Please use this identifier to cite or link to this item:
http://hdl.handle.net/10603/437863
Title: | Deep learning techniques for brain tumor classification using optimization logic |
Researcher: | Leena B |
Guide(s): | Jayanthi A N |
Keywords: | Life Sciences Neuroscience and Behaviour Neurosciences Brain tumor morphology Skull stripping |
University: | Anna University |
Completed Date: | 2022 |
Abstract: | Brain is the most sensitive organ, in which the brain cells are difficult to newlinebe renewed when it is infected by dangerous diseases. The brain tumors are newlinecategorized into two different types namely benign and malignant. The benign newlinetumor may cause change in the structure and shape of the cells. Hence it newlineshould be treated before it spreads to different parts of the brain or infects the newlineother cells. The malignant tumor is more dangerous and can spread inside the newlinebrain if not treated properly. Brain tumor detection is a difficult and sensitive newlinetask that implies the experience of the classifier. In this research work, three newlinemajor contributions have been made. In the first contribution, a novel brain newlinetumor classification method is introduced with different steps such as newline denoising, skull stripping, segmentation, feature extraction and newlineclassification . Initially, the input image is subjected for denoising step, in newlinewhich the entropy-based trilateral filtering process is carried out. Next, the newlinedenoised image is subjected to skull stripping process that includes Otsu newlinethresholding algorithms and morphology segmentation . Next step is the newlinesegmentation process, where these images are implanted for segmenting the newlineimage through the Adaptive CLFAHEand#8223; approach. From the segmented newlineimage, the GLCM based features are extracted. Finally, the classification newlineprocess is done by the new hybridized approach that hybridizes the Bayesian newlinenetwork and DBN. In order to enhance the classification performance, the newline |
Pagination: | xxii, 171p. |
URI: | http://hdl.handle.net/10603/437863 |
Appears in Departments: | Faculty of Information and Communication Engineering |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
01_title.pdf | Attached File | 27.61 kB | Adobe PDF | View/Open |
02_prelim pages.pdf | 2.56 MB | Adobe PDF | View/Open | |
03_content.pdf | 374.01 kB | Adobe PDF | View/Open | |
04_abstract.pdf | 131.32 kB | Adobe PDF | View/Open | |
05_chapter 1.pdf | 415.36 kB | Adobe PDF | View/Open | |
06_chapter 2.pdf | 378.67 kB | Adobe PDF | View/Open | |
07_chapter 3.pdf | 758.26 kB | Adobe PDF | View/Open | |
08_chapter 4.pdf | 1.01 MB | Adobe PDF | View/Open | |
09_chapter 5.pdf | 416.14 kB | Adobe PDF | View/Open | |
10_annexures.pdf | 137.7 kB | Adobe PDF | View/Open | |
80_recommendation.pdf | 92.62 kB | Adobe PDF | View/Open |
Items in Shodhganga are licensed under Creative Commons Licence Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0).
Altmetric Badge: