Abstract
dc:description.abstractWith advancements in technology, turning to machine learning has become a popular choice for aiding clinicians in the diagnoses of breast cancer malignancies. While the neural networking approach has been vetted thoroughly, this work aims to take advantage of traditional machine learning techniques; mainly support vector machine learning and the optimizing of feature extraction. The discrete-wavelet transform is used in the feature extraction stage of machine learning. Previous works that use this feature extraction technique are analyzed and expanded upon by utilizing a variety of different wavelets as well as other color-spaces with the goal of achieving higher result metrics and efficiency. In addition to the Haar wavelet, the Coiflet, Daubechies, and Symlet wavelets are analyzed due to their potential ability to distinguish more distinct features in an image. Furthermore, different approaches of class separation are explored with the aim of being able to classify a single subclass of a malignant tumor successfully. By utilizing the hue layer of the HSV color-space, higher result metrics are achieved by using a third of the training data. By using a third of the input features, this results in tripling the efficiency of the feature extraction stage. These results are compared to the results of neural networking approaches found in literature.
Degree
thesis:*- Name thesis:degree_name
- MS in Electrical Engineering
- Discipline thesis:degree_discipline
- Electrical Engineering
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Lopatka, Kevin
- Contributors dc:contributor
-
- Wayne Pilkington
- Electrical Engineering
- College of Engineering
Subjects
dc:subject × 4Identifiers
dc:identifier.*- Repository record dc:identifier
- https://digitalcommons.calpoly.edu/theses/2991