Automated Breast Volume Scanner (ABVS)-Based Radiomic Nomogram: A Potential Tool for Reducing Unnecessary Biopsies of BI-RADS 4 Lesions

Preoperative discrimination of invasive and non-invasive breast cancer using machine learning based on automated breast volume scanning (ABVS) radiomics and virtual touch quantification (VTQ)

PURPOSE

Evaluating the efficacy of machine learning for preoperative differentiation between invasive and non-invasive breast cancer through integrated automated breast volume scanning (ABVS) radiomics and virtual touch quantification (VTQ) techniques.

METHODS

We conducted an extensive retrospective analysis on a cohort of 171 breast cancer patients, differentiating them into 124 invasive and 47 non-invasive cases. The data was meticulously divided into a training set (n = 119) and a validation set (n = 52), maintaining a 70:30 ratio. Several machine learning models were developed and tested, including Logistic Regression (LR), Random Forest (RF), Decision Tree (DT), and Support Vector Machine (SVM). Their performance was evaluated using the Area Under the Receiver Operating Characteristic (ROC) Curve (AUC), and visualized the feature contributions of the optimal model using Shapley Additive Explanations (SHAP).

RESULTS

Through both univariate and multivariate logistic regression analyses, we identified key independent predictors in differentiating between invasive and non-invasive breast cancer types: coronal plane features, Shear Wave Velocity (SWV), and Radscore. The AUC scores for our machine learning models varied, ranging from 0.625 to 0.880, with the DT model demonstrating a notably high AUC of 0.874 in the validation set.

CONCLUSION

Our findings indicate that machine learning models, which integrate ABVS radiomics and VTQ, are significantly effective in preoperatively distinguishing between invasive and non-invasive breast cancer. Particularly, the DT model stood out in the validation set, establishing it as the primary model in our study. This highlights its potential utility in enhancing clinical decision-making processes.

Predicting hormone receptor status in invasive breast cancer through radiomics analysis of long-axis and short-axis ultrasound planes

The hormone receptor (HR) status plays a significant role in breast cancer, serving as the primary guide for treatment decisions and closely correlating with prognosis. This study aims to investigate the predictive value of radiomics analysis in long-axis and short-axis ultrasound planes for distinguishing between HR-positive and HR-negative breast cancers. A cohort of 505 patients from two hospitals was stratified into discovery (Institute 1, 416 patients) and validation (Institute 2, 89 patients) cohorts. A comprehensive set of 788 ultrasound radiomics features was extracted from both long-axis and short-axis ultrasound planes, respectively. Utilizing least absolute shrinkage and selection operator (LASSO) regression analysis, distinct models were constructed for the long-axis and short-axis data. Subsequently, radiomics scores (Rad-scores) were computed for each patient. Additionally, a combined model was formulated by integrating data from long-axis and short-axis Rad-scores along with clinical factors. The diagnostic efficacy of all models was evaluated by the area under the receiver operating characteristic (ROC) curve (AUC). The long-axis and short-axis models, consisting of 11 features and 15 features, respectively, were established, yielding AUCs of 0.743 and 0.751 in the discovery cohort, and 0.795 and 0.744 in the validation cohort. The calculated long-axis and short-axis Rad-scores exhibited significant differences between HR-positive and HR-negative groups across all cohorts (all p < 0.001). Univariate analysis identified ultrasound-reported tumor size as an independent predictor. The combined model, incorporating long-axis and short-axis Rad-scores along with tumor size, achieved superior AUCs of 0.788 and 0.822 in the discovery and validation cohorts, respectively. The combined model effectively distinguishes between HR-positive and HR-negative breast cancers based on ultrasound radiomics features and tumor size, which may offer a valuable tool to facilitate treatment decision making and prognostic assessment.

Nomogram Utilizing ABVS Radiomics and Clinical Factors for Predicting ≤ 3 Positive Axillary Lymph Nodes in HR+ /HER2- Breast Cancer with 1-2 Positive Sentinel Nodes

BACKGROUND

In HR+ /HER2- breast cancer patients with ≤ 3 positive axillary lymph nodes (ALNs), genomic tests can streamline chemotherapy decisions. Current studies, centered on tumor metrics, miss broader patient insights. Automated Breast Volume Scanning (ABVS) provides advanced 3D imaging, and its potential synergy with radiomics for ALN evaluation is untapped.

