N. Krishnavardhan; Bhavani R.; Ch. V. S. Satyamurty; Spandana Shivanadhuni; Rajesh Nekkanti; Ashwini Barbadekar

Multi-Task Learning for Brain Tumor Diagnosis: MRI-Glioma Net with Residual Attention and Grading Head

PDF (807KB), PP.110-125

Views: 0 Downloads: 0

Author(s)

N. Krishnavardhan ^1,* Bhavani R. ² Ch. V. S. Satyamurty ³ Spandana Shivanadhuni ⁴ Rajesh Nekkanti ⁵ Ashwini Barbadekar ⁶

1. Department of Data Science, Geethanjali College of Engineering and Technology, Hyderabad, 501301, India

2. Department of Computer Science and Engineering, Vel Tech Rangarajan Dr. Sagunthala R&D Institute of Science and Technology, Chennai, 600062, India

3. Department of Information Technology, CVR College of Engineering, Hyderabad, 501510, India

4. Department of Computer Science and Engineering, Koneru Lakshmaiah Education Foundation, Hyderabad, 500043, India

5. Department of Electrical and Electronics Engineering, Aditya University, Surampalem, 533437, India

6. Department of Electronics and Telecommunication Engineering, Vishwakarma Institute of Technology, Pune, 411037, India

* Corresponding author.

DOI: https://doi.org/10.5815/ijigsp.2026.05.07

Received: 18 Feb. 2026 / Revised: 5 Mar. 2026 / Accepted: 31 Mar. 2026 / Published: 8 Oct. 2026

Index Terms

Brain tumor Segmentation, Multi-scale Features, Residual Learning, Glioma Grading, Dice Similarity Coefficient, Tumor Core.

Abstract

To accurately segment brain tumors and grade gliomas using multi-modal MRI data, MRI-Glioma Net was built as a new 3D Res-UNet framework. The model uses residual learning, multi-scale feature extraction, and attention-enhanced fusion to delineate three heterogeneous tumor subregions: whole tumor (WT), tumor core (TC), and enhancing tumor (ET). It leverages T1, T2, FLAIR, and T1ce sequences. It uses fused latent features to incorporate a specialized classification head for glioma grading (low-grade vs. high-grade), which improves discriminative capability. A glioma dataset with extensive cross-validation was used for evaluation across multiple institutions. MRI-Glioma Net outperformed baseline models such as 3D UNet and Res UNet, achieving Dice Similarity Coefficients (DSCs) of 0.94 (WT), 0.90 (TC), and 0.88 (ET), respectively. An IoU of 0.86, HD95 of 5.4 mm, and a volumetric similarity of 0.95 were all recorded by the model. In terms of grading, it achieved better results than attention UNet and nn UNet, with 95% accuracy, 94% precision, 93% recall, and 93.5% F1-score. With a 7.2 GB GPU usage, a 2.1s inference time per volume, and only a 0.9% accuracy loss post-quantization, the model's efficiency metrics demonstrate its lightweight deployment potential. Additionally, 95% confidence intervals are computed for key metrics to reflect variability across folds. Statistical significance of improvements over baseline models (3D U-Net, Res-UNet, Attention U-Net, nnU-Net) is evaluated using paired statistical tests (e.g., Wilcoxon signed-rank test), confirming that performance gains are not due to random variation. Efficiency is further validated using inference time per volume, GPU memory consumption, and post-quantization performance, ensuring practical deployment feasibility. The anatomical fidelity is shown by qualitative overlays to be superior, and interpretability is improved by Grad-CAM and error maps. MRI-Glioma Net offers a feasible, effective, and interpretable way to classify gliomas and diagnose tumors in real time in clinical settings. MRI-Glioma Net represents a major step forward in neuro-oncology imaging; its strong performance across segmentation and grading tasks suggests it could be useful for pre-surgical planning, prognosis, and monitoring treatment response.

