R-AttNet: Residual Encoder-Induced Attention Network for Brain Lesion Localization

Authors

  • Pradyumna Kumar Sahoo Gandhi Institute of Engineering and Technology University, Gunupur, India
  • Bhramara Bar Biswal Gandhi Institute of Engineering and Technology University, Gunupur, India
  • Deepak Kumar Sahoo Sri Sri University, Cuttack, India https://orcid.org/0000-0002-1816-1756

DOI:

https://doi.org/10.3991/ijoe.v22i05.59425

Keywords:

Brain Tumor, MRI, Residual Encoder, Attention Map

Abstract


Early brain tumor localization is a crucial task for better treatment planning. Computer vision plays a significant role in vision-assisted diagnosis of tumors. In the last few decades, various vision-assisted techniques have been developed by different researchers. However, these existing techniques are incapable of accurately separating tumors of varying sizes from different modalities. Therefore, in this paper, we introduce R-AttNet, a deep learning-based residual system for accurate brain tumor localization. The developed framework has three phases of novelties: the developed residual encoder network (RAN) comprises various residual blocks that make the model computationally efficient and can extract the required multiscale details effectively. The designed spatial attention mechanism acts as a bridge between the encoder and decoder that highlights the prominent details, which are highly essential for tumor localization. The proposed decoder network provides the extracted tumor mask while retaining the spatial dependency among the pixels efficiently. The effectiveness of the developed algorithm is corroborated using visual demonstration as well as objective analysis. The findings of this study, compared against various existing learning-based systems, may be suitable for clinical settings.

Author Biographies

Pradyumna Kumar Sahoo, Gandhi Institute of Engineering and Technology University, Gunupur, India

Pradyumna Kumar Sahoo is a Ph.D. scholar in the Department of Computer Science and Engineering at the Gandhi Institute of Engineering and Technology University, Gunupur, Odisha. His research focuses on developing machine learning algorithms for large-scale data analytics, with a particular emphasis on graph neural networks and their applications in social network analysis. He holds a Master’s degree in Computer Science from the Biju Patnaik University of Technology (BPUT), Odisha. Pradyumna has showcased his research at prestigious conferences such as NeurIPS and ICML. In addition to his academic pursuits, he mentors undergraduate students and actively contributes to initiatives that promote women’s participation in technology. (EMAIL: pradyumna.sahoo@giet.edu).

Deepak Kumar Sahoo, Sri Sri University, Cuttack, India

Deepak Kumar Sahoo, earned his Master in computer science and engineering from International Institute of Information Technology (IIIT-Bhubaneswar). He earned his Ph.D in computer science and engineering International Institute of Information Technology (IIIT-Bhubaneswar). He has worked for Odia-vertical at IIIT-Bhubaneswar, in a Consortia project headed by IIIT-Bombay having the project title “Cross Lingual Information Access”. He has over 15 years of teaching and research experience in various organizations. He has authored 34 publications out of which 19 papers in different journals indexed in SCI & Scopus, 11 conference and 4 book chapters. (EMAIL: deepak.s@srisriuniversity.edu.in).

References

[1] F. J. Dorfner, J. B. Patel, J. Kalpathy-Cramer, E. R. Gerstner, and C. P. Bridge, “A review of deep learning for brain tumor analysis in MRI,” npj Precis. Oncol., vol. 9, no. 1, p. 2, Jan. 2025, doi: 10.1038/s41698-024-00789-2.

[2] A. K. Sahoo, P. Parida, M. K. Panda, K. Muralibabu, and A. S. Mohanty, “MultiTumor Analyzer (MTA-20–55): A network for efficient classification of detected brain tumors from MRI images,” Biocybern. Biomed. Eng., vol. 44, no. 3, pp. 617–634, Jul. 2024, doi: 10.1016/j.bbe.2024.06.003.

[3] M. Ariful Islam, M. F. Mridha, M. Safran, S. Alfarhood, and M. Mohsin Kabir, “Revolutionizing Brain Tumor Detection Using Explainable AI in MRI Images,” NMR Biomed., vol. 38, no. 3, Mar. 2025, doi: 10.1002/nbm.70001.

