Brain Tumor Classification and Segmentation using DTCW Transform, Back Propagation Neural Network and Spatial Fuzzy C-Means Clustering
Rahul Mapari1, Sangeeta Kakarwal2, Ratnadeep Deshmukh3

1Rahul Mapari*, Computer Science and Engineering Department, Maharashtra Institute of Technology, Aurangabad, India.
2Sangeeta Kakarwal, Computer Science and Engineering Department, PES College of Engineering, Aurangabad, India.
3Ratnadeep Deshmukh, Department of Computer Science and IT, Dr. Babasabheb Ambedkar Marathwada University, Aurangabad, India.
Manuscript received on February 10, 2020. | Revised Manuscript received on February 20, 2020. | Manuscript published on March 10, 2020. | PP: 2073-2079 | Volume-9 Issue-5, March 2020. | Retrieval Number: L24891081219/2020©BEIESP | DOI: 10.35940/ijitee.L2489.039520
Open Access | Ethics and Policies | Cite | Mendeley
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (

Abstract: A novel method is presented in this paper for finding brain tumor and classifying it using the back-propagation neural network is proposed. Spatial Fuzzy C-Means clustering is utilized for the segmentation of image to identify the influenced area of brain MRI picture. Automated detection of tumors in brain MR images is urgent in many diagnosis processes. Because of noise, blurred edges, the detection, and classification of brain tumor are very difficult. This paper presents one programmed brain tumor identification strategy to expand the exactness and yield and diminishing the determination time. The objective is ordering the tissues to three classes of typical, start and malignant. The size and the location tumor is very important for doctors for defining the treatment of tumor. The proposed determination strategy comprises of four phases, pre-processing of MR images, feature extraction, and classification. The features are extracted using Dual-Tree Complex wavelet transformation (DTCWT). Back Propagation Neural Network (BPN) is employed for finding brain tumor in MRI images. In the last stage, a productive scheme is proposed for segmentation depends on the Spatial Fuzzy C-Means Clustering. The performance analysis clearly proves that the proposed scheme is more efficient and the efficiency of the scheme is measured with sensitivity and specificity. The evaluation is performed on the image data set of 15 MRI images of brain. 
Keywords: Spatial Fuzzy C-Means Clustering, Back Propagation Neural Network, MRI, Dual tree complex wavelet transformation.
Scope of the Article: Clustering