A3Net: A Lightweight Attention-Guided Encoder–Decoder for Multi-Class Brain Tumor Segmentation with Cross-Dataset Domain Gap Characterization

Abstract

Accurate multi-class segmentation of brain tumors from T1-weighted contrast-enhanced MRI is essential for surgical planning and treatment monitoring, yet existing high-performance architectures exceed 30 M parameters, limiting deployment in resource-constrained clinical settings. This work introduces A3Net, a lightweight encoder–decoder combining depth-wise separable convolutions, squeeze-and-excitation channel attention, and summation-based skip fusion with a cross-entropy + Dice composite loss. On the cleaned BRISC 2025 benchmark (47 duplicates removed by MD5 audit) under 2-seed × 5-fold cross-validation, A3Net attains 77.79 ± 0.29% mIoU and 87.14 ± 0.17% Dice at 2.11 M parameters, 1.39 GFLOPs, and 8.15 ms latency, achieving statistical parity with Attention U-Net (P=0.334) at 15.0× lower parameter cost. A3Net achieves significant per-class advantages on meningioma (+1.54 pp IoU, P=0.007) and pituitary (+0.42 pp IoU, P=0.016), while Attention U-Net retains a significant advantage on glioma (−1.49 pp IoU, p < 0.001). In-domain evaluation on Figshare yields 74.66 ± 0.50% mIoU (+1.65 pp over Attention U-Net), and a supplementary probe on Akter-Seg confirms efficiency parity with Attention U-Net (83.30 ± 0.44% mIoU vs. 83.21%) at 15.0× fewer parameters across two primary and one supplementary benchmark. Zero-shot cross-dataset transfer exposes a strongly asymmetric domain gap (−12.64 pp BRISC→Figshare; −29.90 pp Figshare→BRISC); multi-source standardized training (BRISC ∪ Figshare, N = 6,655) provides a partial +3.05 pp recovery under target-inclusive training, which is a fundamentally different experimental condition from the target-blind zero-shot setting in which the −29.90 pp gap is measured; the two quantities are therefore not directly commensurable. Class-distribution mismatch, rather than intensity shift, is identified as the dominant residual barrier. Overall, A3Net offers a statistically validated, computationally efficient solution for multi-class brain tumor segmentation with demonstrated cross-dataset generalization.

Keywords

• Brain tumor segmentation
• MRI
• Lightweight deep learning
• Squeeze-and-excitation attention
• Cross-dataset generalization

Ethical Statement

Ethics committee approval was not required for this study because it did not involve new data collection from humans or animals; all MRI data used (BRISC 2025, Figshare, Akter-Seg, and the Mendeley external dataset) are publicly available, de-identified, secondary datasets used strictly for computational model development and evaluation.

Thanks

The authors did not receive administrative, technical, or in-kind support beyond the resources acknowledged in the Funding statement.