[4] A. K. Sahoo, P. Parida, and K. Muralibabu, “Hybrid deep neural network with clustering algorithms for effective gliomas segmentation,” Int. J. Syst. Assur. Eng. Manag., vol. 15, no. 3, pp. 964–980, Mar. 2024, doi: 10.1007/s13198-023-02183-w.

[5] A. K. Sahoo, P. Parida, K. Muralibabu, and S. Dash, “An improved DNN with FFCM method for multimodal brain tumor segmentation,” Intell. Syst. with Appl., vol. 18, p. 200245, May 2023, doi: 10.1016/j.iswa.2023.200245.

[6] A. K. Sahoo and P. Parida, “Automatic clustering based approach for brain tumor extraction,” in Journal of Physics: Conference Series, Goa, India, May 2021, p. 012007. doi: 10.1088/1742-6596/1921/1/012007.

[7] A. K. Sahoo and P. Parida, “A Clustering Based Approach For Meningioma Tumors Extraction From Brain MRI Images,” in 2020 IEEE International Symposium on Sustainable Energy, Signal Processing and Cyber Security (iSSSC), IEEE, Dec. 2020, pp. 1–5. doi: 10.1109/iSSSC50941.2020.9358849.

[8] A. S. Mohanty, K. C. Patra, and P. Parida, “Toddler ASD Classification Using Machine Learning Techniques,” Int. J. online Biomed. Eng., vol. 17, no. 7, pp. 156–171, Jul. 2021, doi: 10.3991/ijoe.v17i07.23497.

[9] R. Ray, S. Jena, P. Parida, L. Dash, and S. K. Biswal, “Empowering Diabetic Eye Disease Detection: Leveraging Differential Evolution for Optimized Convolution Neural Networks,” Int. J. Online Biomed. Eng., vol. 20, no. 10, pp. 86–100, Jul. 2024, doi: 10.3991/ijoe.v20i10.49187.

[10] T. A. Ton Komar Azaharan, A. K. Mahamad, S. Saon, Muladi, and S. W. Mudjanarko, “Investigation of VGG-16, ResNet-50 and AlexNet Performance for Brain Tumor Detection,” Int. J. Online Biomed. Eng., vol. 19, no. 08, pp. 97–109, Jun. 2023, doi: 10.3991/ijoe.v19i08.38619.

[11] Chandrakala C.B, P. S., C. Pujari, S. Ketavarapu, S. Awatramani, and S. Gohil, “Health-Lens: A Health Diagnosis Companion,” Int. J. Interact. Mob. Technol., vol. 19, no. 12, pp. 68–102, Jun. 2025, doi: 10.3991/ijim.v19i12.51525.

[12] W. El-Shafai et al., “Hybrid Segmentation Approach for Different Medical Image Modalities,” Comput. Mater. Contin., vol. 73, no. 2, pp. 3455–3472, 2022, doi: 10.32604/cmc.2022.028722.

[13] P. E. Kasar, S. M. Jadhav, and V. Kansal, “MRI Modality-based Brain Tumor Segmentation Using Deep Neural Networks,” May 12, 2021. doi: 10.21203/rs.3.rs-496162/v1.

[14] Y. Jiang, Y. Zhang, X. Lin, J. Dong, T. Cheng, and J. Liang, “SwinBTS: A Method for 3D Multimodal Brain Tumor Segmentation Using Swin Transformer,” Brain Sci., vol. 12, no. 6, p. 797, Jun. 2022, doi: 10.3390/brainsci12060797.

[15] X. Liu, P. Gao, T. Yu, F. Wang, and R.-Y. Yuan, “CSWin-UNet: Transformer UNet with cross-shaped windows for medical image segmentation,” Inf. Fusion, vol. 113, p. 102634, Jan. 2025, doi: 10.1016/j.inffus.2024.102634.

[16] M. R. Islam, M. Qaraqe, and E. Serpedin, “CoST-UNet: Convolution and swin transformer based deep learning architecture for cardiac segmentation,” Biomed. Signal Process. Control, vol. 96, p. 106633, Oct. 2024, doi: 10.1016/j.bspc.2024.106633.

[17] Y. Liu, Y. Ma, Z. Zhu, J. Cheng, and X. Chen, “TransSea: Hybrid CNN–Transformer With Semantic Awareness for 3-D Brain Tumor Segmentation,” IEEE Trans. Instrum. Meas., vol. 73, pp. 16–31, 2024, doi: 10.1109/TIM.2024.3413130.

