Abstract
For MRI brain tumor image segmentation, it is necessary to have high real-time and accurate edge localization. Therefore, we propose a yolov8n based on dynamic serpentine convolution and multi-feature attention method (DMA-YOLOV8n). The method combines dynamic serpentine convolution and multi-feature attention mechanism, which can effectively adapt to different brain tumor tissue edge morphology changes and more accurately segmented to obtain brain tumor and locate its edge position. First, dynamic serpentine convolution is used to replace standard convolution in C2f. module. Then, drawing on the idea of skip connection in U-Net model, multi- feature fusion is used to connect multilayer sampling information to retain more feature details and improve edge segmentation accuracy. Finally, dual attention mechanism is added to multilayer feature fusion to pay more attention to brain tumor tissue. DMA-YOLOV8n is applied to brain MRI images from Kaggle_3M dataset. Experimental results show the method has mAP50: 0.806 and mAP50:95: 0.490.
Similar content being viewed by others
Data availability
The datasets generated and/or analysed during the current study are available in the repository, https://aistudio.baidu.com/datasetdetail/205256 and https://pan.baidu.com/s/13d_KSRxGe3qUtEsg7jupVQ?pwd=j56k and https://pan.baidu.com/s/13d_KSRxGe3qUtEsg7jupVQ?pwd=j56k
Abbreviations
- MRI:
-
Magnetic resonance imaging
- SIFT:
-
Scale-invariant feature transformation
- SURF:
-
Speeded up robust features
- CNN:
-
Convolutional neural network
- FCN:
-
Fully convolutional networks
- MFA:
-
Moth flame algorithm
- SPPF:
-
Spatial pyramid pooling–fast
- FPN:
-
Feature pyramid network
- PAN:
-
Path aggregation network
References
Obayya, M. et al. A novel U-net model for brain tumor segmentation from MRI images. Alex. Eng. J. https://doi.org/10.1016/J.AEJ.2025.04.051 (2025).
Rogers, C. M., Jones, P. S. & Weinberg, J. S. Intraoperative MRI for brain tumors. J. Neurooncol. 151, 479–490 (2021).
ZebariA, N. et al. A deep learning fusion model for accurate classification of brain tumours in magnetic resonance images. CAAI Trans. Intell. Technol. 9(4), 790–804 (2024).
Dheepa, G. & Chithra, P. L. An efficient encoder-decoder CNN for brain tumor segmentation in MRI images. IETE J. Res. 69(12), 8647–8658 (2023).
Shaheema, B. & Muppalaneni, B. N. Multimodal brain image segmentation: A recent review, challenges and future perspectives. Multim. Tools Appl. https://doi.org/10.1007/S11042-025-21068-7 (2025) ((prepublish)).
Gao, Y. et al. Medical image segmentation: A comprehensive review of deep learning-based methods. Tomography 11(5), 52–52. https://doi.org/10.3390/TOMOGRAPHY11050052 (2025).
Pandey, M. & Gupta, A. Tumorous kidney segmentation in abdominal CT images using active contour and 3D-UNet. Ir. J. Med. Sci. 192(3), 1401–1409 (2023).
Pang, Z. F. et al. Adaptive weighted curvature-based active contour for ultrasonic and 3T/5T MR image segmentation. Signal Process 205, 108881 (2023).
Hosny, K. M. et al. Multilevel segmentation of 2D and volumetric medical images using hybrid coronavirus optimization algorithm. Comput. Biol. Med. 150, 106003 (2022).
Katherine, Rulaningtyas, R., Ain, K. CT scan image segmentation based on hounsfield unit values using Otsu thresholding method. in Journal of Physics: Conference Series. Bristol, United Kingdom: IOP Publishing, 012080 (2021).
Zhang, Q., Min, B., Hang, Y., Chen, H. & Qiu, J. A full-scale lung image segmentation algorithm based on hybrid skip connection and attention mechanism. Sci. Rep. 14(1), 23233 (2024).
Huda, N. & Mahamud, K. R. K. CNN-based image segmentation approach in brain tumor classification: A review †. Eng. Proc. 84(1), 66–66. https://doi.org/10.3390/ENGPROC2025084066 (2025).
Kadry, S. et al. CNN segmentation of skin melanoma in pre-processed dermoscopy images. Proce. Comput. Sci. 235, 2775–2782 (2024).
Li, B. et al. TransRNetFuse: A highly accurate and precise boundary FCN-transformer feature integration for medical image segmentation. Complex Intell. Syst. 11(5), 213–213. https://doi.org/10.1007/S40747-025-01847-3 (2025).
