Abstract The aim of this study was to construct a radiomics nomogram for prediction of breast masses (BMs) by analyzing the clinical characteristics of the patients as well as radiomics features of two-dimensional (2D) ultrasound images and strain elastography images. In this retrospective study, 219 patients diagnosed with BMs were enrolled and randomly divided into training set and testing set in a 7:3 ratio. Radiomics nomogram was constructed based on clinical features and Radscore to compare area under the receiver operating characteristic curve (AUC) with another models. The AUCs of the training set were 0.83, 0.91, 0.92, 0.96, and 0.99 for the clinical model, elastography radiomics model, 2D radiomics model, bimodal radiomics model, and nomogram, respectively, and the AUCs of the testing set were 0.86, 0.87, 0.91, 0.93, and 0.95, respectively. There were significant differences in AUC between nomogram and another models ( p < 0.05). 2D ultrasound radiomics model and strain elastography radiomics model were of diagnostic value in identifying BMs. The bimodal radiomics model was superior to these two single-modal radiomics models. Nomogram can further enhance the diagnosis of BMs and contribute valuable information for clinical decision making.
A Thu, study studied this question.