Gray Matter Segmentation of Brain MRI Using Hybrid Enhanced Independent Component Analysis
| Autor: | M. Satya Sai Ram, Shaik Basheera |
|---|---|
| Rok vydání: | 2021 |
| Předmět: |
020205 medical informatics
business.industry Computer science Anatomical segmentation Pattern recognition 02 engineering and technology Mixture model Computer Graphics and Computer-Aided Design Independent component analysis Gray (unit) 030218 nuclear medicine & medical imaging Computer Science Applications 03 medical and health sciences 0302 clinical medicine 0202 electrical engineering electronic engineering information engineering Brain mri Segmentation Computer Vision and Pattern Recognition Artificial intelligence business |
| Zdroj: | International Journal of Image and Graphics. :2150029 |
| ISSN: | 1793-6756 0219-4678 |
| DOI: | 10.1142/s0219467821500297 |
| Popis: | One of the primary pre-processing tasks of medical image analysis is segmentation; it is used to diagnose the abnormalities in the tissues. As the brain is a complex organ, anatomical segmentation of brain tissues is a challenging task. Segmented gray matter is analyzed for early diagnosis of neurodegenerative disorders. In this endeavor, we used enhanced independent component analysis to perform segmentation of gray matter in noise-free and noisy environments. We used modified [Formula: see text]-means, expectation–maximization and hidden Markov random field to provide better spatial relation to overcome inhomogeneity, noise and low contrast. Our objective is achieved using the following two steps: (i) Irrelevant tissues are stripped from the MRI using skull stripping algorithm. In this algorithm, sequence of threshold, morphological operations and active contour are applied to strip the unwanted tissues. (ii) Enhanced independent component analysis is used to perform segmentation of gray matter. The proposed approach is applied on both T1w MRI and T2w MRI images at different noise environments such as salt and pepper noise, speckle noise and Rician noise. We evaluated the performance of the approach using Jaccard index, Dice coefficient and accuracy. The parameters are further compared with existing frameworks. This approach gives better segmentation of gray matter for the diagnosis of atrophy changes in brain MRI. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|