Categorical and Binary Brain Tumor Classification Using Transfer Learning Techniques

Abstract

The quality and length of life may be affected by brain tumors, which are created when cells in the head region proliferate out of control. Patients with misdiagnosed or late-diagnosed brain tumors and untreated patients have a lower chance of survival. Images obtained from MR imaging equipment are typically used to diagnose brain cancers. Given the rising number of patients and the high doctor density, computer-assisted techniques are particularly helpful in the diagnosis and categorization of brain tumors. In this study, transfer learning techniques were used to classify brain tumors from MRI data. In the study, a 4-class dataset made up of glioma, meningioma, pituitary, and no-tumor was used in addition to a binary data set of tumor and no-tumor. Repetitive and unneeded regions in the images were eliminated by applying image preprocessing techniques to the datasets. Following that, classification was performed using the EfficientNet, XceptionNet, and CoAtNet models, which modified the last layer and used the weight values of the models trained on very large datasets (imagenet). As a result, show that CoAtNet performed best in multiclassification validation accuracy (98.26) and EfficientNet in binary classification (99.98). When compared to high-success studies with similar datasets, it was observed that the success metrics were quite close to those of these studies.

Keywords

• transfer learning
• coatnet
• efficientnet
• xceptionnet
• classification
• brain tumor

Transfer Öğrenme Teknikleri Kullanılarak Kategorik ve İkili Beyin Tümörü Sınıflandırması

Abstract

Beyin bölgesindeki hücrelerin kontrol dışı çoğalmasıyla oluşan beyin tümörleri yaşam kalitesini ve uzunluğunu etkileyebilir. Yanlış veya geç tanı konulan beyin tümörlü hastaların veya tedavi edilmeyen hastaların hayatta kalma şansı daha düşüktür. MR görüntüleme ekipmanından elde edilen görüntüler tipik olarak beyin kanserlerini teşhis etmek için kullanılır. Artan hasta sayısı ve yüksek doktor yoğunluğu göz önüne alındığında, bilgisayar destekli teknikler özellikle beyin tümörlerinin teşhisinde ve sınıflandırılmasında yardımcı olmaktadır. Bu çalışmada, MRI verilerinden beyin tümörlerini sınıflandırmak için transfer öğrenme teknikleri kullanılmıştır. Çalışmada tümörlü ve tümörsüz ikili veri setine ek olarak glioma, menenjiyom, hipofiz ve tümörsüz görüntülerinden oluşan 4 sınıflı bir veri seti kullanılmıştır. Veri setlerine görüntü ön işleme teknikleri uygulanarak görüntülerdeki tekrarlı bölgeler ve gereksiz bölgeler ortadan kaldırılmıştır. Ardından son katmanı modifiye edilen EfficientNet, XceptionNet ve CoAtNet modelleri ile bu modellerin çok büyük veri setleri (imagenet) üzerinde eğitilmiş ağırlık değerleri kullanılarak sınıflandırma yapılmıştır. Sonuç olarak, CoAtNet'in çoklu sınıflandırma validasyon doğruluğunda (98.26) ve EfficientNet'in ikili sınıflandırmada (99.98) en iyi performansı gösterdiği görülmüştür. Benzer veri setlerine sahip yüksek başarılı çalışmalarla karşılaştırıldığında, başarı metriklerinin bu çalışmalara oldukça yakın olduğu görülmüştür.

Keywords

• transfer öğrenme
• coatnet
• efficientnet
• xceptionnet
• sınıflandırma
• beyin tümörü

Thanks

CAIAC'22 de sunulmuştur. Konferans komitesine teşekkür ederiz.

References

1. [1] Louis D.N., Perry A., Reifenberger G., Deimling A.V., Figarella-Branger D., Cavenee W.K., Ohgaki H., Wiestler O.D., Kleihues P., and Ellison D.W., The 2016 World Health Organization classification of tumors of the central nervous system: A summary, Acta Neuropathol., 131 (2016) 803–820.
2. [2] De Angelis L.M., Brain tumors, New England J. Med, 344(2): (2001) 114-123. https://doi.org/10.1056/NEJM2001011134 40207.
3. [3] Gladson C.L., Prayson R.A., Liu W., The pathobiology of glioma tumors, Annual Review of Pathology: Mechanisms of Disease, 5 (2010) 33-50. https://doi.org/10.1146 /annurev-pathol-121808-102109.
4. [4] Mehrotra R., Ansari M.A., Agrawal R., Anand R.S., A Transfer Learning approach for AI-based classification of brain tumors, Mach. Learn. Appl., 2 (2020) 10–19.
5. [5] Pereira S., Pinto A., Alves V., and Silva C.A., Brain tumor segmentation using convolutional neural networks in MRI images, IEEE Trans. Med. Imaging, 35 (2018) 1240–1251.
6. [6] Dundar T.T., Yurtsever I., Pehlivanoglu M.K., Yildiz U., Eker A., Demir M.A., Mutluer A.S., Tektaş R., Kazan M.S., Kitis S., Gokoglu A., Dogan I., Duru N., Machine Learning-Based Surgical Planning for Neurosurgery: Artificial Intelligent Approaches to the Cranium, Front Surg., 9 (2022) 863633. doi: 10.3389/fsurg.2022.863633. PMID: 35574559; PMCID: PMC9099011.
7. [7] Hamada A., Br35H Brain Tumor Detection 2020 Dataset, Available online: https://www.kaggle.com/ahmedhamada0 /braintumor-detection.
8. [8] Sartaj,“Brain Tumor Classification (MRI) Dataset”, Available online: https://www.kaggle.com/datasets/sartajbhuvaji /brain-tumor-classification-mri.