Cite This Paper

N. Krishnavardhan, Bhavani R., Ch. V. S. Satyamurty, Spandana Shivanadhuni, Rajesh Nekkanti, Ashwini Barbadekar, "Multi-Task Learning for Brain Tumor Diagnosis: MRI-Glioma Net with Residual Attention and Grading Head", International Journal of Image, Graphics and Signal Processing(IJIGSP), Vol.18, No.5, pp. 110-125, 2026. DOI:10.5815/ijigsp.2026.05.07

Reference

[1]J. Amin, M. Sharif, M. Raza, & M. Yasmin, “Detection of brain tumor based on features fusion and machine learning.” Journal of Ambient Intelligence and Humanized Computing, vol. 15, no. 1, pp.983-999, 2024.
[2]M. F. Khan, A. Iftikhar, H. Anwar, & S. A. Ramay, “Brain tumor segmentation and classification using optimized deep learning.” Journal of Computing & Biomedical Informatics, vol.7, no. 01, pp.632-640.
[3]R. Sajjanar, U. D. Dixit, & V. K. Vagga, “Advancements in hybrid approaches for brain tumor segmentation in MRI:” a comprehensive review of machine learning and deep learning techniques. Multimedia Tools and Applications, vol.83, no. 10, pp. 30505-30539, 2024.
[4]S. Anantharajan, S. Gunasekaran, & T. Subramanian, “MRI brain tumor detection using deep learning and machine learning approaches.” Measurement: Sensors, vol.31, 101026, 2024.
[5]S. Karim, G. Tong, Y. Yu, A. A. Laghari A. A. Khan, M. Ibrar, & F. Mehmood, “Developments in brain tumor segmentation using MRI:” deep learning insights and future perspectives, IEEE Access, vol.12, pp.26875-26896, 2024.
[6]M. F. Almufareh, M. Imran, A. Khan, M. Humayun, & M. Asim, “Automated brain tumor segmentation and classification in MRI using YOLO-based deep learning.” IEEE Access, vol.12, pp.16189-16207, 2024.
[7]U. Bhimavarapu, N. Chintalapudi, & G. Battineni, “Brain tumor detection and categorization with segmentation of improved unsupervised clustering approach and machine learning classifier.” Bioengineering, vol.11, no. 3, 266.
[8]C. M. Umarani, S. G. Gollagi, S. Allagi, K. Sambrekar, & S. B. Ankali, “Advancements in deep learning techniques for brain tumor segmentation:” A survey. Informatics in Medicine Unlocked, vol.50, 101576, 2024.
[9]Z. U. Abidin, R. A. Naqvi, A. Haider, H. S. Kim, D. Jeong, & S. W. Lee, “Recent deep learning-based brain tumor segmentation models using multi-modality magnetic resonance imaging:” A prospective survey. Frontiers in Bioengineering and Biotechnology, vol.12, 1392807, 2024.
[10]F. Ullah, M. Nadeem, & M. Abrar, “Revolutionizing brain tumor segmentation in MRI with dynamic fusion of handcrafted features and global pathway-based deep learning.” KSII Transactions on Internet and Information Systems (TIIS), vol.18, no. 1, pp.105-125, 2024.
[11]S. Stephe, S. B. Kumar, A. Thirumalraj, & V. Dzhyvak, “Transformer based attention guided network for segmentation and hybrid network for classification of liver tumor from CT scan images.” 2024.
[12]S. Ali, R. Khurram, K. U. Rehman, A. Yasin, Z. Shaukat, Z. Sakhawat, & G. Mujtaba, “An improved 3D U-Net-based deep learning system for brain tumor segmentation using multi-modal MRI.” Multimedia Tools and Applications, vol.83, no 37, pp. 85027-85046, 2024.
[13]J. N. Benedict, S. Shanmugapriya, S. Senthil Pandi, & P. Kumar, “Accurate Segmentation and Classification of Brain Tumor Using Deep Learning Approaches.” In 2024 Second International Conference on Advances in Information Technology (ICAIT) .Vol. 1, pp. 1-6, July, 2024. IEEE.
[14]M. S. Ullah, M. A. Khan, H. M. Albarakati, R. Damaševičius, & S. Alsenan, “Multimodal brain tumor segmentation and classification from MRI scans based on optimized DeepLabV3+ and interpreted networks information fusion empowered with explainable AI.” Computers in biology and medicine, vol. 182, 109183, 2024.
[15]A. Verma, S. N. Shivhare, S. P. Singh, N. Kumar, & A. Nayyar, “Comprehensive Review on MRI-Based Brain Tumor Segmentation:” A Comparative Study from 2017 Onwards: Comprehensive Review on MRI-Based...: A. Verma et al. Archives of Computational Methods in Engineering, vol.31, no. 8, 2024.
[16]L. Dénes-Fazakas, L. Kovács, G. Eigner, & L. Szilágyi, “Enhancing Brain Tumor Diagnosis with L-Net: A Novel Deep Learning Approach for MRI Image Segmentation and Classification.” Biomedicines, vol.12, no.10, 2388, 2024.
[17]R. Ahsan, I. Shahzadi, F. Najeeb, & H. Omer, “Brain tumor detection and segmentation using deep learning.” Magnetic Resonance Materials in Physics, Biology and Medicine, vol. 38, no. 1, pp.13-22, 2025.
[18]P. K. Tiwary, P. Johri, A. Katiyar, &M. K. Chhipa,” Deep Learning-Based MRI Brain Tumor Segmentation with EfficientNet-Enhanced UNet.” IEEE Access, 2025.
[19]D. Rastogi, P. Kadry, M. Donelli, S. Kadry, A. A. Khan, G. Espa, & J. Kim, “Deep learning-integrated MRI brain tumor analysis:” feature extraction, segmentation, and Survival Prediction using Replicator and volumetric networks. Scientific Reports, vol.15, no. 1, 1437, 2025.
[20]U. A. Bhatti, J. Liu, M. Huang, & Y. Zhang, FF-UUNet: “Feature fusion based deep learning-powered enhanced framework for accurate brain tumor segmentation in MRI images.” Image and Vision Computing, 105635, 2025.
[21]Z. Schwehr, & S. Achanta, “Brain tumor segmentation based on deep learning, attention mechanisms, and energy-based uncertainty predictions.” Multimedia Tools and Applications, pp.1-20, 2025.
[22]M. A. Ilani, D. Shi, & Y. M. Banad, “T1-weighted MRI-based brain tumor classification using hybrid deep learning models.” Scientific Reports, vol.15, no. 1, 7010, 2025.
[23]K. Lakshmi, S. Amaran, G. Subbu Lakshmi, S. Padmini, G. P. Joshi, & W. Cho, “Explainable artificial intelligence with UUNet based segmentation and Bayesian machine learning for classification of brain tumors using MRI images.” Scientific Reports, vol. 15, no. 1, 690, 2025.
[24]P. V. Kusuma, & S. C. M. Reddy, “Brain tumor segmentation and classification using MRI: Modified segnet model and hybrid deep learning architecture with improved texture features.” Computational Biology and Chemistry, vol.117, 108381, 2025.
[25]S. Stephen, B. Manjunath, V. Revathi, & A. Thirumal raj,” Osteosarcoma cancer detection using ghost-faster RCNN model from histopathological images.” Iran Journal of Computer Science, vol 8, no. 1, pp. 217-231, 2025.
[26]S. Baswaraju, A. Thirumal raj, & B. Manjunath, “Unlocking the potential of deep learning in knee bone Cancer diagnosis using MSCSA-Net segmentation and MLGC-LTNet classification.” In Sustainable Development Using Private AI, pp. 190-213, 2024. CRC Press
[27]https://www.kaggle.com/datasets/awsaf49/brats2020-training-data

International Journal of Image, Graphics and Signal Processing (IJIGSP)