Evrişimsel Sinir Ağları Tabanlı MR Görüntülerinden Beyin Tümörü Tespiti

Yıl 2025, Cilt: 15 Sayı: 4, 1341 - 1366, 15.12.2025

Kübra Bozoğlan , Burakhan Çubukçu

https://doi.org/10.31466/kfbd.1485913

Öz

Bu çalışmanın amacı Manyetik Rezonans (MR) görüntülerini kullanarak derin öğrenme yöntemleri ile beyin tümörlerini tespit etmek ve derin öğrenme yöntemlerinin başarımlarını kıyaslamaktır. Beyin tümörleri, günümüzde artış gösteren ölümcül bir hastalık tehdidi haline gelmiştir. Hızlı büyüme eğilimleri göz önüne alındığında, erken teşhis ve doğru tedavi, hastaların hayatta kalma şansını artırmak için oldukça önem arz etmektedir. MR görüntülerinin incelenmesi, bu teşhis sürecinin temelini oluşturmaktadır. Bu çalışmada, beyin MR görüntülerinden tümörleri otomatik olarak tespit eden ve sınıflandıran, uzmanlara yardımcı olabilecek yeni bir bilgisayar destekli sistem sunulmaktadır. Geliştirilen sistem, Evrişimsel Sinir Ağları (CNN) adı verilen derin öğrenme mimarisine dayanmaktadır. Çalışmada, farklı öğrenme aktarım mimarilerinden VGG16, ResNet50 ve DenseNet121 kullanılmıştır. Bu modeller Figshare, SARTAJ ve Br35H veri setlerinin birleşiminden oluşturulan bir veri seti üzerinde test edilerek kıyaslanmıştır. Elde edilen bulgular, VGG16 mimarisinin %99,05'lik doğruluk oranıyla en yüksek başarıyı yakaladığını göstermiştir. ResNet50 mimarisi ise %73,40’lık başarı oranıyla modeller arasında en düşük başarı gösteren model olmuştur. Bu bulgular ışığında, evrişimsel sinir ağları tabanlı otomatik tümör tespit sisteminin, beyin tümörlerinin erken teşhisinde ve tedavisinde önemli bir rol oynayabileceği öngörülmektedir. Sistemin, uzman radyologların iş yükünü hafifletmesi ve teşhis sürecinin daha hızlı ve doğru bir şekilde gerçekleşmesine katkıda bulunması beklenmektedir.

Anahtar Kelimeler

Beyin Tümörü , Derin Öğrenme , CNN , VGG16 , ResNet50 , DenseNet121

Brain Tumor Detection from MR Images Based on Convolutional Neural Networks

Öz

The purpose of this study is to detect brain tumors using deep learning methods on Magnetic Resonance (MR) images and to compare the performances of various deep learning models. Brain tumors are increasingly becoming a serious threat as a fatal disease. Given their rapid growth, early diagnosis and accurate treatment are essential for improving patient survival rates. Examining MR images is fundamental to this diagnostic process. This study introduces a novel computer-aided system that automatically detects and classifies tumors in brain MR images, assisting medical experts. The developed system is based on a deep learning architecture known as Convolutional Neural Networks (CNN). The study employed different transfer learning architectures, including VGG16, ResNet50, and DenseNet121. These models were tested and compared using a dataset created from the combination of the Figshare, SARTAJ, and Br35H datasets. The results showed that the VGG16 architecture achieved the highest accuracy rate at 99.05%. In contrast, the ResNet50 architecture had the lowest performance among the models with an accuracy rate of 73.40%. Based on these findings, it is anticipated that the CNN-based automatic tumor detection system can play a significant role in the early diagnosis and treatment of brain tumors. The system is expected to alleviate the workload of expert radiologists and contribute to a faster and more accurate diagnostic process.