9. [9] Kushwaha V., Maidamwar P., BTFCNN: Design of a brain tumor classification model using fused convolutional neural networks, 2022 10th International Conference on Emerging Trends in Engineering and Technology-Signal and Information Processing (ICETET-SIP-22), (2022) 1-6, doi: 10.1109/ICETET-SIP-2254415.2022.9791734.
10. [10] Kang J., Ullah Z., and Gwak J., MRI-Based Brain Tumor Classification Using Ensemble of Deep Features and Machine Learning Classifiers, Sensors (Basel), 21(6) (2021) 2222. doi: 10.3390/s21062222.
11. [11] Amran G.A., Alsharam M.S., Blajam A.O.A., Hasan A.A., Alfaifi M.Y., Amran M.H., Gumaei A., Eldin S.M., Brain Tumor Classification and Detection Using Hybrid Deep Tumor Network, Electronics, 11(21) (2022) 3457. https://doi.org/10.3390/electronics11213457.
12. [12] Özdem K., Özkaya Ç., Atay Y., Çeltikçi E., Börcek A., Demirezen U., and Sağıroğlu Ş., A GA-Based CNN Model for Brain Tumor Classification, 2022 7th International Conference on Computer Science and Engineering (UBMK), (2022) 418-423, doi: 10.1109/UBMK55850.2022.9919461
13. [13] Papageorgiou V., Brain Tumor Detection Based on Features Extracted and Classified Using a Low- Complexity Neural Network”, Traitement du Signal., 38 (2021) 547-554. doi:10.18280/ts.380302.
14. [14] Sert E., Ӧzyurt F., and Doğanteklin A., A new approach for brain tumor diagnosis system: Single image super resolution based maximum fuzzy entropy segmentation and convolutional neural network, Medical Hypotheses, 133 (2019) 109413.
15. [15] Ӧzyurt F., Sert E., Avci E., and Doğanteklin E., Brain tumor detection on Convolutional Neural Networks with neutrosophic expert maximum fuzzy sure entropy, Measurement, 147 (2019) 106830. https://doi.org/10.1016 /j.measurement.2019.07.058.
16. [16] Sajja V.R., and Kalluri H.K., Classification of brain tumors using convolutional neural networks over various SVM methods, Ingénierie des Systèmes d’Information, 25(4): (2020) 489-495. https://doi.org/10.18280/isi.250412
17. [17] Mittal A., Kumar D., AiCNNs (artificially-integrated convolutional neural networks) for brain tumor prediction, EAI Endorse Transactions on Pervasive Health and Technology, 5 (2019) 1-18. http://doi.org/10.4108/eai.12-2-2019.161976.
18. [18] Sultan H.H., Salem N.M., Al-Atabany W., Multi-classification on brain tumor images using deep neural network, IEEE Access, 7 (2019) 69215-69225.
19. [19] Ari A., Alcin O.F., Hanbay D., Brain MR image classification based on deep features by using extreme learning machines, Biomedical Journal of Scientific and Technical Research, 25: (2020) 19137-19144.
20. [20] Zhao L., Jia K., Multiscale CNNs for brain tumor segmentation and diagnosis, Computational and Mathematical Methods in Medicine, (2016) 1-17. https://doi.org/10.1155/2016/8356294A.
21. [21] Hamada A., Br35H: Brain Tumor Detection, (2020), https://www.kaggle.com/datasets/ahmedhamada0/brain-tumor-detection.
22. [22] Sartaj, Classify MRI images into four classes, (2020), https://www.kaggle.com/datasets/sartajbhuvaji/brain-tumor-classification-mri.
23. [23] Szegedy C., Liu W., Jia Y., Sermanet P., Reed S., Anguelov D., Rabinovich A., Going deeper with convolutions, In Proceedings of the IEEE conference on computer vision and pattern recognition, (2015) 1-9.
24. [24] Lumini A., Nanni L., Deep learning and transfer learning features for plankton classification, Ecological informatics, 51 (2019) 33-43.
25. [25] Dandıl E., Serin Z. Derin, Sinir Ağları Kullanarak Histopatolojik Görüntülerde Meme Kanseri Tespiti. Avrupa Bilim ve Teknoloji Dergisi, (2020) 451-463.