Breast image classification using machine learning techniques

Empirical thesis.Bibliography: pages 59-60.1. Introduction -- 2. Literature review -- 3. Mammographic images -- 4. Tamura's features -- 5. Feature data -- 6. Support vector machines -- 7. Results & discussion -- 8. Conclusion -- 9. Future work -- 10. Abbreviations -- Appendix -- Bibliography.Proper diagnosis of breast images may be performed through a combination of image processing and machine learning techniques. Provided an accurate diagnosis, the time in which a doctor would personally analyse the images is reduced.Practically, speaking, the diagnosis process consists of feature detection and classification. This research project investigates the performance of Tamura's texture analysis in combination with support vector machines (SVM) as a means to carry out this process. As the provided data consists of three classes, normal, benign and cancer, multiclass SVM techniques will be investigated.Mode of access: World wide web.1 online resource (xiii, 60 pages illustrations