IRO Journals
Journal of Innovative Image Processing

A Dual-Backbone CNN Framework with Gated Cross-Attention Fusion and Mixup Focal Loss for Robust and Explainable Classification of Brain Tumors

Open Access
Published: 22 May, 2026
Pages: 702-723
  • Cmak Zeelan Basha , Cmak Zeelan Basha
    Department of Computer Science and Engineering, Koneru Lakshmaiah Education Foundation, Vaddeswaram, Guntur, Andhra Pradesh, India
    Department of Computer Science and Engineering, Koneru Lakshmaiah Education Foundation, Vaddeswaram, Guntur, Andhra Pradesh, India
  • Prasanth Yalla Prasanth Yalla
    Department of Computer Science and Engineering, Koneru Lakshmaiah Education Foundation, Vaddeswaram, Guntur, Andhra Pradesh, India
    Department of Computer Science and Engineering, Koneru Lakshmaiah Education Foundation, Vaddeswaram, Guntur, Andhra Pradesh, India
view PDF
view PDF

How to Cite

Basha, Cmak Zeelan, and Prasanth Yalla. 2026. “A Dual-Backbone CNN Framework With Gated Cross-Attention Fusion and Mixup Focal Loss for Robust and Explainable Classification of Brain Tumors”. Journal of Innovative Image Processing 8 (2): 702-23. https://doi.org/10.36548/jiip.2026.2.014.
ACM ACS APA ABNT Chicago Harvard IEEE MLA Turabian Vancouver AMA

Download Citation

Endnote/Zotero/Mendeley (RIS) BibTeX

Keywords

Brain Tumor Classification
MRI
Dual-Backbone CNN
Cross-Attention
Focal Loss
Grad-CAM

Abstract

Diagnosing and classifying brain tumors based on MRI scans is a difficult and complicated process due to differences in tumor types, low contrast, and class imbalance.  In this study, we propose a framework that uses a dual-backbone convolutional neural network (CNN) featuring EfficientNetV2-S and ConvNeXt-Tiny for multi-scale feature extraction and classification. The gated cross-attention fusion module enables adaptive bidirectional interaction among the feature representations. In addition, Mixup regularization, focal loss, and label smoothing improve robustness, calibration, and class imbalance handling. The model employs a two-stage transfer learning framework, supported by exponential moving average (EMA)-based stabilization. The framework achieves 95.73% classification accuracy, a macro-F1 score of 95.7%, and a macro-average ROC–AUC of 0.9967 when evaluated on a multi-class MRI brain tumor dataset containing 7,023 images. The proposed fusion and training strategy demonstrate effectiveness through comparative and ablation analyses. Furthermore, Grad-CAM visualizations show that the model focuses on tumor-related regions. The proposed dual-backbone gated cross-attention framework for brain tumor classification demonstrates both high performance and good interpretability. The model also shows potential for clinical decision-support systems but requires additional multi-center validation.