References

1. Akter, A., Nosheen, N., Ahmed, S., Hossain, M. Z., Yousuf, M. A., Almoyad, M. A. A., Hasan, K. F., & Moni, M. A. (2024). Robust clinical applicable CNN and U-Net based algorithm for MRI classification and segmentation for brain tumor. Expert Systems with Applications, 238, Article 122347. https://doi.org/10.1016/j.eswa.2023.122347
2. Badža, M. M., & Barjaktarović, M. C. (2021). Segmentation of brain tumors from MRI images using convolutional autoencoder. Applied Sciences, 11(9), Article 4317. https://doi.org/10.3390/app11094317
3. Bakas, S., Reyes, M., Jakab, A., Bauer, S., Rempfler, M., Crimi, A., Shin, H. G., Berger, C., Ha, S. M., Rozycki, M., Prastawa, M., Alberts, E., Rieger, J., Sako, C., Wiest, R., Pfeiffer, D., Kirschke, J. S., Koumoutsakos, P., Davatzikos, C., ... Menze, B. H. (2018). Identifying the best machine learning algorithms for brain tumor segmentation, progression assessment, and overall survival prediction in the BRATS challenge (arXiv:1811.02629). arXiv. https://doi.org/10.48550/arXiv.1811.02629
4. Bhamboo, A. K., Mahala, S., Kumar, Y., Shekhawat, N. S., & Sharma, K. (2025). Trust-refined U-Net ensemble for uncertainty-aware brain tumor segmentation. TechRxiv. https://doi.org/10.36227/techrxiv.175492140.09803812/v1
5. Bhuiyan, M. R. I., Bhat, S., Qahqaie, M., Rahman, M. M., & Islam, M. S. (2026). LoGSAM: Parameter-efficient cross-modal grounding for MRI segmentation (arXiv:2603.17576). arXiv. https://doi.org/10.48550/arXiv.2603.17576
6. Bindhu, R., & Vimala, S. (2026). A-Lite UACNet: Hybrid attention U-Net with active contour for efficient brain tumor segmentation. In Proceedings of the 2026 IEEE Recent Advances in Electrical, Electronics, Ubiquitous Communication, and Computational Intelligence (RAEEUCCI) (pp. 1–6). IEEE. https://doi.org/10.1109/RAEEUCCI67649.2026.11504790
7. Cao, H., Wang, Y., Chen, J., Jiang, D., Zhang, X., Tian, Q., & Wang, M. (2022). Swin-UNet: UNet-like pure transformer for medical image segmentation. In L. Karlinsky, T. Michaeli, & K. Nishino (Eds.), Computer Vision – ECCV 2022 Workshops (Lecture Notes in Computer Science, Vol. 13803, pp. 205–218). Springer. https://doi.org/10.1007/978-3-031-25066-8_9
8. Chaurasia, A., & Culurciello, E. (2017). LinkNet: Exploiting encoder representations for efficient semantic segmentation. In Proceedings of the 2017 IEEE Visual Communications and Image Processing (VCIP) (pp. 1–4). IEEE. https://doi.org/10.1109/VCIP.2017.8305148

