| Paper Details: | Downloads: 2443 |
| Serial Number: | P1121802600
|
| Title: | High Precision Automated Breast Cancer Diagnosis based on Multi-modal Texture Analysis of Mammogram Images
|
| Authors: | Nahed Tawfik and Mahmoud Fakhr and Heba A. Elnemr
|
| Abstract: | Breast Cancer is considered one of the most widespread cancers among women. Detection and classification of the disease at early stages may lead to a suitable treatment that in turn reduces the number of deaths. This paper presents an automatic methodology for breast cancer diagnosis based on hybrid texture and statistical features. An image segmentation procedure, to extract the region of interest (ROI), is first performed. The proposed automatic segmentation technique is carried out through two steps, pre-processing and post-processing. The proposed method was assessed using Mini Mammographic Image Analysis Society (Mini-MIAS) dataset and several experiments are conducted using various blends of the utilized features. The results show that the maximum accuracy and specificity are 100%.The results were validated using k-fold cross-validation method.
|
| Keywords: | Breast Cancer, feature extraction, DCT, CT, GLCM, STD, ENT, SVM, and Mini-MIAS dataset.
|
| Journal/Conference: | International Journal of Artificial Intelligence and Machine Learning
|
| Volume: | 18
|
| Issue: | 1
|
| Submission Date: | 10/15/2017 12:00:00 AM
|
| Review Date: | 2/21/2018 12:00:00 AM
|
| Publishing Date: | 3/18/2018 12:00:00 AM
|
| Article Downloads: | 2443
|
| Download: |
|