Authors : Ameer Hamza; Tang Siying; Yaoyao Ran; Ali Sajid; Muhammad Abubakr; Yihong Zhang

Volume/Issue : Volume 11 - 2026, Issue 4 - April

Google Scholar : https://tinyurl.com/mshz5uha

Scribd : https://tinyurl.com/2f9x66t4

DOI : https://doi.org/10.38124/ijisrt/26apr532

Note : A published paper may take 4-5 working days from the publication date to appear in PlumX Metrics, Semantic Scholar, and ResearchGate.

Abstract : Accurate segmentation of glioblastoma subregions from multimodal MRI with enhanced boundary delineation of tumor core is critical for optimal surgical and radiotherapy outcomes, yet current manual annotation approaches remain labor-intensive, subjective, and prone to significant inter-observer variability, limiting treatment planning efficacy. Attention U-Net with deep supervision and Monte Carlo Dropout-based uncertainty quantification is proposed that integrates attention gating at skip connections to enhance tumor-relevant feature detection and improve segmentation of clinically critical enhancing tumor regions on the BraTS2020 Dataset. The symmetric encoder-decoder design with progressive filter expansion (16→32→64→128→256), multi-scale deep supervision across four hierarchical decoder levels with optimized weighted loss functions (0.6, 0.2, 0.15, 0.05), and 3×3×3 convolutions, achieving the highest Enhancing Tumor Dice of 0.847 and Tumor Core Hausdorff Distance of 1.79 mm with an improvement over competing methods (nnU-Net, H2NF-Net, TransUNet) while maintaining robust performance across whole tumor segmentation and boundary delineation. This framework establishes a robust, objective standard for glioblastoma segmentation that reduces manual annotation burden, improves clinical consistency, and achieves superior boundary accuracy with 1.79 mm Hausdorff Distance, enhancing surgical and radiotherapy guidance for better patient care and improved treatment outcomes.

Keywords : BraTS2020 Dataset, Brain Tumor Segmentation, Attention U-Net, Deep Supervision, Boundary Enhancement HD95.

References :

