Otsu ve Rocchio Metotlarıyla Beyin Tümörü Tespiti

Beynimiz, kafatası içinde bulunan ve merkezi sinir sisteminin en karmaşık organıdır. En karmaşık organımız olan beynimiz vücudumuzun tüm fonksiyonlarını kontrol eder. Beyin tümörleri, beyindeki hücrelerin kontrolsüz bir şekilde büyümesiyle ortaya çıkar. Beyin tümörlerini erken teşhis etmek genellikle daha fazla tedavi imkanı sağlar. Beyin tümörlerinin teşhisinde en çok manyetik rezonans görüntülemeden yararlanılır. Bu çalışmada, Otsu ve Rocchio metotları kullanılarak bölütleme sistemleri geliştirildi. Beyin MR görüntüsünü girdi olarak alan, kafatası ayırma, ön-işleme, segmentasyon ve art-işleme işlemlerini gerçekleştiren sistemler tasarlandı ve uygulandı. Ön-işlemeden önce, kafatası bölgesi beyin MR görüntü veri setindeki görüntülerden çıkarılır. Ön-işlemede çeşitli filtreleme ve morfolojik tekniklerle beyin görüntülerinin kalitesi artırılır ve görüntülerin gürültüsü ortadan kaldırılır. Bölütlemede ise Otsu metodu ile eşik değerlerinin belirlenmesi ile beyindeki tümörlü bölge tespit edilir. Art-işlemede, beyin tümörü veri setinin eğitim veri seti kullanılarak Rocchio sınıflandırıcı metodu eğitilir ve belirlenen tümörlü bölgelerin en uygun olanı bulunur. Böylece en doğru tümörlü bölge tespit edilerek optimize edilmiş olur. Test safhasında, sistemlerin başarılarını değerlendirmek amacıyla doğruluk, kesinlik ve seçicilik metrikleriyle sistemlerin başarıları karşılaştırılmıştır. Art-işleme sonucunda başarının önemli ölçüde arttığı görülmüştür.

Brain Tumour Detection with Otsu and Rocchio Methods

Our brain is the most complicated organ of the central nervous system, located inside the skull. Our most complex organ, our brain, controls all the functions of our body. Brain tumours occur when cells in the brain grow uncontrollably. Detecting brain tumours early usually provides more treatment opportunities. Magnetic resonance imaging is mostly used in the diagnosis of brain tumours. In this study, segmentation systems were developed using Otsu and Rocchio methods. Systems that take brain MR images as input and perform skull separation, pre-processing, segmentation and post-processing have been designed and implemented. Before pre-processing, the skull region is extracted from the images in the brain MR image dataset. In pre-processing, the quality of brain images is improved and the noise of the images is eliminated by various filtering and morphological techniques. In segmentation, the tumour region in the brain is determined by detecting the threshold values with the Otsu method. In post-processing, the Rocchio classifier method is trained using the training dataset of the brain tumour dataset and the most suitable one of the determined tumour regions is found. Thus, the most accurate tumour region is detected and optimized. In the test phase, the success of the systems was compared with the accuracy, precision and selectivity metrics to evaluate the success of the systems. As a result of post-processing, it was observed that success of the system is increased significantly.

___

  • AlAzawee, W. S. (1995). Computer-Aided Brain Tumor Edge Extraction Using Morphological Operations. MSc Thesis, Western Michigan University, Kalamazoo, USA.
  • Ali, S. M., Abood, L. K. & Abdoon, R. S. (2013). Brain tumor extraction in MRI images using clustering and morphological operations techniques. Int J Geograph Inform Syst Appl Remote Sens, vol. 4(1).
  • Aşlıyan, R. & Atbakan, İ. (2020). AutomatIc BraIn Tumor SegmentatIon wIth K-Means, Fuzzy C-Means, Self-Organizing Map and Otsu Methods. Journal of Selcuk-Technic, 267-281.
  • Ayachi, R. & Amor, N. (2009). Brain tumor segmentation using support vector machines. European Conference on Symbolic and Quantitative Approaches to Reasoning with Uncertainty, 736-747.
  • Batista, J. & Kitney, R. (1995). Extraction of tumors from MR images of the brain by texture and clustering. Conference on Image Analysis and Processing, 235-240.
  • Brain Tumor Dataset. (2019) Website. [Online]. Available:https://figshare.com/articles/brain_tumor_dataset /1512427.
  • Dharshini, R. & Hemanandhini, S. (2016). Brain tumor segmentation based on Self Organising Map and Discrete Wavelet Transform. International Conference on Computer Communication and Informatics (ICCCI), Coimbatore, India.
  • Hastie, T., Tibshirani, R. & Friedman, J. (2001). The Elements of Statistical Learning. Springer Series in Statistics. Springer Verlag, New York.
  • Juang, L. & Wu, M. (2010). MRI brain lesion image detection based on color-converted k-means clustering segmentation. Measurement, 43(7), 941-949.
  • Karaddi, S., Babu, P. & Reddy, R. (2018). Detection of Brain Tumor Using Otsu-Region Based Method of Segmentation. International Conference on Computing Methodologies and Communication (ICCMC), 128-134.
  • Kathirvel, R. & Batri, K. (2017). Detection and diagnosis of meningioma brain tumor using ANFIS classifier. Imaging Syst Technol, 27, 187-192.
  • Park, H., Jeonand, M. & Rosen, J. B. (2003). Lower dimensional representation of text data based on centroids and least squares. BIT, 43(2), 1–22.
  • Pereira, S., Pinto, A., Alves, V. and Silva, C. A. (2016). Brain tumor segmentation using convolutional neural networks in MRI images. IEEE Trans Med Imag, 35, 1240-1251.
  • Prakash, R. M. & Kumari, R. S. S. (2016). Fuzzy C means integrated with spatial information and contrast enhancement for segmentation of MR brain images. Int J Imag Sys Techol., 26, 116-123.
  • Saraswathi, D., Priya, B. L. & Lakshmi, R. P. (2019). Brain Tumor Segmentation and Classification using Self Organizing Map. International Conference on System, Computation, Automation and Networking (ICSCAN), Pondicherry, India.
  • Sharma, M. & Mukharjee, S. (2013). Brain Tumor Segmentation using Genetic Algorithm and Artificial Neural Network Fuzzy Inference System ANFIS. Advances in Intelligent Systems and Computing, Springer, Berlin, Heidelberg, 177, 329-339.
  • Wu, M., Lin, C. & Chang, C. (2007). Brain tumor detection using color-based k-means clustering segmentation. Conference on Intelligent Information Hiding and Multimedia Signal Processing, 245-250.
  • Zhang, J., Ma, K., Er, M. & Chong, V. (2004). Tumor segmentation from magnetic resonance imaging by learning via one-class support vector machine. Workshop on Advanced Image Technology, 207-211.