Quantitative breast lesion classification based on multichannel distributions in shear-wave imaging

摘要: A computer-aided diagnosis (CAD) system based on the quantified color distributions in shear wave elastography (SWE) was developed.The combination of proposed CAD using SWE features and BI-RADS assessment would provide a promising diagnostic suggestion.Spatial correlation combined with kPa-weighted information as feature to reduce influence noises far from tumor. Background objectivesA shear-wave developed evaluate malignancies breast tumors. MethodsFor 57 benign 31 malignant tumors, 18 were extracted regions interest (ROI), including tumor peritumoral areas. In ROI, histogram each channel described moments such mean, variance, skewness, kurtosis. Moreover, three channels vector tissue elasticity. The then logistic regression classifier for classification. ResultsThe performance achieved an accuracy 81%. Combining obtained Az improvement 0.77 0.89 (p-value <0.05). ConclusionsThe suggestion.