Please use this identifier to cite or link to this item:
Title: Analysis and Classification of Breast density using Mammographic Images
Researcher: Kumar Indrajeet
Guide(s): Bhadauria HarvendraSingh, Virmani Jitendra
Keywords: Breast density, Feature extraction, Support vector machine, Ensemble of neural network, Hybrid hierarchical classifier
Engineering and Technology,Computer Science,Computer Science Artificial Intelligence
University: Uttarakhand Technical University
Completed Date: 19-2-2018
Abstract: The present research work has been carried out with an aim to enhance the diagnostic potential of mammography imaging modality for classification of breast density. To achieve this objective, the design and implementation of an interactive framework for classification of breast density using digitized screen film mammograms are proposed in the present study. The research objectives for the present work were formulated keeping in view the needs of the radiologists, based on the practical difficulties faced by them in routine clinical practice.The fact has been also observed that the breast density classification systems can be designed using either segmented breast tissue or a predefined ROI on benchmark dataset or dataset collected by an individual research group. Accordingly, the present study performed for developing an efficient 4 class and 2 class breast density classification systems using ROI based approach on a benchmark dataset. newlineThe study was conducted on a comprehensive image dataset of 480 MLO view digitized screen film mammograms. The same set of mammographic images have been used for 2 class breast density classification by considering cases belonging to BIRADS I and BIRADS II classes in fatty image class and cases belonging to BIRADS III and BIRADS IV classes in dense image class.A classification accuracy of 84.1 percent has been achieved by using hybrid hierarchical classification framework which is consisting of less number of classifier with respect to another module. newlineThus it has been concluded that the principal component analysis and multiresolution texture descriptors based computerized framework should be used in the clinical practice for the discrimination between fatty and dense mammograms using digitized screen film mammograms newline newline
Pagination: 197 pages
Appears in Departments:Department of Computer Science and Engineering

Files in This Item:
File Description SizeFormat 
10-chapter 3.pdfAttached File3.65 MBAdobe PDFView/Open
11-chapter 4.pdf1.75 MBAdobe PDFView/Open
12-chapter 5.pdf1.42 MBAdobe PDFView/Open
13-chapter 6.pdf1.52 MBAdobe PDFView/Open
14-chapter 7.pdf1.06 MBAdobe PDFView/Open
15-chapter 8.pdf1.17 MBAdobe PDFView/Open
16-chapter 9.pdf179.26 kBAdobe PDFView/Open
17-list of publications.pdf241.81 kBAdobe PDFView/Open
18-thesis_references.pdf233.1 kBAdobe PDFView/Open
1-title page.pdf50.01 kBAdobe PDFView/Open
2-certificate.pdf347.38 kBAdobe PDFView/Open
3-contents.pdf208.79 kBAdobe PDFView/Open
4-list of tables.pdf169.63 kBAdobe PDFView/Open
5-list of figures.pdf225.65 kBAdobe PDFView/Open
6-list of abbreviations.pdf243.48 kBAdobe PDFView/Open
7-acknowledgements.pdf155.1 kBAdobe PDFView/Open
8-chapter 1.pdf854.08 kBAdobe PDFView/Open
9-chapter 2.pdf506.71 kBAdobe PDFView/Open

Items in Shodhganga are protected by copyright, with all rights reserved, unless otherwise indicated.