9. Chen, L.-C., Zhu, Y., Papandreou, G., Schroff, F., & Adam, H. (2018). Encoder–decoder with atrous separable convolution for semantic image segmentation. In V. Ferrari, M. Hebert, C. Sminchisescu, & Y. Weiss (Eds.), Computer Vision – ECCV 2018 (Lecture Notes in Computer Science, Vol. 11211, pp. 833–851). Springer. https://doi.org/10.1007/978-3-030-01234-2_49
10. Cheng, J. (2017). Brain tumor dataset [Data set]. Figshare. https://doi.org/10.6084/m9.figshare.1512427
11. Chollet, F. (2017). Xception: Deep learning with depthwise separable convolutions. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (pp. 1251–1258). IEEE. https://doi.org/10.1109/CVPR.2017.195
12. Díaz-Pernas, F. J., Martínez-Zarzuela, M., Antón-Rodríguez, M., & González-Ortega, D. (2021). A deep learning approach for brain tumor classification and segmentation using a multiscale convolutional neural network. Healthcare, 9(2), Article 153. https://doi.org/10.3390/healthcare9020153
13. Dou, Q., Ouyang, C., Chen, C., Chen, H., Glocker, B., Zhuang, X., & Heng, P.-A. (2019). PnP-AdaNet: Plug-and-play adversarial domain adaptation network with a benchmark at cross-modality cardiac segmentation. IEEE Access, 7, 99–111. https://doi.org/10.1109/ACCESS.2019.2929258
14. Elfwing, S., Uchibe, E., & Doya, K. (2018). Sigmoid-weighted linear units for neural network function approximation in reinforcement learning. Neural Networks, 107, 3–11. https://doi.org/10.1016/j.neunet.2017.12.012
15. Fahim, T. A., Alam, F. B., & Hossain, M. A. (2025). Brain tumor detection, classification and segmentation by deep learning models from MRI images: Recent approaches, challenges and future directions. Array, 28, Article 100571. https://doi.org/10.1016/j.array.2025.100571
16. Fateh, A., Rezvani, Y., Moayedi, S., Shirani, M., Rahimi, A., & Kashef, R. (2026). BRISC: Annotated dataset for brain tumor segmentation and classification. Scientific Data, 13, Article 361. https://doi.org/10.1038/s41597-026-06753-y
17. Filimonova, E., Leone, A., Carbone, F., D'Agata, F., Rossi, R., & Vercelli, A. (2025). Real-world multi-institutional glioblastoma MRI dataset with standardized preprocessing, expert-validated segmentation, and MGMT promoter profiling (v1.0.1) [Data set]. OpenNeuro. https://doi.org/10.18112/openneuro.ds007045.v1.0.1
18. Ganin, Y., Ustinova, E., Ajakan, H., Germain, P., Larochelle, H., Laviolette, F., Marchand, M., & Lempitsky, V. (2016). Domain-adversarial training of neural networks. Journal of Machine Learning Research, 17(59), 1–35. https://jmlr.org/papers/v17/15-239.html
19. Guan, H., & Liu, M. (2022). Domain adaptation for medical image analysis: A survey. IEEE Transactions on Biomedical Engineering, 69(3), 1173–1185. https://doi.org/10.1109/TBME.2021.3117407
20. Havaei, M., Davy, A., Warde-Farley, D., Biard, A., Courville, A., Bengio, Y., Pal, C., Jodoin, P.-M., & Larochelle, H. (2017). Brain tumor segmentation with deep neural networks. Medical Image Analysis, 35, 18–31. https://doi.org/10.1016/j.media.2016.05.004
21. Hira, M. I. K., Hossain, M. S., Bithee, M. M. A., Sara, U., Hasan, M. M., Towsif, A. A., & Ahmed, M. K. (2026). Brain tumor MRI dataset (Glioma, Meningioma, Pituitary, No Tumor) (V6) [Data set]. Mendeley Data. https://doi.org/10.17632/zwr4ntf94j.6
22. Howard, A. G., Zhu, M., Chen, B., Kalenichenko, D., Wang, W., Weyand, T., Andreetto, M., & Adam, H. (2017). MobileNets: Efficient convolutional neural networks for mobile vision applications (arXiv:1704.04861). arXiv. https://doi.org/10.48550/arXiv.1704.04861
23. Howard, A., Sandler, M., Chu, G., Chen, L.-C., Chen, B., Tan, M., Wang, W., Yukrut, Y., Vasudevan, V., Le, Q. V., & Adam, H. (2019). Searching for MobileNetV3. In Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV) (pp. 1314–1324). IEEE. https://doi.org/10.1109/ICCV.2019.00140
24. Hu, J., Shen, L., & Sun, G. (2018). Squeeze-and-excitation networks. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (pp. 7132–7141). IEEE. https://doi.org/10.1109/CVPR.2018.00745
25. Huang, X., & Belongie, S. (2017). Arbitrary style transfer in real-time with adaptive instance normalization. In Proceedings of the IEEE International Conference on Computer Vision (ICCV) (pp. 1501–1510). IEEE. https://doi.org/10.1109/ICCV.2017.167