1. Grech, N., et al., Rising Incidence of Glioblastoma Multiforme in a Well-Defined Population. Cureus, 2020. 12(5): p. e8195.
2. Reihanian, Z., et al., Impact of Age and Gender on Survival of Glioblastoma Multiforme Patients: A Multicenter Retrospective Study. Cancer Rep (Hoboken), 2024. 7(11): p. e70050.
3. Huang, J., et al., Brain tumor segmentation using deep learning: high performance with minimized MRI data. Frontiers in Radiology, 2025. Volume 5 - 2025.
4. Yin, X.X., et al., U-Net-Based Medical Image Segmentation. J Healthc Eng, 2022. 2022: p. 4189781.
5. Abidin, Z.U., et al., Recent deep learning-based brain tumor segmentation models using multi-modality magnetic resonance imaging: a prospective survey. Front Bioeng Biotechnol, 2024. 12: p. 1392807.
6. Akter, A., et al., Robust clinical applicable CNN and U-Net based algorithm for MRI classification and segmentation for brain tumor. Expert Systems with Applications, 2024. 238: p. 122347.
7. Lu, N.-H., et al., Deep learning-driven brain tumor classification and segmentation using non-contrast MRI. Scientific Reports, 2025. 15(1): p. 27831.
8. Visser, M., et al., Inter-rater agreement in glioma segmentations on longitudinal MRI. Neuroimage Clin, 2019. 22: p. 101727.
9. Ma, S., J. Tang, and F. Guo, Multi-Task Deep Supervision on Attention R2U-Net for Brain Tumor Segmentation. Front Oncol, 2021. 11: p. 704850.
10. Park, M., et al., ES-UNet: efficient 3D medical image segmentation with enhanced skip connections in 3D UNet. BMC Med Imaging, 2025. 25(1): p. 327.
11. Li, Y., et al., Automated segmentation of vertebral cortex with 3D U-Net-based deep convolutional neural network. Front Bioeng Biotechnol, 2022. 10: p. 996723.
12. Zhou, Z., et al., UNet++: Redesigning Skip Connections to Exploit Multiscale Features in Image Segmentation. IEEE Transactions on Medical Imaging, 2020. 39(6): p. 1856–1867.
13. Schlosser, M., S. Prettner, and T. Ivanovska. On Analysis of Swin UNETR: a Robust Architecture for Medical Image Segmentation. in 2023 1st International Conference on Optimization Techniques for Learning (ICOTL). 2023.
14. Chen, J., et al., TransUNet: Rethinking the U-Net architecture design for medical image segmentation through the lens of transformers. Medical Image Analysis, 2024. 97: p. 103280.
15. Das, N. and S. Das, Attention-UNet architectures with pretrained backbones for multi-class cardiac MR image segmentation. Curr Probl Cardiol, 2024. 49(1 Pt C): p. 102129.
16. Gu, R., et al., CA-Net: Comprehensive Attention Convolutional Neural Networks for Explainable Medical Image Segmentation. IEEE Trans Med Imaging, 2021. 40(2): p. 699–711.
17. Ye, Q., Y. Shi, and S. Guo, CAGs-Net: A Novel Adjacent-Context Network With Channel Attention Gate for 3D Brain Tumor Image Segmentation. Int J Biomed Imaging, 2025. 2025: p. 6656059.
18. Bareja, R., et al., nnU-Net-based Segmentation of Tumor Subcompartments in Pediatric Medulloblastoma Using Multiparametric MRI: A Multi-institutional Study. Radiol Artif Intell, 2024. 6(5): p. e230115.
19. Zhang, J., et al., Advances in attention mechanisms for medical image segmentation. Computer Science Review, 2025. 56: p. 100721.
20. Nalepa, J., M. Marcinkiewicz, and M. Kawulok, Data Augmentation for Brain-Tumor Segmentation: A Review. Front Comput Neurosci, 2019. 13: p. 83.
21. Li, J., et al., Rethinking Deep Supervision for Brain Tumor Segmentation. IEEE Transactions on Artificial Intelligence, 2024. 5(5): p. 2103–2116.
22. Isensee, F., Jäger, P. F., Full, P. M., Vollmuth, P., & Maier-Hein, K. H. (2020). nnU-Net for brain tumor segmentation. In International MICCAI Brainlesion Workshop (pp. 118-132). Springer, Cham.
23. van Marrewijk, B.M., et al., 3D plant segmentation: Comparing a 2D-to-3D segmentation method with state-of-the-art 3D segmentation algorithms. Biosystems Engineering, 2025. 254: p. 104147.
24. Usman Akbar, M., et al., Brain tumor segmentation using synthetic MR images - A comparison of GANs and diffusion models. Scientific Data, 2024. 11(1): p. 259.
25. Skandarani, Y., P.M. Jodoin, and A. Lalande, GANs for Medical Image Synthesis: An Empirical Study. J Imaging, 2023. 9(3).
26. Eijgelaar, R.S., et al., Robust Deep Learning–based Segmentation of Glioblastoma on Routine Clinical MRI Scans Using Sparsified Training. Radiology: Artificial Intelligence, 2020. 2(5): p. e190103.
27. Maqsood, R., et al., Optimal Res-UNET architecture with deep supervision for tumor segmentation. Front Med (Lausanne), 2025. 12: p. 1593016.
28. Wang, H., et al., MCA-UNet: multi-scale cross co-attentional U-Net for automatic medical image segmentation. Health Inf Sci Syst, 2023. 11(1): p. 10.
29. Jones, C.K., et al., Direct quantification of epistemic and aleatoric uncertainty in 3D U-net segmentation. J Med Imaging (Bellingham), 2022. 9(3): p. 034002.
30. Gal, Y., & Ghahramani, Z. (2016). Dropout as a Bayesian approximation: Representing model uncertainty in deep learning. In M. F. Balcan & K. Q. Weinberger (Eds.), Proceedings of the 33rd International Conference on Machine Learning (Vol. 48, pp. 1050–1059). PMLR.
31. Mehrtash, A., et al., Confidence Calibration and Predictive Uncertainty Estimation for Deep Medical Image Segmentation. IEEE Trans Med Imaging, 2020. 39(12): p. 3868–3878.
32. Gade, M., et al., Impact of uncertainty quantification through conformal prediction on volume assessment from deep learning-based MRI prostate segmentation. Insights Imaging, 2024. 15(1): p. 286.
