False-positive reduction in computer-aided mass detection using mammographic texture analysis and classification.

BACKGROUND AND OBJECTIVE: The aim of computer-aided-detection (CAD) systems for mammograms is to assist radiologists by marking region of interest (ROIs) depicting abnormalities. However, the confusing appearance of some normal tissues that visually look like masses results in a large proportion of marked ROIs with normal tissues. This paper copes with this problem and proposes a framework to reduce false positive masses detected by CAD. METHODS: To avoid the error induced by the segmentation step, we proposed a segmentation-free framework with particular attention to improve feature extraction and classification steps. We investigated for the first time in mammogram analysis, Hilbert's image representation, Kolmogorov-Smirnov distance and maximum subregion descriptors. Then, a feature selection step is performed to select the most discriminative features. Moreover, we considered several classifiers such as Random Forest, Support Vector Machine and Decision Tree to distinguish between normal tissues and masses. Our experiments were carried out on a large dataset of 10168 ROIs (8254 normal tissues and 1914 masses) constructed from the Digital Database for Screening Mammography (DDSM). To simulate practical scenario, our normal regions are false positives asserted by a CAD system from healthy cases. RESULTS: The combination of all the descriptors yields better results than each feature set used alone, and the difference is statistically significant. Besides, the feature selection steps yields a statistically significant increase in the accuracy values for the three classifiers. Finally, the random forest achieves the highest accuracy (81.09%), outperforming the SVM classifier (80.01%)) and decision tree (79.12%), but the difference is not statistically significant. CONCLUSIONS: The accuracy of discrimination between normal and abnormal ROIs in mammograms obtained with the proposed gray level texture features sets are encouraging and comparable to these obtained with multiresolution features. Combination of several features as well as feature selection steps improve the results. To improve false positives reduction in CAD systems for breast cancer diagnosis, these features could be combined with multiresolution features.

Breast cancer diagnosis in digitized mammograms using curvelet moments.

BACKGROUND: Feature extraction is a key issue in designing a computer aided diagnosis system. Recent researches on breast cancer diagnosis have reported the effectiveness of multiscale transforms (wavelets and curvelets) for mammogram analysis and have shown the superiority of curvelet transform. However, the curse of dimensionality problem arises when using the curvelet coefficients and therefore a reduction method is required to extract a reduced set of discriminative features. METHODS: This paper deals with this problem and proposes a feature extraction method based on curvelet transform and moment theory for mammogram description. First, we performed discrete curvelet transform and we computed the four first-order moments from curvelet coefficients distribution. Hence, two feature sets can be obtained: moments from each band and moments from each level. In this work, both sets are studied. Then, the t-test ranking technique was applied to select the best features from each set. Finally, a k-nearest neighbor classifier was used to distinguish between normal and abnormal breast tissues and to classify tumors as malignant or benign. Experiments were performed on 252 mammograms from the Mammographic Image Analysis Society (mini-MIAS) database using the leave-one-out cross validation as well as on 11553 mammograms from the Digital Database for Screening Mammography (DDSM) database using 2×5-fold cross validation. RESULTS: Experimental results prove the effectiveness and the superiority of curvelet moments for mammogram analysis. Indeed, results on the mini-MIAS database show that curvelet moments yield an accuracy of 91.27% (resp. 81.35 %) with 10 (resp. 8) features for abnormality (resp. malignancy) detection. In addition, empirical comparisons of the proposed method against state-of-the-art curvelet-based methods on the DDSM database show that the suggested method does not only lead to a more reduced feature set, but it also statistically outperforms all the compared methods in terms of accuracy. CONCLUSIONS: In summary, curvelet moments are an efficient and effective way to extract a reduced set of discriminative features for breast cancer diagnosis.