Ledan, Q. et al. Automatic segmentation method using FCN with multi-scale dilated convolution for medical ultrasound image. Vis. Comput. 39(11), 5953–5969 (2022).
Vu, A. T. et al. A modified Bi-directional convolutional U-Net (BCDU-Net) neural network approach for lung CT image segmentation. J. Biomimet. Biomater. Biomed. Eng. https://doi.org/10.4028/P-CDM8F6 (2025).
Su, R. et al. MSU-net: Multi-scale U-net for 2D medical image segmentation. Front. Genet. 12, 639930 (2021).
Yang, T. et al. An approach for plant leaf image segmentation based on YOLOV8 and the improved DEEPLABV3+. Plants 12(19), 3438 (2023).
Nimmagadda, R. & Devi, K. P. A deep learning approach for brain tumour classification and detection in MRI images using YOLOv7. Front. Oncol. https://doi.org/10.3389/FONC.2025.1508326 (2025).
Haseeb, A. N. et al. Advance brain tumor segmentation using feature fusion methods with deep U-Net model with CNN for MRI data. J. King Saud Univ. Comput. Inf. Sci. 35(9), 101793 (2023).
Cai, Z., Zhou, K. & Liao, Z. A systematic review of YOLO-based object detection in medical imaging: Advances, challenges, and future directions. Comput. Mater. Contin. 85(2), 2255–2303. https://doi.org/10.32604/CMC.2025.067994 (2025).
Lin, Z. et al. Research on improved YOLOv8 algorithm for insulator defect detection. J. Real-Time Image Process. 21(1), 22 (2024).
Chen, L., Sun, Q. & Wang, F. Attention-adaptive and deformable convolutional modules for dynamic scene deblurring. Inform. Sci. 546, 368–377 (2021).
Tong, S. et al. Hybrid attention mechanism of feature fusion for medical image segmentation. IET Image Process. 18(1), 77–87 (2023).
Yawu, Z. et al. WRANet: Wavelet integrated residual attention U-Net network for medical image segmentation. Complex Intell. Syst. 9(6), 11–13 (2023).
Nannan, Z. H. A. O. & Feichen, G. A. O. Improved YOLOv8-based algorithm for instance segmentation in traffic scenes. Comput. Eng. https://doi.org/10.19678/j.issn.1000-3428.0068677 (2025).
Li, W. et al. BCE YOLOv8: A novel YOLO model for brain tumor instance segmentation. Biomed. Phys. Eng. Express 11(5), 055035–055035. https://doi.org/10.1088/2057-1976/ADF8F1 (2025).
Tan, X. & Zhao, J. Enhancing YOLOv8 for improved instance segmentation of automotive surface damage. Comput. Eng. Appl. 60(14), 197–208 (2024).
Cui, H. et al. Improved YOLOv8-seg for laryngeal structure recognition in medical images. Am. J. Transl. Res. 17(5), 3293–3306. https://doi.org/10.62347/BIHI3707 (2025).
Lei, W. A. N. G. et al. RCSA-YOLO: Improved SAR Ship instance segmentation of YOLOv8. Comput. Eng. Appl. 60(18), 103–113 (2024).
Acknowledgements
We would like to thank everyone who has contributed to this article. We would also like to thank the anonymous reviewers and editors for their helpful suggestions and comments.
Funding
This study was funded by the program of the General Program of Nantong Science and Technology Bureau in 2024 under Grant No. MSZ2024110, the Program of the Nantong Institute of Technology Doctoral Research Start-up Fund No.2025XKB30.
Author information
Authors and Affiliations
Contributions
Yiliu Hang: Writing–review & editing, Writing–original draft, Validation, Resources, Methodology, Investigation, Project Administration, Funding Acquisition. Qiong Zhang: Writing–review & editing, Methodology, Data Curation, Formal Analysis, Software. Li Li: Formal Analysis, Project Administration, Writing–review & editing. Chunhua Lin: Writing–review & editing,Resources, Methodology, Formal Analysis.
Corresponding author
Ethics declarations
Competing interests
The authors declare no competing interests.
Ethical approval
Dataset belongs to public databases. The patients involved in the database have obtained ethical approval. Users can download relevant data for free for research and publish relevant articles. Our study is based on open source data, so there are no ethical issues and other conflicts of interest.
Additional information
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.
About this article
Cite this article
Hang, Y., Zhang, Q., Li, L. et al. Yolov8n based on dynamic serpentine convolution and multi-feature attention for MRI brain cranial tumor segmentation. Sci Rep (2026). https://doi.org/10.1038/s41598-026-42502-2
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41598-026-42502-2