26. Isensee, F., Jaeger, P. F., Kohl, S. A. A., Petersen, J., & Maier-Hein, K. H. (2021). nnU-Net: A self-configuring method for deep learning-based biomedical image segmentation. Nature Methods, 18(2), 203–211. https://doi.org/10.1038/s41592-020-01008-z
27. Kamnitsas, K., Baumgartner, C., Ledig, C., Newcombe, V., Simpson, J., Kane, A., Menon, D., Nori, A., Criminisi, A., Rueckert, D., & Glocker, B. (2017). Unsupervised domain adaptation in brain lesion segmentation with adversarial networks. In S. Ourelin, L. Joskowicz, M. R. Sabuncu, G. Unal, & W. Wells (Eds.), Information Processing in Medical Imaging (IPMI 2017) (Lecture Notes in Computer Science, Vol. 10265, pp. 597–609). Springer. https://doi.org/10.1007/978-3-319-59050-9_47
28. Kuzhan, Y., & Günay, M. (2026). Comparative evaluation of CNN and transformer-based models for brain tumor MRI segmentation. Journal of Computer Science and Technology Studies, 8(7), 75–83. https://alkindipublisher.com/index.php/jcsts/article/view/12720/11395
29. Loshchilov, I., & Hutter, F. (2017). SGDR: Stochastic gradient descent with warm restarts [Paper presentation]. 5th International Conference on Learning Representations (ICLR 2017), Toulon, France. https://doi.org/10.48550/arXiv.1608.03983
30. Loshchilov, I., & Hutter, F. (2019). Decoupled weight decay regularization [Paper presentation]. 7th International Conference on Learning Representations (ICLR 2019), New Orleans, LA, United States. https://doi.org/10.48550/arXiv.1711.05101
31. Louis, D. N., Perry, A., Wesseling, P., Brat, D. J., Cree, I. A., Figarella-Branger, D., Hawkins, C., Ng, H. K., Pfister, S. M., Reifenberger, G., Soffietti, R., von Deimling, A., & Ellison, D. W. (2021). The 2021 WHO classification of tumors of the central nervous system: A summary. Neuro-Oncology, 23(8), 1231–1251. https://doi.org/10.1093/neuonc/noab106
32. Mayala, S., Herdlevær, I., Haugsøen, J. B., Anandan, S., Gavasso, S., & Brun, M. (2022). Brain tumor segmentation based on minimum spanning tree. Frontiers in Signal Processing, 2, Article 816186. https://doi.org/10.3389/frsip.2022.816186
33. Mehta, S., Rastegari, M., Shapiro, L., & Hajishirzi, H. (2019). ESPNetv2: A light-weight, power efficient, and general purpose CNN. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) (pp. 9190–9200). IEEE. https://doi.org/10.1109/CVPR.2019.00941
34. Menze, B. H., Jakab, A., Bauer, S., Kalpathy-Cramer, J., Farahani, K., Kirby, J., Burren, Y., Porz, N., Slotboom, J., Wiest, R., Lanczi, L., Gerstner, E., Weber, M. A., Arbel, T., Avants, B. B., Ayache, N., Buendia, P., Collins, D. L., Cordier, N., ... Van Leemput, K. (2015). The multimodal brain tumor image segmentation benchmark (BRATS). IEEE Transactions on Medical Ima Sandler ging, 34(10), 1993–2024. https://doi.org/10.1109/TMI.2014.2377694
35. Milletari, F., Navab, N., & Ahmadi, S.-A. (2016). V-Net: Fully convolutional neural networks for volumetric medical image segmentation. In Proceedings of the 2016 Fourth International Conference on 3D Vision (3DV) (pp. 565–571). IEEE. https://doi.org/10.1109/3DV.2016.79
36. Musthafa, N., Masud, M. M., Memon, Q., & Farhad, M. (2025). VISION: View-specific integrated segmentation-classification framework for accurate brain tumor detection in MRI scans. PLOS ONE, 20(5), Article e0332395. https://doi.org/10.1371/journal.pone.0332395
37. Oktay, O., Schlemper, J., Le Folgoc, L., Lee, M., Heinrich, M., Misawa, K., Mori, K., McDonagh, S., Glocker, B., & Rueckert, D. (2018). Attention U-Net: Learning where to look for the pancreas (arXiv:1804.03999). arXiv. https://doi.org/10.48550/arXiv.1804.03999
38. Ostrom, Q. T., Price, M., Neff, C., Cioffi, G., Waite, K. A., Kruchko, C., & Barnholtz-Sloan, J. S. (2022). CBTRUS statistical report: Primary brain and other central nervous system tumors diagnosed in the United States in 2015–2019. Neuro-Oncology, 24(Suppl 5), v1–v95. https://doi.org/10.1093/neuonc/noac202
39. Pacal, I., & Deveci, M. (2026). The next generation of convolutional neural networks in neuro-oncology: A comparative analysis of performance, model complexity, and efficiency in brain tumor diagnosis. Brain Network Disorders, Article 100001. https://doi.org/10.1016/j.bnd.2026.02.001