33. Tsaneva-Atanasova, K., G. Pederzanil, and M. Laviola, Decoding uncertainty for clinical decision-making. Philos Trans A Math Phys Eng Sci, 2025. 383(2292): p. 20240207.
34. Brown, K.E., S. Talbert, and D.A. Talbert, A QUEST for Model Assessment: Identifying Difficult Subgroups via Epistemic Uncertainty Quantification. AMIA Annu Symp Proc, 2023. 2023: p. 854–863.
35. O.K, S., et al., Uncertainty-aware segmentation quality prediction via deep learning Bayesian Modeling: Comprehensive evaluation and interpretation on skin cancer and liver segmentation. Computerized Medical Imaging and Graphics, 2025. 123: p. 102547.
36. Tamo, J.B., et al. Uncertainty-Aware Ensemble Learning Models for Out-of-Distribution Medical Imaging Analysis. in 2023 IEEE International Conference on Bioinformatics and Biomedicine (BIBM). 2023.
37. Hoebel, K.V., et al., Expert-centered Evaluation of Deep Learning Algorithms for Brain Tumor Segmentation. Radiol Artif Intell, 2024. 6(1): p. e220231.
38. Kumar, A., et al., A flexible 2.5D medical image segmentation approach with in-slice and cross-slice attention. Computers in Biology and Medicine, 2024. 182: p. 109173.
39. Hicks, S.A., et al., On evaluation metrics for medical applications of artificial intelligence. Sci Rep, 2022. 12(1): p. 5979.
40. Åkesson, J., J. Töger, and E. Heiberg, Random effects during training: Implications for deep learning-based medical image segmentation. Computers in Biology and Medicine, 2024. 180: p. 108944.
41. Yeung, M., et al., Unified Focal loss: Generalising Dice and cross entropy-based losses to handle class imbalanced medical image segmentation. Computerized Medical Imaging and Graphics, 2022. 95: p. 102026.
42. Mukherkjee, D., et al., Brain tumor image generation using an aggregation of GAN models with style transfer. Sci Rep, 2022. 12(1): p. 9141.
43. Nisa, Z.U., et al., Beyond Accuracy: Evaluating certainty of AI models for brain tumour detection. Computers in Biology and Medicine, 2025. 193: p. 110375.
44. Pemberton, H.G., et al., Multi-class glioma segmentation on real-world data with missing MRI sequences: comparison of three deep learning algorithms. Scientific Reports, 2023. 13(1): p. 18911.
45. El Jurdi, R., Varoquaux, G., & Colliot, O. (2023). Confidence intervals for performance estimates in brain MRI segmentation. arXiv. https://arxiv.org/abs/2307.10926v3.
46. Yuan, D., et al., μ-Net: Medical image segmentation using efficient and effective deep supervision. Computers in Biology and Medicine, 2023. 160: p. 106963.
47. Chen, W., et al., MAUNet: a mixed attention U-net with spatial multi-dimensional convolution and contextual feature calibration for 3D brain tumor segmentation in multimodal MRI. Front Neurosci, 2025. 19: p. 1682603.
48. Zhang, Y., Y. Han, and J. Zhang, MAU-Net: Mixed attention U-Net for MRI brain tumor segmentation. Math Biosci Eng, 2023. 20(12): p. 20510–20527.
49. Tufail, A.B., et al., On Improved 3D-CNN-Based Binary and Multiclass Classification of Alzheimer's Disease Using Neuroimaging Modalities and Data Augmentation Methods. J Healthc Eng, 2022. 2022: p. 1302170.
50. Alwadee, E.J., et al., LATUP-Net: A lightweight 3D attention U-Net with parallel convolutions for brain tumor segmentation. Comput Biol Med, 2025. 184: p. 109353.
51. Talukder, M.A., et al., ACU-Net: Attention-based convolutional U-Net model for segmenting brain tumors in fMRI images. DIGITAL HEALTH, 2025. 11: p. 20552076251320288.
52. Mzoughi, H., et al., Deep Multi-Scale 3D Convolutional Neural Network (CNN) for MRI Gliomas Brain Tumor Classification. J Digit Imaging, 2020. 33(4): p. 903–915.
53. Karimijafarbigloo, S., et al., MS-Former: Multi-Scale Self-Guided Transformer for Medical Image Segmentation, in Medical Imaging with Deep Learning, O. Ipek, et al., Editors. 2024, PMLR: Proceedings of Machine Learning Research. p. 680––694.
54. Fang, X. and P. Yan, Multi-Organ Segmentation Over Partially Labeled Datasets With Multi-Scale Feature Abstraction. IEEE Trans Med Imaging, 2020. 39(11): p. 3619–3629.
55. Taha, A.A. and A. Hanbury, Metrics for evaluating 3D medical image segmentation: analysis, selection, and tool. BMC Med Imaging, 2015. 15: p. 29.
56. Oktay, O., et al., Attention u-net: Learning where to look for the pancreas. arXiv preprint arXiv:1804.03999, 2018.
57. Gal, Y. and Z. Ghahramani. Dropout as a bayesian approximation: Representing model uncertainty in deep learning. in international conference on machine learning. 2016. PMLR.
58. Isensee, F., et al., Automated design of deep learning methods for biomedical image segmentation. arXiv preprint arXiv:1904.08128, 2019.
59. Abrar, M., et al., Enhancing brain tumor segmentation using attention based convolutional UNet on MRI images. Sci Rep, 2025. 15(1): p. 36603.
60. Chen, J., et al., TransUNet: Rethinking the U-Net architecture design for medical image segmentation through the lens of transformers. Medical Image Analysis, 2024. 97: p. 103280.
61. Hatamizadeh, A., et al. Swin unetr: Swin transformers for semantic segmentation of brain tumors in mri images. in International MICCAI brainlesion workshop. 2021. Springer.
62. Wu, X., et al., FCFDiff-Net: full-conditional feature diffusion embedded network for 3D brain tumor segmentation. Quant Imaging Med Surg, 2025. 15(5): p. 4217–4234.
63. Chen, W., et al., MAUNet: a mixed attention U-net with spatial multi-dimensional convolution and contextual feature calibration for 3D brain tumor segmentation in multimodal MRI. Frontiers in Neuroscience, 2025. Volume 19 - 2025.