Anahtar Kelimeler

Brain Tumor , Deep Learning , CNN , VGG16 , ResNet50 , DenseNet121

Kaynakça

• Afshar, P., Plataniotis, K. N., & Mohammadi, A. (2019). Capsule Networks’ Interpretability for Brain Tumor Classification Via Radiomics Analyses. International Conference on Image Processing, ICIP, 2019-September, 3816-3820. https://doi.org/10.1109/ICIP.2019.8803615
• Ali, H., Khan, M. A., Tariq, U., & Zhang, Y. (2020). Brain tumor classification using deep learning: A comprehensive review. Journal of Healthcare Engineering, 2020, Article 8897181.
• Asaad Zebari, N., A. H. Alkurdi, A., B. Marqas, R., & Shamal Salih, M. (2023). Enhancing Brain Tumor Classification with Data Augmentation and DenseNet121. Academic Journal of Nawroz University, 12(4), 323-334. https://doi.org/10.25007/ajnu.v12n4a1985
• Cengil, E., Çinar, A., & Güler, Z. (2017, Ekim 30). A GPU-based convolutional neural network approach for image classification. IDAP 2017 - International Artificial Intelligence and Data Processing Symposium. https://doi.org/10.1109/IDAP.2017.8090194
• Charfi, S., Lahmyed, R., & Rangarajan, L. (2014). A Novel Approach For Brain Tumor Detection Using Neural Network. IMPACT: International Journal of Research in Engineering & Technology (IMPACT: IJRET), 2(7), 93-104. www.impactjournals.us
• Citak-Er, F., Firat, Z., Kovanlikaya, I., Ture, U., & Ozturk-Isik, E. (2018). Machine-learning in grading of gliomas based on multi-parametric magnetic resonance imaging at 3T. Computers in Biology and Medicine, 99, 154-160. https://doi.org/10.1016/J.COMPBIOMED.2018.06.009
• Daşgın, A. (2023). Covid19 Yayılımını Azaltmak İçin Yüz Maskesinin Evrişimsel Sinir Ağı Modelleri ile Tespiti. Aksaray Üniversitesi.
• Ersoy, E., & Karal, Ö. (2012). Yapay Sinir Ağları ve İnsan Beyni. İnsan ve Toplum Bilimleri Araştırmaları Dergisi, 1(2), 188-205.
• Fırıldak, K., & Talu, M. F. (2019). Evrişimsel sinir ağlarında kullanılan transfer öğrenme yaklaşımlarının incelenmesi. Computer Science, 4(2), 88-95.
• Goodfellow, I., Bengio, Y., & Courville, A. (2016). Deep learning. MIT Press.
• Gökalp, S., & Aydın, İ. (2021). Farklı Derin Sinir Ağı Modellerinin Duygu Tanımadaki Performanslarının Karşılaştırılması. Muş Alparslan Üniversitesi Mühendislik-Mimarlık Fakültesi Dergisi, 2(1), 35-43.
• Gupta, R., Koundal, D., & Goyal, N. (2023). Brain tumor classification using deep transfer learning with DenseNet-121. Frontiers in Computational Neuroscience, 17, 11719945.
• Gupta, T., Gandhi, T. K., Gupta, R. K., & Panigrahi, B. K. (2020). Classification of patients with tumor using MR FLAIR images. Pattern Recognition Letters, 139, 112-117. https://doi.org/10.1016/J.PATREC.2017.10.037
• Habibi Aghdam, H., Jahani Heravi, E., & Puig, D. (2016). A practical approach for detection and classification of traffic signs using Convolutional Neural Networks. Robotics and Autonomous Systems, 84, 97-112. https://doi.org/10.1016/J.ROBOT.2016.07.003
• Hazra, A., Dey, A., Gupta, S. K., & Ansari, A. (2017). Brain Tumor Detection Based on Segmentation using MATLAB. International Conference on Energy, Communication, Data Analytics and Soft Computing (ICECDS-2017), 425-430.
• Herek, D., & Karabulut, N. (2010). Manyetik Rezonans Görüntüleme. TTD Toraks Cerrahisi Bülteni, 1(3), 214-222.
• Huda, S., Yearwood, J., Jelinek, H. F., Hassan, M. M., Fortino, G., & Buckland, M. (2016). A Hybrid Feature Selection with Ensemble Classification for Imbalanced Healthcare Data: A Case Study for Brain Tumor Diagnosis. IEEE Access, 4, 9145-9154. https://doi.org/10.1109/ACCESS.2016.2647238
• Karuppathal, R., & Palanisamy, V. (2014). Fuzzy Based Automatic Detection and Classification Approach For MRI-Braın Tumor. ARPN Journal of Engineering and Applied Sciences, 9(12), 2770-2779. www.arpnjournals.com
• Kaur, T., & Gandhi, T. K. (2020). Automated brain image classification based on VGG-16 and transfer learning. Journal of Critical Reviews, 7(8), 742–747.
• Kingma, D. P., & Ba, J. (2015). Adam: A method for stochastic optimization. International Conference on Learning Representations (ICLR). [https://arxiv.org/abs/1412.6980](https://arxiv.org/abs/1412.6980)