40. Paszke, A., Chaurasia, A., Kim, S., & Culurciello, E. (2016). ENet: A deep neural network architecture for real-time semantic segmentation (arXiv:1606.02147). arXiv. https://doi.org/10.48550/arXiv.1606.02147
41. Pereira, S., Pinto, A., Alves, V., & Silva, C. A. (2016). Brain tumor segmentation using convolutional neural networks in MRI images. IEEE Transactions on Medical Imaging, 35(5), 1240–1251. https://doi.org/10.1109/TMI.2016.2538465
42. Preetha, R., Priyadarsini, M. J. P., & Nisha, J. S. (2025). Brain tumor segmentation using multi-scale attention U-Net with EfficientNetB4 encoder for enhanced MRI analysis. Scientific Reports, 15(1), Article 9914. https://doi.org/10.1038/s41598-025-94267-9
43. Rashid, M. M., Singh, S., & Ahmed, A. (2026). Deep learning based methods for brain tumor segmentation in MRI: A review. DMPedia Lecture Notes in Multidisciplinary Research, 1, 76. https://digitalmanuscriptpedia.com/conferences/index.php/DMP-LNMR/article/view/76
44. Razzaghi, P., Abbasi, K., Shirazi, M., & Rashidi, S. (2022). Multimodal brain tumor detection using multimodal deep transfer learning. Applied Soft Computing, 129, Article 109631. https://doi.org/10.1016/j.asoc.2022.109631
45. Ronneberger, O., Fischer, P., & Brox, T. (2015). U-Net: Convolutional networks for biomedical image segmentation. In N. Navab, J. Hornegger, W. M. Wells, & A. Frangi (Eds.), Medical Image Computing and Computer-Assisted Intervention – MICCAI 2015 (Lecture Notes in Computer Science, Vol. 9351, pp. 234–241). Springer. https://doi.org/10.1007/978-3-319-24574-4_28
46. Sandler, M., Howard, A., Zhu, M., Zhmoginov, A., & Chen, L.-C. (2018). MobileNetV2: Inverted residuals and linear bottlenecks. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (pp. 4510–4520). IEEE. https://doi.org/10.1109/CVPR.2018.00474
47. Sobhaninia, Z., Rezaei, S., Karimi, N., Emami, A., & Samavi, S. (2020). Brain tumor segmentation by cascaded deep neural networks using multiple image scales. In Proceedings of the 2020 28th Iranian Conference on Electrical Engineering (ICEE) (pp. 1–4). IEEE. https://doi.org/10.1109/ICEE50131.2020.9260876
48. Tan, M., & Le, Q. V. (2019). EfficientNet: Rethinking model scaling for convolutional neural networks. In Proceedings of the 36th International Conference on Machine Learning (ICML) (pp. 6105–6114). PMLR. https://doi.org/10.48550/arXiv.1905.11946
49. Thahiruddin, M., & Wulandari, A. (2025). CNNs vs. hybrid transformers for brain tumor classification on the BRISC dataset. Journal of Applied Technology and Information Management, 6(1), 1–12. https://api.semanticscholar.org/CorpusID:281623477
50. Tibshirani, R. J., & Efron, B. (1993). An introduction to the bootstrap. Monographs on statistics and applied probability, 57(1), 1-436.
51. Ullah, M. S., Khan, M. A., Albarakati, H. M., Damasevicius, R., & Alsenan, S. (2024). 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, 182, Article 109183. https://doi.org/10.1016/j.compbiomed.2024.109183
52. Wang, M., & Deng, W. (2018). Deep visual domain adaptation: A survey. Neurocomputing, 312, 135–153. https://doi.org/10.1016/j.neucom.2018.05.083
53. Woo, S., Park, J., Lee, J.-Y., & Kweon, I. S. (2018). CBAM: Convolutional block attention module. In V. Ferrari, M. Hebert, C. Sminchisescu, & Y. Weiss (Eds.), Computer Vision – ECCV 2018 (Lecture Notes in Computer Science, Vol. 11211, pp. 3–19). Springer. https://doi.org/10.1007/978-3-030-01234-2_1
54. Xie, E., Wang, W., Yu, Z., Anandkumar, A., Alvarez, J. M., & Luo, P. (2021). SegFormer: Simple and efficient design for semantic segmentation with transformers. Advances in Neural Information Processing Systems, 34, 12077–12090.
55. Yu, C., Gao, C., Wang, J., Yu, G., Shen, C., & Sang, N. (2021). BiSeNet V2: Bilateral network with guided aggregation for real-time semantic segmentation. International Journal of Computer Vision, 129(11), 3051–3068.
56. Zhu, J.-Y., Park, T., Isola, P., & Efros, A. A. (2017). Unpaired image-to-image translation using cycle-consistent adversarial networks. In Proceedings of the IEEE International Conference on Computer Vision (ICCV) (pp. 2223–2232). IEEE. https://doi.org/10.1109/ICCV.2017.244