Abstract:Rationale and objectives: Dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) radiomics had been used to evaluate lymphovascular invasion (LVI) in patients with breast cancer. However, no studies had explored the associations between features from delayed phase as well as multiphases DCE-MRI and the LVI status. Thus, we aimed to develop an efficient nomogram based on multiphases DCE-MRI to predict the LVI status in invasive (IBC) breast cancer patients. Materials and methods: A retrospective analysis was conducted on preoperative clinical, pathological, and DCE-MRI data of 173 breast cancer patients. All patients were randomly assigned into training set (n=121) and validation set (n=52) in 7:3 ratio. The clinical, pathologic, and conventional MRI characteristics were then subjected to univariate and multivariate logistic regression analysis, and the clinical risk factors with P < 0.05 in the multivariate logistic regression were used to build clinical models. Different single-phase models (early phase, peak phase, and terminal phase), as well as a multiphases model integrating radiomics features from multiple phases, were established. Furthermore, a preoperative radiomics nomogram model was constructed by combining the rad-score of the multiphases model with clinicopathologic independent risk factors. Finally, the performance of the multiphases model, clinical model, and rad-score was compared using receiver operating characteristic (ROC) curves, area under the curve (AUC) values, and decision curve analysis (DCA). The clinical utility of the rad-score was evaluated using calibration curves, and Delong test was used to compare the differences in AUC values among the different models. Results: The axillary lymph nodes (ALN) status and Ki-67 had been identified as clinicopathologic independent predictors and a clinical model had been constructed. Image features that were extracted from the terminal phase of the DCE-MRI exhibited notably superior predictive performances compared to features from the other single phases. Particularly, in the multiphases model, terminal phase features were identified as potentially providing more predictive information. Among the nine features that were found to be associated with LVI in the multiphase model, one was derived from the early phase, two from the peak phase, and six from the terminal phase, indicating that terminal phase features contributed significantly more information towards predicting LVI. Evaluation of the nomogram performance revealed promising results in both the training set (AUCs: clinical model vs. multiphase model vs. nomogram=0.734 vs. 0.840 vs. 0.876) and the validation set (AUCs: clinical model vs. multiphase model vs. nomogram=0.765 vs. 0.753 vs. 0.832). Conclusion: The DCE-MRI-based radiomics model demonstrated utility in predicting LVI status, features of the terminal phase offered more valuable information particularly. The preoperative radiomics nomogram enhanced the diagnostic capability of identifying LVI status in IBC patients, and might aid clinicians in making personalized treatment decisions.

Exploring habitats-based spatial distributions: improving predictions of lymphovascular invasion in invasive breast cancer

Non-invasive Prediction of Lymph Node Status for Patients with Early-Stage Invasive Breast Cancer Based on a Morphological Feature from Ultrasound Images

Habitat-based radiomics enhances the ability to predict lymphovascular space invasion in cervical cancer: a multi-center study

Intra‐ and Peritumoral Based Radiomics for Assessment of Lymphovascular Invasion in Invasive Breast Cancer

Using Machine Learning Methods to Assess Lymphovascular Invasion and Survival in Breast Cancer: Performance of Combining Preoperative Clinical and MRI Characteristics

Development of an Intratumoral and Peritumoral Radiomics Nomogram Using Digital Breast Tomosynthesis for Preoperative Assessment of Lymphovascular Invasion in Invasive Breast Cancer

Multiparametric MRI-based radiomics nomogram for preoperative prediction of lymphovascular invasion and clinical outcomes in patients with breast invasive ductal carcinoma

Multiphases DCE-MRI Radiomics Nomogram for Preoperative Prediction of Lymphovascular Invasion in Invasive Breast Cancer

Radiomics-based analysis of dynamic contrast-enhanced magnetic resonance image: A prediction nomogram for lymphovascular invasion in breast cancer

Habitat Imaging With Tumoral and Peritumoral Radiomics for Prediction of Lung Adenocarcinoma Invasiveness on Preoperative Chest CT: A Multicenter Study

Assessment of Lymphovascular Invasion in Breast Cancer Using a Combined MRI Morphological Features, Radiomics, and Deep Learning Approach Based on Dynamic Contrast‐Enhanced MRI

Delta-Radiomics Based on Dynamic Contrast-Enhanced MRI for Predicting Lymphovascular Invasion in Invasive Breast Cancer

Predicting lymphovascular invasion in clinically node-negative breast cancer detected by abbreviated magnetic resonance imaging: Transfer learning vs. radiomics

The Association Between Tumor Radiomic Analysis and Peritumor Habitat‐Derived Radiomic Analysis on Gadoxetate Disodium‐Enhanced MRI With Microvascular Invasion in Hepatocellular Carcinoma

Preoperative Prediction of Microvascular Invasion in Hepatocellular Carcinoma Using Diffusion-Weighted Imaging–based Habitat Imaging

Ultrasound Radiomics-Based Nomogram to Predict Lymphovascular Invasion in Invasive Breast Cancer: a Multicenter, Retrospective Study.

Intratumoral and peritumoral habitat imaging based on multiparametric MRI to predict cervical stromal invasion in early-stage endometrial carcinoma

Quantification of Intratumoral Heterogeneity Using Habitat-Based MRI Radiomics to Identify HER2-positive, -Low and -Zero Breast Cancers: a Multicenter Study

Preoperative prediction of lymphovascular invasion in patients with T1 breast invasive ductal carcinoma based on radiomics nomogram using grayscale ultrasound

Multiparametric MRI-based radiomics for the prediction of microvascular invasion in hepatocellular carcinoma

Leveraging multimodal MRI-based radiomics analysis with diverse machine learning models to evaluate lymphovascular invasion in clinically node-negative breast cancer