Brain Tumor Boundary Segmentation of MR Imaging using Spatial Domain Image Processing
| Autor: | Chandrika Malkanthi, Maheshi B. Dissanayake |
|---|---|
| Rok vydání: | 2017 |
| Předmět: |
0301 basic medicine
Ground truth medicine.diagnostic_test Pixel Computer science business.industry ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION Scale-space segmentation Boundary (topology) Magnetic resonance imaging Line drawing algorithm Pattern recognition Image processing 02 engineering and technology General Medicine 03 medical and health sciences ComputingMethodologies_PATTERNRECOGNITION 030104 developmental biology 0202 electrical engineering electronic engineering information engineering medicine 020201 artificial intelligence & image processing Segmentation Artificial intelligence business |
| Zdroj: | International Journal for Innovation Education and Research. 5:1-9 |
| ISSN: | 2411-2933 2411-3123 |
| DOI: | 10.31686/ijier.vol5.iss10.621 |
| Popis: | Extracting information for medical purposes from magnetic resonance imaging is critically important for diagnostic and treatment plans. In this paper, a simple algorithm for tumor segmentation of Magnetic resonance imaging (MRI) is introduced. The novelty incorporates, preserving fine details of the input image while detecting the boundary accurately. Tumor segmentation is carried out by set of pre processing steps followed by morphological operations. Rough contour of the tumor is localized to reduce the search space for the boundary. Line drawing algorithm in cooperated with pixel selection criteria is used to detect the accurate boundary. The algorithm is evaluated in terms of the performance and accuracy with radiologist labelled ground truth MRI scans. Simulation results show that the proposed algorithm provides better identification with above 95% of accuracy, for clearly distinguishable tumors in relation to conventional contour detection methods. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|