Applications of Intelligent Optimization in Biology and Medicine by Aboul-Ella Hassanien Crina Grosan & Mohamed Fahmy Tolba

Author:Aboul-Ella Hassanien, Crina Grosan & Mohamed Fahmy Tolba
Language: eng
Format: epub
Publisher: Springer International Publishing, Cham

7.2 Methodology

7.2.1 Dataset Description

In the present work, mammograms from a standard benchmark database Mammographic Image Analysis Society (MIAS) have been used. The database contains 322 images, 1024 1024 pixels in size with 256 gray scale tones with a horizontal and vertical resolution of 96 dpi. The database includes the location and nature of abnormality if any. The database also includes the nature of the breast tissue with respect to the density. The images in the database are classified into 3 categories as fatty, fatty-glandular and dense-glandular [41]. The previous related studies for breast density classification using MIAS dataset have either considered 3 class classification into fatty, fatty-glandular or dense-glandular categories or two class classification into fatty or dense by combining the fatty-glandular or dense-glandular categories as dense class [10, 15, 16, 19, 20, 24]. For the present work 3-class classification has been reduced to a 2-class classification problem by combining the fatty-glandular and dense-glandular mammograms into a single category of dense mammograms resulting in 106 mammograms belonging to fatty class and 216 mammograms belonging to dense class. From each image ROIs are extracted and whole database is bifurcated to form training and testing datasets. The final dataset has 322 ROIs stored in Intel Core I3-2310 M, 2.10 GHZ with 3 GB RAM. The description of the database is shown in Fig. 7.3.

Fig. 7.3Dataset Description

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.