Nomogram Model for Predicting Minimal Breast Cancer Based on Clinical and Ultrasonic Characteristics.

Abstract

Purpose: To construct a nomogram prediction model on minimal breast cancer (≦ 10 mm) based on clinical and ultrasound parameters.

Methods: Clinical and ultrasound data of 433 patients with minimal breast lesions was conducted in this retrospective study. Patients were randomly divided into a training set and a validation set with a ratio of 7:3. Independent risk factors for minimal breast cancer were selected by the least absolute shrinkage and selection operator (LASSO) regression and multivariable logistic regression analysis to construct a nomogram prediction model. The calibration curve, the clinical decision curve analysis (DCA) and the area under the curve (AUC) of the receiver operating characteristic (ROC) curve were used to evaluate the diagnostic efficacy of the model.

Results: Age, margin, shape, and breast density were independent risk factors for malignant minimal breast lesions (P < 0.05). The AUC of the training set and validation set of the nomogram prediction model were 0.875, the sensitivity were 75.0% and 88.9%, the specificity were 83.8% and 77.7%, respectively. The mean absolute error (MAE) of the training set and validation set of the calibration curve were 0.01 and 0.024, respectively.

Conclusion: The nomogram prediction model has good discrimination, calibration and clinical practical value in the training set and validation set. The minimal breast cancer prediction model based on clinical and ultrasonic features possesses high clinical value, facilitating the early diagnosis of minimal breast cancer.