DB-FGA-Net: Dual Backbone Frequency Gated Attention Network for Multi-Class Brain Tumor Classification with Grad-CAM Interpretability

TL;DR

DB-FGA-Net introduces a dual-backbone brain tumor classifier that fuses VGG16 and Xception with a Frequency-Gated Attention (FGA) module to capture complementary local and global features from MRI. The FGA block combines channel-spatial co-attention with frequency-domain cues and dynamic gating, delivering augmentation-free state-of-the-art performance on the 7K-DS dataset and strong cross-dataset generalization to 3K-DS, all while providing Grad-CAM interpretability. The approach is validated across 4-, 3-, and 2-class settings and supported by a GUI for real-time classification and tumor localization, enhancing clinical trust and usability. Overall, DB-FGA-Net demonstrates robust, interpretable, and deployment-friendly performance for multi-class brain tumor diagnosis in MRI data.

Abstract

Brain tumors are a challenging problem in neuro-oncology, where early and precise diagnosis is important for successful treatment. Deep learning-based brain tumor classification methods often rely on heavy data augmentation which can limit generalization and trust in clinical applications. In this paper, we propose a double-backbone network integrating VGG16 and Xception with a Frequency-Gated Attention (FGA) Block to capture complementary local and global features. Unlike previous studies, our model achieves state-of-the-art performance without augmentation which demonstrates robustness to variably sized and distributed datasets. For further transparency, Grad-CAM is integrated to visualize the tumor regions based on which the model is giving prediction, bridging the gap between model prediction and clinical interpretability. The proposed framework achieves 99.24\% accuracy on the 7K-DS dataset for the 4-class setting, along with 98.68\% and 99.85\% in the 3-class and 2-class settings, respectively. On the independent 3K-DS dataset, the model generalizes with 95.77\% accuracy, outperforming baseline and state-of-the-art methods. To further support clinical usability, we developed a graphical user interface (GUI) that provides real-time classification and Grad-CAM-based tumor localization. These findings suggest that augmentation-free, interpretable, and deployable deep learning models such as DB-FGA-Net hold strong potential for reliable clinical translation in brain tumor diagnosis.

Figures (14)

• Figure 1: Sample MRI images from the 7K-DS dataset (Glioma, Meningioma, No Tumor and Pituitary).
• Figure 2: Sample MRI images from the 3K-DS dataset (Glioma, Meningioma, No Tumor and Pituitary).
• Figure 3: Visual representation of the Workflow of the proposed approach.
• Figure 4: Baseline architecture: input images are processed through a CNN backbone, followed by Global Average Pooling, Dropout, and a Softmax classifier.
• Figure 5: Detailed structure of the FGA block, illustrating the sequential channel and spatial attention mechanisms applied to enhance feature maps in the CNN architecture.