TY  - JOUR
AU  - Alam, Nur
AU  - Zhu, Yutong
AU  - Shao, Jiaqi
AU  - Usman, Muhammad
AU  - Fayaz, Muhammad
PY  - 2025
DA  - 2025/07/09
TI  - A Novel Deep Learning Framework for Brain Tumor Classification Using Improved Swin Transformer V2
JO  - ICCK Transactions on Advanced Computing and Systems
T2  - ICCK Transactions on Advanced Computing and Systems
JF  - ICCK Transactions on Advanced Computing and Systems
VL  - 1
IS  - 3
SP  - 154
EP  - 163
DO  - 10.62762/TACS.2025.807755
UR  - https://www.icck.org/article/abs/TACS.2025.807755
KW  - brain tumor classification
KW  - deep learning
KW  - MRI scans
KW  - computational efficiency
KW  - medical image analysis
AB  - Brain tumors pose a serious threat to global health, making accurate and early detection essential for effective treatment planning. While Magnetic Resonance Imaging (MRI) is widely used for diagnosis, manual interpretation is time-consuming and subject to error. This has prompted increasing use of deep learning for automated tumor classification. We propose a novel framework based on the Swin Transformer V2 architecture for classifying brain tumors in MRI scans into glioma, meningioma, pituitary tumor, and non-tumor categories. The design introduces two core innovations: a Dual-Branch Down-sampling module and an Enhanced Attention Mechanism, which improve multi-scale feature representation and computational efficiency. Using a dataset of 7,023 grayscale MRI images, the proposed model achieved an accuracy of 98.97%, outperforming ResNet50 (90.39%) and DenseNet121 (93.20%). It maintained precision, recall, and F1-scores above 98% across all classes and showed improved training efficiency. These results demonstrate the model’s potential as a robust and efficient diagnostic support system for brain tumor classification.
SN  - 3068-7969
PB  - Institute of Central Computation and Knowledge
LA  - English
ER  - 

@article{Alam2025A,
  author = {Nur Alam and Yutong Zhu and Jiaqi Shao and Muhammad Usman and Muhammad Fayaz},
  title = {A Novel Deep Learning Framework for Brain Tumor Classification Using Improved Swin Transformer V2},
  journal = {ICCK Transactions on Advanced Computing and Systems},
  year = {2025},
  volume = {1},
  number = {3},
  pages = {154-163},
  doi = {10.62762/TACS.2025.807755},
  url = {https://www.icck.org/article/abs/TACS.2025.807755},
  abstract = {Brain tumors pose a serious threat to global health, making accurate and early detection essential for effective treatment planning. While Magnetic Resonance Imaging (MRI) is widely used for diagnosis, manual interpretation is time-consuming and subject to error. This has prompted increasing use of deep learning for automated tumor classification. We propose a novel framework based on the Swin Transformer V2 architecture for classifying brain tumors in MRI scans into glioma, meningioma, pituitary tumor, and non-tumor categories. The design introduces two core innovations: a Dual-Branch Down-sampling module and an Enhanced Attention Mechanism, which improve multi-scale feature representation and computational efficiency. Using a dataset of 7,023 grayscale MRI images, the proposed model achieved an accuracy of 98.97\%, outperforming ResNet50 (90.39\%) and DenseNet121 (93.20\%). It maintained precision, recall, and F1-scores above 98\% across all classes and showed improved training efficiency. These results demonstrate the model’s potential as a robust and efficient diagnostic support system for brain tumor classification.},
  keywords = {brain tumor classification, deep learning, MRI scans, computational efficiency, medical image analysis},
  issn = {3068-7969},
  publisher = {Institute of Central Computation and Knowledge}
}