OBJECTIVE

This study sought to combine ABVS radiomics and clinical characteristics in a nomogram to predict ≤ 3 positive ALNs in HR+ /HER2- breast cancer patients with 1-2 positive sentinel lymph nodes (SLNs), guiding clinicians in genetic test candidate selection.

METHODS

We enrolled 511 early-stage breast cancer patients: 362 from A Hospital for training and 149 from B Hospital for validation. Using LASSO logistic regression, primary features were identified. A clinical-radiomics nomogram was developed to predict the likelihood of ≤ 3 positive ALNs in HR+ /HER2- patients with 1-2 positive SLNs. We assessed the discriminative capability of the nomogram using the ROC curve. The model's calibration was confirmed through a calibration curve, while its fit was evaluated using the Hosmer-Lemeshow (HL) test. To determine the clinical net benefits, we employed the Decision Curve Analysis (DCA).

RESULTS

In the training group, 81.2% patients had ≤ 3 metastatic ALNs, and 83.2% in the validation group. We developed a clinical-radiomics nomogram by analyzing clinical characteristics and rad-scores. Factors like positive SLNs (OR=0.077), absence of negative SLNs (OR=11.138), lymphovascular invasion (OR=0.248), and rad-score (OR=0.003) significantly correlated with ≤ 3 positive ALNs. The clinical-radiomics nomogram, with an AUC of 0.910 in training and 0.882 in validation, outperformed the rad-score-free clinical nomogram (AUCs of 0.796 and 0.782). Calibration curves and the HL test (P values 0.688 and 0.691) confirmed its robustness. DCA showed the clinical-radiomics nomogram provided superior net benefits in predicting ALN burden across specific threshold probabilities.

CONCLUSION

We developed a clinical-radiomics nomogram that integrated radiomics from ABVS images and clinical data to predict the presence of ≤ 3 positive ALNs in HR+ /HER2- patients with 1-2 positive SLNs, aiding oncologists in identifying candidates for genomic tests, bypassing ALND. In the era of precision medicine, combining genomic tests with SLN biopsy refines both surgical and systemic patient treatments.

Multimodal ultrasound imaging for diagnostic differentiation of sclerosing adenosis from invasive ductal carcinoma

Background

Sclerosing adenosis (SA) is a common proliferative benign lesion without atypia in the breast that may mimic invasive ductal carcinoma (IDC) on medical imaging, leading to it often being misdiagnosed and mistreated. Consequently, the purpose of this study was to assess the diagnostic value of multimodal ultrasound imaging in distinguishing SA from IDC.

Methods

Multimodal ultrasound imaging, including automated breast volume scan (ABVS), elasticity imaging (EI), and color Doppler flow imaging (CDFI), were performed on 120 consecutive patients comprising 122 breast lesions (54 SA, 68 IDC). All lesions were pathologically confirmed. Multimodal ultrasound imaging features were compared between the two groups. Binary logistic regression analysis based on ABVS, EI, and CDFI was conducted to formulate a logistic regression equation for differentiating SA from IDC. The diagnostic performances of ABVS, EI, CDFI, and their combination were compared by the receiver operating characteristic (ROC) curve analysis.

Results

The sensitivity, specificity, and accuracy of ABVS, EI, CDFI, and their combination in differentiating SA from IDC were, respectively, 75.00%, 72.22%, and 73.77%; 86.76%, 72.22%, and 80.33%; 73.53%, 64.81%, and 69.67%; and 88.24%, 74.07%, and 81.97%. Combining multimodal ultrasound imaging yielded an area under the curve (AUC) of 0.895 (95% confidence interval: 0.827-0.943), which was higher than that of ABVS, EI, and CDFI, with AUC values of 0.736, 0.795, and 0.692, respectively, and the difference was statistically significant (ABVS vs. combined model, P<0.001; CDFI vs. combined model, P<0.001; EI vs. combined model, P<0.001). There was no significant difference in the diagnostic efficacy among the three imaging modalities (ABVS vs. EI, P=0.266; ABVS vs. CDFI, P=0.4671; EI vs. CDFI, P=0.051). Compared with those in IDC, the calcification (16.67% vs. 57.35%; P<0.001) and retraction phenomena in the coronal planes (18.52% vs. 57.35%; P<0.001) were less common in patients with SA, while circumscribed margin (38.89% vs. 5.88%; P<0.001), vascularity grade 0-I (64.81% vs. 26.47%; P<0.001), and elasticity scores 1-3 (72.22% vs. 13.24%; P<0.001) were more frequently found in patients with SA. Patients with SA were significantly younger than were patients with IDC (43±11 vs. 54±11 years; P<0.001), and the lesion size was smaller in patients with SA than in those with IDC (median size 1.0 cm; interquartile range (IQR), 0.9 cm vs. median size 1.3 cm; IQR, 1.3 cm; P<0.001).

