Comparison of Multi-resolution Analysis Patterns for Texture Classification of Breast Tumors Based On DCE-MRI

Alexia Tzalavra (1) , Kalliopi Dalakleidi (1) , Evangelia I. Zacharaki (2, 3) , Nikolaos Tsiaparas (1) , Fotios Constantinidis (4) , Nikos Paragios (2, 3) , Konstantina S. Nikita (1)

1 School of Electrical and Computer Engineering [Athènes]
2 CVN - Centre de vision numérique
3 GALEN - Organ Modeling through Extraction, Representation and Understanding of Medical Image Content
4 NHSGGC - NHS Greater Glasgow & Clyde [Glasgow]

Although Fourier and Wavelet Transform have been widely used for texture classification methods in medical images, the discrimination performance of FDCT has not been investigated so far in respect to breast cancer detection. Ιn this paper, three multi-resolution transforms , namely the Discrete Wavelet Transform (DWT), the Stationary Wavelet Transform (SWT) and the Fast Discrete Curvelet Transform (FDCT) were comparatively assessed with respect to their ability to discriminate between malignant and benign breast tumors in Dynamic Contrast-Enhanced Magnetic Resonance Images (DCE-MRI). The mean and entropy of the detail sub-images for each decomposition scheme were used as texture features, which were subsequently fed as input into several classifiers. FDCT features fed to a Linear Discriminant Analysis (LDA) classifier produced the highest overall classification performance (93,18 % Accuracy).

Mots clés: Classification; DCE-MRI; Texture; Wavelet; Breast tumor diagnosis

Domaines: Vision par ordinateur et reconnaissance de formes [cs.CV]; Imagerie médicale

Alexia Tzalavra, Kalliopi Dalakleidi, Evangelia I. Zacharaki, Nikolaos Tsiaparas, Fotios Constantinidis, et al.. Comparison of Multi-resolution Analysis Patterns for Texture Classification of Breast Tumors Based On DCE-MRI. MICCAI 2016 - 7th International Workshop on Machine Learning in Medical Imaging, Oct 2016, Athens, Greece. pp.296-304, ⟨10.1007/978-3-319-47157-0_36⟩. ⟨hal-01359118⟩