[18] N. Rasool, J. Iqbal Bhat, N. Ahmad Wani, N. Ahmad, and M. Alshara, “TransResUNet: Revolutionizing Glioma Brain Tumor Segmentation Through Transformer-Enhanced Residual UNet,” IEEE Access, vol. 12, pp. 72105–72116, 2024, doi: 10.1109/ACCESS.2024.3402947.

[19] W. Zhang, S. Chen, Y. Ma, Y. Liu, and X. Cao, “ETUNet:Exploring efficient transformer enhanced UNet for 3D brain tumor segmentation,” Comput. Biol. Med., vol. 171, p. 108005, Mar. 2024, doi: 10.1016/j.compbiomed.2024.108005.

[20] F. Ghazouani, P. Vera, and S. Ruan, “Efficient brain tumor segmentation using Swin transformer and enhanced local self-attention,” Int. J. Comput. Assist. Radiol. Surg., vol. 19, no. 2, pp. 273–281, Oct. 2023, doi: 10.1007/s11548-023-03024-8.

[21] G. Ramasamy, T. Singh, and X. Yuan, “Multi-Modal Semantic Segmentation Model using Encoder Based Link-Net Architecture for BraTS 2020 Challenge,” Procedia Comput. Sci., vol. 218, pp. 732–740, 2023, doi: 10.1016/j.procs.2023.01.053.

[22] T. M. Angona and M. R. H. Mondal, “An attention based residual U-Net with swin transformer for brain MRI segmentation,” Array, vol. 25, p. 100376, Mar. 2025, doi: 10.1016/j.array.2025.100376.

[23] F. D. Hernandez-Gutierrez et al., “Brain Tumor Segmentation from Optimal MRI Slices Using a Lightweight U-Net,” Technologies, vol. 12, no. 10, p. 183, Oct. 2024, doi: 10.3390/technologies12100183.

[24] Jun Cheng, “Brain Tumor Dataset,” Figshare. [Online]. Available: https://figshare.com/articles/dataset/brain_tumor_dataset/1512427

[25] S. Bakas et al., “Identifying the Best Machine Learning Algorithms for Brain Tumor Segmentation, Progression Assessment, and Overall Survival Prediction in the BRATS Challenge,” Apr. 2019, [Online]. Available: http://arxiv.org/abs/1811.02629

[26] Q.-H. Trinh, T.-H. N. Mau, R. Zosimov, and M.-V. Nguyen, “EfficientNet for Brain-Lesion classification,” Aug. 2022, [Online]. Available: http://arxiv.org/abs/2208.04616

[27] A. K. Sahoo, P. Parida, and K. Muralibabu, “Effective Use of Clustering Techniques for Brain Tumor Segmentation,” in 2023 IEEE 3rd International Conference on Applied Electromagnetics, Signal Processing, & Communication (AESPC), IEEE, Nov. 2023, pp. 1–5. doi: 10.1109/AESPC59761.2023.10390467.

[28] J. Sun et al., “MVSI-Net: Multi-view attention and multi-scale feature interaction for brain tumor segmentation,” Biomed. Signal Process. Control, vol. 95, p. 106484, Sep. 2024, doi: 10.1016/j.bspc.2024.106484.

[29] S. Rajput, R. Kapdi, M. Roy, and M. S. Raval, “A triplanar ensemble model for brain tumor segmentation with volumetric multiparametric magnetic resonance images,” Healthc. Anal., vol. 5, p. 100307, Jun. 2024, doi: 10.1016/j.health.2024.100307.

Cover iJOE – Vol 22, No 05 (2026)

Downloads

Published

2026-05-11

How to Cite

Sahoo, P. K., Biswal, B. B., & Sahoo, D. K. (2026). R-AttNet: Residual Encoder-Induced Attention Network for Brain Lesion Localization. International Journal of Online and Biomedical Engineering (iJOE), 22(05), pp. 70–86. https://doi.org/10.3991/ijoe.v22i05.59425

Issue

Vol. 22 No. 05 (2026)

Section

Papers

License

Copyright (c) 2026 Pradyumna Kumar Sahoo, Bhramara Bar Biswal, Deepak Kumar Sahoo

Creative Commons License

This work is licensed under a Creative Commons Attribution 4.0 International License.