• Li, Y., Yu, K., Zhang, Y., Li, X., & Yang, L. (2020). A review of training and evaluation strategies for deep learning in medical image analysis. IEEE Access, 8, 26122–26134.
• Logeswari, T., & Karnan, M. (2010). An Improved Implementation of Brain Tumor Detection Using Segmentation Based on Hierarchical Self Organizing Map. International Journal of Computer Theory and Engineering, 2(4), 591-595.
• Masoudnickparvar. (2023). Brain Tumor MRI Dataset. https://www.kaggle.com/datasets/masoudnickparvar/brain-tumor-mri-dataset
• Masoudnickparvar, A., & Subramaniam, M. (2022). Brain Tumor Classification Using Hybrid Deep Learning Techniques on Figshare, SARTAJ, and Br35H MRI Datasets. Biomedical Signal Processing and Control, 71, 103144.
• Pan, Y., Huang, W., Lin, Z., Zhu, W., Zhou, J., Wong, J., & Ding, Z. (2015). Brain tumor grading based on Neural Networks and Convolutional Neural Networks. Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS, 2015-November, 699-702. https://doi.org/10.1109/EMBC.2015.7318458
• Qassim, H., Verma, A., & Feinzimer, D. (2018). Compressed Residual-VGG16 CNN Model for Big Data Places Image Recognition. 2018 IEEE 8th Annual Computing and Communication Workshop and Conference (CCWC), 169-175.
• Saxena, P., Maheshwari, A., & Maheshwari, S. (2021). Predictive Modeling of Brain Tumor: A Deep Learning Approach. Içinde Advances in Intelligent Systems and Computing (C. 1189, ss. 275-285). Springer, Singapore. https://doi.org/10.1007/978-981-15-6067-5_30
• Seetha, J., & Raja, S. S. (2018a). Brain tumor classification using Convolutional Neural Networks. Biomedical and Pharmacology Journal, 11(3), 1457-1461. https://doi.org/10.13005/bpj/1511
• Seetha, J., & Raja, S. S. (2018b). Brain tumor classification using Convolutional Neural Networks. Biomedical and Pharmacology Journal, 11(3), 1457-1461. https://doi.org/10.13005/bpj/1511
• Shahzadi, I., Meriadeau, F., Tang, T. B., & Quyyum, A. (2019). CNN-LSTM: Cascaded framework for brain tumour classification. 2018 IEEE EMBS Conference on Biomedical Engineering and Sciences, IECBES 2018 - Proceedings, 633-637. https://doi.org/10.1109/IECBES.2018.8626704
• Soltaninejad, M., Yang, G., Lambrou, T., Allinson, N., Jones, T. L., Barrick, T. R., Howe, F. A., & Ye, X. (2017). Automated brain tumour detection and segmentation using superpixel-based extremely randomized trees in FLAIR MRI. International Journal of Computer Assisted Radiology and Surgery, 12(2), 183-203.
• Srivastava, N., Hinton, G., Krizhevsky, A., Sutskever, I., & Salakhutdinov, R. (2014). Dropout: A simple way to prevent neural networks from overfitting. Journal of Machine Learning Research, 15(1), 1929–1958.
• Tamilselvi, R., Nagaraj, A., Beham, M. P., & Sandhiya, M. B. (2020). BRAMSIT: A Database for Brain Tumor Diagnosis and Detection. 2020 Sixth International Conference on Bio Signals, Images, and Instrumentation (ICBSII), 1-5.
• Tan, M., Wang, R., Pang, R., Le, Q. V., & Fei-Fei, L. (2018). MnasNet: Platform-aware neural architecture search for mobile.
• Taşci, B. (2022). Beyin MR Görüntülerinden mRMR Tabanlı Beyin Tümörlerinin Sınıflandırması. Journal of Scientific Reports-B, 6, 1-9.
• Yamashita, R., Nishio, M., Do, R. K. G., & Togashi, K. (2018). Convolutional neural networks: An overview and application in radiology. Insights into Imaging, 9(4), 611–629.
• Wu, W., Chen, A. Y. C., Zhao, L., & Corso, J. J. (2014). Brain tumor detection and segmentation in a CRF (conditional random fields) framework with pixel-pairwise affinity and superpixel-level features. International Journal of Computer Assisted Radiology and Surgery, 9(2), 241-253. https://doi.org/10.1007/S11548-013-0922-7/TABLES/4
• Yaqub, M., Javaid, M. K., Cooper, C., & Noble, J. A. (2014). Investigation of the role of feature selection and weighted voting in random forests for 3-D volumetric segmentation. IEEE Transactions on Medical Imaging, 33(2), 258-271. https://doi.org/10.1109/TMI.2013.2284025
• Zhang, W., Itoh, K., Tanida, J., & Ichioka, Y. (1990). Parallel distributed processing model with local space-invariant interconnections and its optical architecture. Appl. Opt., 29(32), 4790-4797.