Sonoelastomics for Breast Tumor Classification: A Radiomics Approach with Clustering-Based Feature Selection on Sonoelastography

Abstract

A radiomics approach to sonoelastography, called "sonoelastomics," is proposed for classification of benign and malignant breast tumors. From sonoelastograms of breast tumors, a high-throughput 364-dimensional feature set was calculated consisting of shape features, intensity statistics, gray-level co-occurrence matrix texture features and contourlet texture features, which quantified the shape, hardness and hardness heterogeneity of a tumor. The high-throughput features were then selected for feature reduction using hierarchical clustering and three-feature selection metrics. For a data set containing 42 malignant and 75 benign tumors from 117 patients, seven selected sonoelastomic features achieved an area under the receiver operating characteristic curve of 0.917, an accuracy of 88.0%, a sensitivity of 85.7% and a specificity of 89.3% in a validation set via the leave-one-out cross-validation, revealing superiority over the principal component analysis, deep polynomial networks and manually selected features. The sonoelastomic features are valuable in breast tumor differentiation.

Radiomics With Attribute Bagging for Breast Tumor Classification Using Multimodal Ultrasound Images

“…Zhang et al proposed a radiomic approach with HC feature selection on US elastography for breast tumor classification. The data set contained 117 breast tumors (42 malignant and 75 benign).…”

“…The proposed RAB was compared with radiomics by several feature selection/reduction algorithms: (1) principal component analysis (PCA); (2) hierarchical clustering (HC); and (3) the GA (a particular case of RAB with a single classifier). The SVM was used as the classifier for all the methods.…”

“…Radiomics of US images has been used for breast tumor diagnosis . Zhang et al proposed a radiomic approach with clustering‐based feature selection on US elastography for breast tumor classification.…”

“…Radiomics of US images has been used for breast tumor diagnosis . Zhang et al proposed a radiomic approach with clustering‐based feature selection on US elastography for breast tumor classification. A total of 364 radiomic features were calculated, and 7 of them were selected to achieve optimal performance in a receiver operating characteristic curve analysis, with an area under the curve (AUC) of 0.917.…”

A New Challenge for Radiologists: Radiomics in Breast Cancer

“…A radiomics approach on US imaging and specifically on sonoelastograms was proposed by Zhang in 2017, showing that some sonoelastomic features might help to discriminate between benign and malignant breast tumors [ 17 ]. A multicentric and prospective study applied a radiomics approach to DBT for the first time in order to differentiate normal breast tissue from malignant breast tissue in patients with dense breasts [ 18 ].…”