Please use this identifier to cite or link to this item:
http://hdl.handle.net/10603/475539
Title: | Liver Tumor Segmentation and Classification using Outline Preservation and Multilevel Local Region based Sparse Shape Composition method |
Researcher: | Sakthi Saravanan B |
Guide(s): | A Rama |
Keywords: | Computer Science Engineering and Technology Operations Research and Management Science |
University: | Saveetha University |
Completed Date: | 2022 |
Abstract: | Nowadays, liver cancer is the foremost crisis for the person which leads newlineto death. The report shows that 7, 82000 cases are recently detected by the liver newlinecancer and most of the people pass away owed to this liver cancer. There are newlinesome major drawbacks in this existing traditional method for detecting the liver newlinecancer or the tumor that are ineffective towards the liver segmentation. newlineTherefore, liver segmentation offers a greater dataset, time complexity, less newlineaccuracy, quality of loss, high computational cost, and edge information loss. To newlinesolve these problems an efficient method is introduced. newline The planned scheme comprises three stages such as preprocessing, newlinesegmentation and classification. Using this technique, the noises are removed newlineand also the edges are sharpened. The preprocessed images are taken as an newlineinput for the segmentation practice with the benefit of outline preservation based newlinesegmentation (OPBS). The extracted image features from the segmented image newlinegives the essential information regarding classification. With novel similarity newlinesearch based hybrid classification distance the features are classified. newlineA CAD (Computer-Aided Diagnosis) systems acquaint with dissimilar image newlineprocessing recitals for primary recognition of liver disease to uncertain the liver newlinetumor death percentage. A three-dimensional IR, CAD dataset is used in this newlineproposed OPBS-SSHC technique. The performance analysis is measured for the newlinestudy of proposed and existing technique. The OPBS-SSHC analyzed with newlinevarious parameters like volumetric overlap error (VOE), accuracy, precision, newlinerecall, F-measures, and coefficients are jaccard, dice and kappa. The newlineperformance implies that anticipated work than to compare several methods. newlineSeveral classification methods is also compared with future method to analyze it. |
Pagination: | |
URI: | http://hdl.handle.net/10603/475539 |
Appears in Departments: | Department of Engineering |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
01_title.pdf | Attached File | 68.04 kB | Adobe PDF | View/Open |
02_prelimpages.pdf | 1.65 MB | Adobe PDF | View/Open | |
03_content.pdf | 105.51 kB | Adobe PDF | View/Open | |
04_abstract.pdf | 69.52 kB | Adobe PDF | View/Open | |
05_chapter1.pdf | 346.69 kB | Adobe PDF | View/Open | |
06_chapter 2.pdf | 330.47 kB | Adobe PDF | View/Open | |
07_chapter 3.pdf | 718.98 kB | Adobe PDF | View/Open | |
08_chapter4.pdf | 269.18 kB | Adobe PDF | View/Open | |
09_chapter 5.pdf | 293.55 kB | Adobe PDF | View/Open | |
10_annexures.pdf | 42.51 kB | Adobe PDF | View/Open | |
80_recommendation.pdf | 55.92 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: