Araştırma Makalesi

Classification of Brain MRI Images with Vision Transformer: Improving Performance with New Layers and Parameter Optimization

Cilt: 9 Sayı: 2026 15 Eylül 2026
PDF İndir
EN TR

Classification of Brain MRI Images with Vision Transformer: Improving Performance with New Layers and Parameter Optimization

Öz

Recent advances in deep learning have revolutionized fields such as robotics, healthcare, and natural language processing. The Vision Transformer (ViT), which operates on self-attention block, has emerged as an alternative to Convolutional Neural Networks. In this study, an improved ViT architecture is proposed for the categorization of brain MRI images as tumorous and non-tumorous. While the standard ViT backbone is utilized for feature encoding, the conventional classification head has been replaced with a multi-stage architecture comprising sequential fully connected layers with 20, 4, and 2 neurons, integrated with Sigmoid and Linear activation functions. This architectural modification, which has been constructed via Greedy Search approach, aims to refine the feature mapping process and enhance the model's sensitivity towards pathological patterns in medical images. To evaluate the performance, a brain MRI dataset has been split into validation, test and training sets, and data augmentation was performed to prevent overfitting. Standard ViT, Swin Transformer, EfficientNet, ResNet and the proposed ViT models have been trained using an ablation technique to optimize network parameters. For the performance analysis, the models have been evaluated using the metrics such as Accuracy, F1-Score, Precision, Recall, AUC values and ROC curves. The results of this work indicate that the proposed ViT’s head structure improves classification success compared to the standard ViT architecture and the other models. Consequently, by redesigning the classification head and optimizing the network, a 13% improvement has been achieved, reaching macro F1-Score of 95.3% for brain image dataset 1 and macro F1-Score of 99.3% for brain image dataset 2.

Anahtar Kelimeler

Kaynakça

  1. Altun, S., Alkan, A., 2023. LSTM-based deep learning application in brain tumor detection using MR spectroscopy. Journal of the Faculty of Engineering and Architecture of Gazi University, 38(2), 1193–1202. https://doi.org/10.17341/gazimmfd.1069632
  2. Aslan, E., 2024. LSTM-ESA Hibrit Modeli ile MR Görüntülerinden Beyin Tümörünün Sınıflandırılması. Adıyaman Üniversitesi Mühendislik Bilimleri Dergisi, 22(2024), 63–81. https://doi.org/10.54365/adyumbd.1391157
  3. Aslan, M., 2022. Deep Learning-Based Automatic Detection. Fırat Üniversitesi Mühendislik Bilimleri Dergisi, 34(1), 399–407. https://doi.org/10.35234/fumbd.1039825
  4. Aydın, E., Demir, F., Şengür, A., 2024. MR görüntülerinden beyin tümörünün A-ESA tabanlı bir yaklaşımla otomatik sınıflandırılması. International Journal of Pure and Applied Sciences, 10(2), 325–341. https://doi.org/10.29132/ijpas.1398148
  5. 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. European Conference on Computer Vision (ECCV) Workshops, pp. 205–218.
  6. Carion, N., Massa, F., Synnaeve, G., Usunier, N., Kirillov, A., Zagoruyko, S., 2020. End-to-end object detection with transformers. European Conference on Computer Vision (ECCV), pp. 213–231.
  7. Chen, C.-F., Chen, Y., Hebert, M., Wang, H., 2021a. Cross attention in vision transformer. arXiv preprint, arXiv:2106.05786.
  8. Chen, C. F., Fan, Q., Panda, R., 2021b. CrossViT: Cross-Attention Multi-Scale Vision Transformer for Image Classification. IEEE/CVF, International Conference on Computer Vision (ICCV), Montreal, QC, Canada, 2021, pp. 347–356.

Ayrıntılar

Birincil Dil

İngilizce

Konular

Derin Öğrenme, Nöral Ağlar, Makine Öğrenme (Diğer)

Bölüm

Araştırma Makalesi

Yayımlanma Tarihi

15 Eylül 2026

Gönderilme Tarihi

7 Eylül 2025

Kabul Tarihi

9 Nisan 2026

Yayımlandığı Sayı

Yıl 2026 Cilt: 9 Sayı: 2026

Kaynak Göster

APA
Aşlıyan, R., Gör, İ., & Günel, K. (2026). Classification of Brain MRI Images with Vision Transformer: Improving Performance with New Layers and Parameter Optimization. Journal of Intelligent Systems: Theory and Applications, 9(2026), 1-18. https://doi.org/10.38016/jista.1779498
AMA
1.Aşlıyan R, Gör İ, Günel K. Classification of Brain MRI Images with Vision Transformer: Improving Performance with New Layers and Parameter Optimization. jista. 2026;9(2026):1-18. doi:10.38016/jista.1779498
Chicago
Aşlıyan, Rıfat, İclal Gör, ve Korhan Günel. 2026. “Classification of Brain MRI Images with Vision Transformer: Improving Performance with New Layers and Parameter Optimization”. Journal of Intelligent Systems: Theory and Applications 9 (2026): 1-18. https://doi.org/10.38016/jista.1779498.
EndNote
Aşlıyan R, Gör İ, Günel K (01 Eylül 2026) Classification of Brain MRI Images with Vision Transformer: Improving Performance with New Layers and Parameter Optimization. Journal of Intelligent Systems: Theory and Applications 9 2026 1–18.
IEEE
[1]R. Aşlıyan, İ. Gör, ve K. Günel, “Classification of Brain MRI Images with Vision Transformer: Improving Performance with New Layers and Parameter Optimization”, jista, c. 9, sy 2026, ss. 1–18, Eyl. 2026, doi: 10.38016/jista.1779498.
ISNAD
Aşlıyan, Rıfat - Gör, İclal - Günel, Korhan. “Classification of Brain MRI Images with Vision Transformer: Improving Performance with New Layers and Parameter Optimization”. Journal of Intelligent Systems: Theory and Applications 9/2026 (01 Eylül 2026): 1-18. https://doi.org/10.38016/jista.1779498.
JAMA
1.Aşlıyan R, Gör İ, Günel K. Classification of Brain MRI Images with Vision Transformer: Improving Performance with New Layers and Parameter Optimization. jista. 2026;9:1–18.
MLA
Aşlıyan, Rıfat, vd. “Classification of Brain MRI Images with Vision Transformer: Improving Performance with New Layers and Parameter Optimization”. Journal of Intelligent Systems: Theory and Applications, c. 9, sy 2026, Eylül 2026, ss. 1-18, doi:10.38016/jista.1779498.
Vancouver
1.Rıfat Aşlıyan, İclal Gör, Korhan Günel. Classification of Brain MRI Images with Vision Transformer: Improving Performance with New Layers and Parameter Optimization. jista. 01 Eylül 2026;9(2026):1-18. doi:10.38016/jista.1779498

Zeki Sistemler Teori ve Uygulamaları Dergisi