Search

Using a large set of ultrasound features does not necessarily ensure improved quantitative classification of breast tumors; rather, it often degrades the performance of a classifier. In this paper, we propose an effective feature reduction approach in the transform domain for improved multi-class classification of breast tumors. Feature transformation methods, such as empirical mode decomposition (EMD) and discrete wavelet transform (DWT), followed by a filter- or wrapper-based subset selection scheme are used to extract a set of non-redundant and more potential transform domain features through decorrelation of an optimally ordered sequence of N ultrasonic bi-modal (i.e., quantitative ultrasound and elastography) features. The proposed transform domain bi-modal reduced feature set with different conventional classifiers will classify 201 breast tumors into benign-malignant as well as BI-RADS⩽3, 4, and 5 categories. For the latter case, an inadmissible error probability is defined for the subset selection using a wrapper/filter. The classifiers use train truth from histopathology/cytology for binary (i.e., benign-malignant) separation of tumors and then bi-modal BI-RADS scores from the radiologists for separating malignant tumors into BI-RADS category 4 and 5. A comparative performance analysis of several widely used conventional classifiers is also presented to assess their efficacy for the proposed transform domain reduced feature set for classification of breast tumors. The results show that our transform domain bimodal reduced feature set achieves improvement of 5.35%, 3.45%, and 3.98%, respectively, in sensitivity, specificity, and accuracy as compared to that of the original domain optimal feature set for benign-malignant classification of breast tumors. In quantitative classification of breast tumors into BI-RADS categories⩽3, 4, and 5, the proposed transform domain reduced feature set attains improvement of 3.49%, 9.07%, and 3.06%, respectively, in likelihood of malignancy and 4.48% in inadmissible error probability compared to that of the original domain optimal subset. In summary, the construction of a transform domain reduced feature set by extracting complementary information from a large set of available bi-modal features and use of qualitative bi-modal BI-RADS can contribute to improved quantitative classification of breast tumors and thereby help reduce the number of unnecessary biopsies, securing a nearly minimum chance of a life-endangering diagnosis.

Quantitative ultrasound (QUS) based parameters like the effective scatterer diameter (ESD) and mean scatterer spacing (MSS) are gaining attention recently as non-invasive biomarkers for soft tissue characterization. In this work, we propose a multiple QUS parameter based technique that employs ESD and MSS, for binary classification of breast lesions. In order to produce improved ESD estimates, we propose a modified frequency domain technique for ESD estimation of breast tissues from the diffuse component of backscattered radio-frequency (RF) data. Ensemble empirical mode decomposition (EEMD) is performed to separate the diffuse component from the coherent component by decomposing the RF data into their intrinsic mode functions (IMFs). A non-parametric Kolmogorov-Smirnov (K-S) test is employed for automatic IMF selection along with a multi-step system effect minimization process. The ESD is estimated using a nearest neighborhood average regression line fitting algorithm. Furthermore, we use an ameliorated EEMD domain autoregressive (AR) spectral estimation technique for MSS estimation. On using the ESD for binary classification of 159 lesions, we obtain high sensitivity, specificity, accuracy values of 91.07%, 96.12%, and 94.34%, respectively, with an area under the receiver operating characteristics (ROC) curve of 0.94. On combining ESD with MSS we obtain even more improved sensitivity, specificity, and accuracy values of 96.43%, 95.15%, and 95.60%, respectively, with an area under the ROC of 0.96. Such a high classification performance highlights the potential of these QUS parameters to be used as non-invasive biomarkers for breast cancer detection.