A K-Nearest Neighbor Technique for Brain Tumor Segmentation Using Minkowski Distance

作者：Xie, Xiaozhen^*

来源：Journal of Medical Imaging and Health Informatics, 2018, 8(2): 180-185.

DOI：10.1166/jmihi.2018.2285

• Image Segmentation
• Systematic Modelling
• Image Synthetic Analysis
• Iteration
• Minkowski Distance
• K-NN Clustering

Image segmentation is the process of partitioning a digital image into multiple segments. Image segmentation has been gaining widespread significance especially in the area of medical imaging for early detection and diagnosis of tumors. Brain Tumor Segmentation which is a complex and challenging part in the medical image processing has been presented in this paper. A modified KNN based clustering and segmentation approach is discuss in this paper using Minkowski distance as the key parameter to meet the objective. The proposed method consists of two stages namely feature extraction and segmentation. The initial part of the work involves removal of the skull tissue through pre-processing followed by extraction and segmentation. Minkowski distance which is a generalized form of Euclidean distance has been utilised in the proposed work and the segmentation results indicate an increased accuracy of nearly 1.5% over FCM and reduction in computation time as result of reduced iterations. The proposed work has been experimented on T1 and T2 synthetic images from two data sets in the tumor database. The experimental result proves that the proposed methodology performs better compared with the other state-of-the-art methodologies.

• 出版日期2018-2
• 单位西北农林科技大学