Figure 2 | Scientific Reports

Figure 2

From: Development and validation of multivariate models integrating preoperative clinicopathological and radiographic findings to predict HER2 status in gastric cancer

Figure 2

The workflow of this study. (a) Erythrocyte indices, tumor markers, differentiation degree based on biopsy, CT morphological characteristics, CT value-related and texture parameters were extracted. (b) Multivariate models were built based on binomial logistic regression and machine learning algorithm. (c) The overexpression of human epidermal growth factor receptor 2 (HER2) in gastric cancer was tested by IHC based on surgical specimens, samples scored as IHC 2 + (equivocal) were additionally detected by FISH. Diagnostic performance for predicting HER2 status was obtained by ROC curve analysis. MCV mean corpuscular volume, MCH mean corpuscular hemoglobin, MCHC mean corpuscular hemoglobin concentration, RDW red cell distribution width, Hb hemoglobin, CEA carcinoembryonic antigen, CA carbohydrate antigen, CT computed tomography, LASSO least absolute shrinkage and selection operator, SVM support vector machine, IHC immunohistochemistry, FISH fluorescence in situ hybridization, ROC receiver operating characteristic.

Back to article page