Construction of a nomogram model for malignancy risk of breast lesions based on quantitative multimodal ultrasound features and clinical indicators
10.3760/cma.j.cn131148-20241120-00606
- VernacularTitle:基于量化的多模态超声特征及临床指标构建乳腺病灶恶性风险列线图预测模型
- Author:
Quan CHEN
1
;
Yan ZHENG
;
Yang GU
;
Ping HU
;
Qing JIN
;
Fenglin DONG
Author Information
1. 昆山市中医医院超声科，昆山 215300
- Publication Type:Journal Article
- Keywords:
Ultrasonography;
Breast cancer;
Automated breast volume scanner;
Ultrasound elastography;
Quantitative score;
Nomogram
- From:
Chinese Journal of Ultrasonography
2025;34(4):303-310
- CountryChina
- Language:Chinese
-
Abstract:
Objective:To construct a nomogram prediction model for the malignant risk of breast lesions based on quantitative multimodal ultrasound features and clinical indicators，and to explore its clinical application value.Methods:A total of 430 patients with breast diseases（473 lesions）confirmed by pathological examination at Kunshan Traditional Chinese Medicine Hospital from August 2018 to December 2023 were retrospectively analyzed in the study. All the patients underwent routine handheld ultrasound（HHUS），automated breast volume scanning（ABVS）and ultrasound elastography（UE）examinations before biopsy or surgery. Four hundred and seventy-three breast lesions were randomly divided into training set（ n=331）and validation set（ n=142）at a ratio of 7∶3. The ABVS ultrasound features of 331 lesions in the training set were analyzed to identify the risk features of malignant breast lesions. Based on the weight of the regression coefficients β，the ABVS risk features were assigned quantitative scores for each lesion. Using pathology results as the gold standard，ROC curves were plotted to compare the diagnostic efficacy of ABVS quantitative scoring and HHUS in the training set. The variable with higher efficacy was selected for inclusion in the multivariable logistic regression analysis. Based on the identified independent predictive factors，a nomogram was constructed to quantify the malignancy risk of breast lesions. The nomogram was evaluated and validated using the area under the ROC curve（AUC），calibration curve and clinical decision curve analysis（DCA）. Results:The AUC for ABVS quantitative scoring and HHUS in distinguishing benign and malignant breast lesions were 0.941 and 0.903，respectively，with a statistically significant difference（ Z=2.081， P=0.037）. Multivariate Logistic regression analysis showed that age，ABVS quantitative score and elasticity score were independent predictors for the identification of benign and malignant breast lesions（all P<0.05）. A nomogram was constructed based on the aforementioned independent predictive factors. The AUC for the nomogram were 0.968 in the training set and 0.943 in the validation set，indicating good discrimination. The calibration curve indicated that the predicted probabilities from the nomogram were in good agreement with the actual outcomes. DCA indicated strong clinical applicability of the nomogram. Conclusions:The nomogram based on ABVS quantitative scores，elasticity scores，and age demonstrates high diagnostic value，providing a novel method for preoperative assessment of the malignancy risk of breast lesions.
