Sarah Al-Azzawi
AU - Ghaida A. Al-Suhail
PY - 2026
DA - 2026/10/26
JO - BioMedInformatics
DO - 10.3390/biomedinformatics6050092
UR - https://doi.org/10.3390/biomedinformatics6050092
AB - Brain tumor classification from magnetic resonance imaging (MRI) can support computer-aided diagnosis by enabling automated differentiation of tumor and non-tumor categories. This study presents DualXBrNet, a hybrid dual-level fusion framework for multiclass brain tumor classification using the original BRISC2025 dataset. Three pretrained convolutional neural networks (CNNs), ResNet50V2, MobileNetV2, and EfficientNetB0, were optimized to learn complementary representations of glioma, meningioma, pituitary tumor, and no-tumor MRI images. Their deep features were integrated through a learnable Feature Fusion network, while the class probability outputs of the three CNNs and the Feature Fusion model were subsequently combined using a leakage-aware five-fold out-of-fold (OOF) decision-level stacking strategy. A low-complexity Logistic Regression meta-classifier was trained exclusively on the resulting 16-dimensional OOF probability representations to prevent information leakage during meta-learning. On the independent official test set, DualXBrNet achieved 96.80% accuracy and a macro-average ROC-AUC of 99.77%. Oracle, statistical, computational complexity, feature space, and Occlusion Sensitivity analyses were further conducted to assess model complementarity, computational efficiency, representation quality, and interpretability. The results demonstrate that feature-level fusion provides the primary performance improvement while OOF-based decision stacking offers complementary refinement within a leakage-aware evaluation protocol. Overall, the proposed framework emphasizes methodological rigor through leakage-aware meta-learning, independent testing, and comprehensive model analysis, rather than relying on classification accuracy alone.
SN - 2673-7426
ER -