Title :
Two-Level Algorithm for MCs Detection in Mammograms Using Diverse-Adaboost-SVM
Author :
Harirchi, F. ; Radparvar, P. ; Moghaddam, H. Abrishami ; Dehghan, F. ; Giti, M.
Author_Institution :
K. N. Toosi Univ. of Technol., Tehran, Iran
Abstract :
Clustered micro calcifications (MCs) are one of the early signs of breast cancer. In this paper, we propose a new computer aided diagnosis (CAD) system for automatic detection of MCs in two steps. First, pixels corresponding to potential micro calcifications are found using a multilayer feed-forward neural network. The input of this network consists of 4 wavelet and 2 gray-level features. The output of the network is then transformed to potential micro calcification objects using spatial 4-point connectivity. Second, we extract 25 features from the potential MC objects and use Diverse Adaboost SVM (DA-SVM) and 3 other classifiers to detect individual MCs. A free-response operating characteristics (FROC) curve issued to evaluate the performance of the CAD system. The 90.44% mean TP detection rate is achieved at the cost of 1.043 FP per image by using DA-SVM shows a quite satisfactory detection performance of CAD system.
Keywords :
CAD; cancer; feature extraction; feedforward neural nets; learning (artificial intelligence); mammography; medical diagnostic computing; medical image processing; multilayer perceptrons; pattern clustering; support vector machines; CAD system; breast cancer; clustered microcalcifications; computer aided diagnosis system; diverse Adaboost SVM; feature extraction; free-response operating characteristics curve; gray level features; mammograms; multilayer feedforward neural network; satisfactory detection performance; wavelet features; Artificial neural networks; Classification algorithms; Design automation; Feature extraction; Pixel; Support vector machines; Wavelet transforms; Breast Cancer; Diverse Adaboost SVM; Mammogram; Microcalcification;
Conference_Titel :
Pattern Recognition (ICPR), 2010 20th International Conference on
Conference_Location :
Istanbul
Print_ISBN :
978-1-4244-7542-1
DOI :
10.1109/ICPR.2010.75