References

  1. Louis, David N., Arie Perry, Pieter Wesseling, Daniel J. Brat, Ian A. Cree, Dominique Figarella-Branger, Cynthia Hawkins et al. "The 2021 WHO Classification of Tumors of the Central Nervous System: A Summary." Neuro-oncology 23, no. 8 (2021): 1231-1251.
  2. Menze, Bjoern H., Andras Jakab, Stefan Bauer, Jayashree Kalpathy-Cramer, Keyvan Farahani, Justin Kirby, Yuliya Burren et al. "The Multimodal Brain Tumor Image Segmentation Benchmark (BRATS)." IEEE transactions on medical imaging 34, no. 10 (2014): 1993-2024.
  3. Havaei, Mohammad, Axel Davy, David Warde-Farley, Antoine Biard, Aaron Courville, Yoshua Bengio, Chris Pal, Pierre-Marc Jodoin, and Hugo Larochelle. "Brain Tumor Segmentation with Deep Neural Networks." Medical image analysis 35 (2017): 18-31.
  4. He, Kaiming, Xiangyu Zhang, Shaoqing Ren, and Jian Sun. "Deep Residual Learning for Image Recognition." In Proceedings of the IEEE conference on computer vision and pattern recognition, 770-778. 2016.
  5. Litjens, Geert, Thijs Kooi, Babak Ehteshami Bejnordi, Arnaud Arindra Adiyoso Setio, Francesco Ciompi, Mohsen Ghafoorian, Jeroen Awm Van Der Laak, Bram Van Ginneken, and Clara I. Sánchez. "A Survey on Deep Learning in Medical Image Analysis." Medical image analysis 42 (2017): 60-88.
  6. Isty, Maherun Nessa, Raiyan Gani, Jubaer Ahmed, Tasmia Islam, and Shamim Ripon. "Deep Learning Techniques for Early Brain Tumor Detection: A Comparative Study on Models Performance Utilizing Dataset Enhancement." In 2024 IEEE International Conference on Computing, Applications and Systems (COMPAS), IEEE, 2024, 1-6.
  7. Lakineni, Prasanna Kumar, D. Jayanarayana Reddy, M. Chitra, R. Umapriya, L. Vadivel Kannan, and S. R. Barkunan. "Optimal Feature Selection and Classification Using Convolutional Neural Network-Based Plant Disease Prediction." In 2023 IEEE International Conference on Integrated Circuits and Communication Systems (ICICACS), IEEE, 2023, 1-6.
  8. Dosovitskiy, Alexey, Lucas Beyer, Alexander Kolesnikov, Dirk Weissenborn, Xiaohua Zhai, Thomas Unterthiner, Mostafa Dehghani et al. "An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale." arXiv preprint arXiv:2010.11929 (2020).
  9. Zhang, Hongyi, Moustapha Cisse, Yann N. Dauphin, and David Lopez-Paz. "Mixup: Beyond Empirical Risk Minimization." arXiv preprint arXiv:1710.09412 (2017).
  10. Selvaraju, Ramprasaath R., Michael Cogswell, Abhishek Das, Ramakrishna Vedantam, Devi Parikh, and Dhruv Batra. "Grad-Cam: Visual Explanations from Deep Networks Via Gradient-Based Localization." In Proceedings of the IEEE international conference on computer vision, 2017, 618-626.
  11. Haralick, Robert M., Karthikeyan Shanmugam, and Its' Hak Dinstein. "Textural Features for Image Classification." IEEE Transactions on systems, man, and cybernetics 6 (1973): 610-621.
  12. Cheng, Jun, Wei Huang, Shuangliang Cao, Ru Yang, Wei Yang, Zhaoqiang Yun, Zhijian Wang, and Qianjin Feng. "Enhanced Performance of Brain Tumor Classification via Tumor Region Augmentation and Partition." PloS one 10, no. 10 (2015): e0140381.
  13. Ulli, Sai Shanmukh, Hanish Akkineni, Uma Santhosh Vuyyuru, Sai Ram Pathyala, Uttej Kumar Nannapaneni, and B. Suvarna. "Brain Tumor Detection Using Modified VGG-19 Model." In 2024 International Conference on Electrical Electronics and Computing Technologies (ICEECT), vol. 1, IEEE, 2024, 1-5.
  14. Khan, Amjad Rehman, Siraj Khan, Majid Harouni, Rashid Abbasi, Sajid Iqbal, and Zahid Mehmood. "Brain Tumor Segmentation Using K‐Means Clustering and Deep Learning with Synthetic Data Augmentation for Classification." Microscopy Research and Technique 84, no. 7 (2021): 1389-1399.
  15. Chollet, François. "Xception: Deep Learning with Depthwise Separable Convolutions." In Proceedings of the IEEE conference on computer vision and pattern recognition, 2017, 1251-1258.
  16. Hu, Jie, Li Shen, and Gang Sun. "Squeeze-and-Excitation Networks." In Proceedings of the IEEE conference on computer vision and pattern recognition, 2018, 7132-7141.
  17. Woo, Sanghyun, Jongchan Park, Joon-Young Lee, and In So Kweon. "Cbam: Convolutional Block Attention Module." In Proceedings of the European conference on computer vision (ECCV), 2018, 3-19.
  18. Baseer, K. K., K. Sivakumar, Duggineni Veeraiah, Gunjan Chhabra, Prasanna Kumar Lakineni, M. Jahir Pasha, Ramu Gandikota, and Gopakumar Harikrishnan. "Healthcare Diagnostics with an Adaptive Deep Learning Model Integrated with the Internet of Medical Things (IoMT) for Predicting Heart Disease." Biomedical Signal Processing and Control 92 (2024): 105988.
  19. Cui, Yin, Menglin Jia, Tsung-Yi Lin, Yang Song, and Serge Belongie. "Class-Balanced Loss Based on Effective Number of Samples." In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 2019, 9268-9277.
  20. Yun, Sangdoo, Dongyoon Han, Seong Joon Oh, Sanghyuk Chun, Junsuk Choe, and Youngjoon Yoo. "Cutmix: Regularization Strategy to Train Strong Classifiers with Localizable Features." In Proceedings of the IEEE/CVF international conference on computer vision, 2019, 6023-6032.
  21. Yahyaoui, Hela, Fethi Ghazouani, and Imed Riadh Farah. "Deep Learning Guided by an Ontology for Medical Images Classification Using a Multimodal Fusion." In 2021 International congress of advanced technology and engineering (ICOTEN), IEEE, 2021, 1-6.
  22. Ribeiro, Marco Tulio, Sameer Singh, and Carlos Guestrin. "" Why Should I Trust You?" Explaining the predictions of any classifier." In Proceedings of the 22nd ACM SIGKDD international conference on knowledge discovery and data mining, 2016, 1135-1144.
  23. Lundberg, Scott M., and Su-In Lee. "A Unified Approach to Interpreting Model Predictions." Advances in neural information processing systems 30 (2017).
  24. El-Assiouti, Omar S., Ghada Hamed, Hadeer El-Saadawy, Hala M. Ebied, and Dina Khattab. "RegionInpaint, Cutoff and RegionMix: Introducing Novel Augmentation Techniques for Enhancing the Generalization of Brain Tumor Identification." IEEE Access 11 (2023): 83232-83250.
  25. Chaki, J. “Brain Tumor MRI Dataset”. IEEE Dataport, 2023.