01whole.pdf (16.99 MB)
Download file

Breast image classification using machine learning techniques

Download (16.99 MB)
thesis
posted on 28.03.2022, 18:56 authored by Jonathan Wu
Among all the common forms of cancer, breast cancer is the most prevalent one. Statistics show that breast cancer causes the second highest mortality in women worldwide. Accurate classification of the breast cancer is of high importance for the proper diagnosis. The survival rate of breast cancer affected people is greatly increased if the cancer is detected early. Ultrasound, MRI or CT imaging techniques are used to capture the current condition of the breast. Digital signal processing, along with modern computer aided techniques are used to analyse these images, and proper diagnosis of the images can save the diagnoses time of the doctor. For the correct classification of the images, Feature detection and extraction of the features play a vital role. Various feature detection techniques are available, But for this thesis we will use the wavelet transform to perform feature detection. The features extracted from the images using the wavelet transform will be utilised for classification purposes. For the classification task we will use a support vector machine.

History

Table of Contents

1. Introduction -- 2. Background and related work -- 3. Wavelet transform -- 4. Invariant scattering convolution networks -- 5. Results -- 6. Conclusions -- 7. Abbreviations -- Appendices -- Bibliography.

Notes

Empirical thesis. Bibliography: pages 89-90

Awarding Institution

Macquarie University

Degree Type

Thesis bachelor honours

Degree

BSc (Hons), Macquarie University, Faculty of Science and Engineering, School of Engineering

Department, Centre or School

School of Engineering

Year of Award

2016

Principal Supervisor

Yinan Kong

Rights

Copyright Jonathan Wu 2016. Copyright disclaimer: http://mq.edu.au/library/copyright

Language

English

Extent

1 online resource (xv, 90 pages diagrams, graphs, tables)

Former Identifiers

mq:70310 http://hdl.handle.net/1959.14/1262421

Usage metrics

Keywords

Exports