Feature Extraction using Balanced Multiwavelets for Classification of Microcalcification Clusters with application in breast cancer diagnostic
| Main Authors: | D.M.Garge, Dr.V.N.Bapat |
|---|---|
| Format: | Article Journal |
| Bahasa: | eng |
| Terbitan: |
, 2011
|
| Subjects: | |
| Online Access: |
https://zenodo.org/record/1438759 |
Daftar Isi:
- Abstract: This paper presents an approach for early breast cancer diagnostic by employing newer member of multiwavelet family: Balanced Multiwavelets. Detection and classification of microcalcification cluster is based on subband image decomposition. Detection of microcalcification is achieved by decomposing the mammogram and then reconstructing it from from the subbands containing only high frequency components. For this type of approach, we have applied different type of balanced multiwavelets for feature extraction. We used these results as an input to classification system. The proposed methodology is tested using Mammographic Image Analysis Society (MIAS) database. Results are presented as the receiver operating charactesistic (ROC) performance and quantified by the area under the ROC curve. Key words: breast cancer, multiwavelets, balancing, microcalcification http://www.ijbst.org/Home/papers-published/ijbst-2011-volume-4-issue-5