Conclusions

The preliminary results suggested that multimodal ultrasound imaging can improve the diagnostic accuracy of SA and provide additional information for differential diagnosis of SA and IDC.

A clinical-radiomics nomogram based on multimodal ultrasound for predicting the malignancy risk in solid hypoechoic breast lesions

Background

In routine clinical examinations, solid hypoechoic breast lesions are frequently encountered, but accurately distinguishing them poses a challenge. This study proposed a clinical-radiomics nomogram based on multimodal ultrasound that enhances the diagnostic accuracy for solid hypoechoic breast lesions.

Method

This retrospective study analyzed ultrasound strain elastography (SE) and automated breast volume scanner images (ABVS) of 423 solid hypoechoic breast lesions from 423 female patients in our hospital between August 2019 and May 2022. They were assigned to the training (n=296) and validation (n=127) groups in a 7:3 ratio by generating random numbers. Radiomics features were extracted and screened from ABVS and SE images, followed by the calculation of the radiomics score (Radscore) based on these features. Subsequently, a nomogram was constructed through multivariate logistic regression to assess the malignancy risk in breast lesions by combining Radscore with Breast Imaging Reporting and Data System (BI-RADS) scores and clinical risk factors associated with breast malignant lesions. The diagnostic performance, calibration performance, and clinical usefulness of the nomogram were assessed by the area under the curve (AUC) of the receiver operating characteristic curve, the calibration curve, and the decision analysis curve, respectively.

Results

The diagnostic performance of the nomogram is significantly superior to that of both the clinical diagnostic model (BI-RADS model) and the multimodal radiomics model (SE+ABVS radiomics model) in training (AUC: 0.972 vs 0.930 vs 0.941) and validation group (AUC:0.964 vs 0.916 vs 0.933). In addition, the nomogram also exhibited a favorable goodness-of-fit and could lead to greater net benefits for patients.

Conclusion

The nomogram enables a more effective assessment of the malignancy risk of solid hypoechoic breast lesions; therefore, it can serve as a new and efficient diagnostic tool for clinical diagnosis.

Optimized Radiomics Nomogram Based on Automated Breast Ultrasound System: A Potential Tool for Preoperative Prediction of Metastatic Lymph Node Burden in Breast Cancer

Background

Axillary lymph node dissection (ALND) can be safely avoided in women with T1 or T2 primary invasive breast cancer (BC) and one to two metastatic sentinel lymph nodes (SLNs). However, cancellation of ALND based solely on SLN biopsy (SLNB) may lead to adverse outcomes. Therefore, preoperative assessment of LN tumor burden becomes a new focus for ALN status.

Objective

This study aimed to develop and validate a nomogram incorporating the radiomics score (rad-score) based on automated breast ultrasound system (ABUS) and other clinicopathological features for evaluating the ALN status in patients with early-stage BC preoperatively.

Methods

Totally 354 and 163 patients constituted the training and validation cohorts. They were divided into ALN low burden (<3 metastatic LNs) and high burden (≥3 metastatic LNs) based on the histopathological diagnosis. The radiomics features of the segmented breast tumor in ABUS images were extracted and selected to generate the rad-score of each patient. These rad-scores, along with the ALN burden predictors identified from the clinicopathologic characteristics, were included in the multivariate analysis to establish a nomogram. It was further evaluated in the training and validation cohorts.

Results

High ALN burdens accounted for 11.2% and 10.8% in the training and validation cohorts. The rad-score for each patient was developed based on 7 radiomics features extracted from the ABUS images. The radiomics nomogram was built with the rad-score, tumor size, US-reported LN status, and ABUS retraction phenomenon. It achieved better predictive efficacy than the nomogram without the rad-score and exhibited favorable discrimination, calibration and clinical utility in both cohorts.

Conclusion

We developed an ABUS-based radiomics nomogram for the preoperative prediction of ALN burden in BC patients. It would be utilized for the identification of patients with low ALN burden if further validated, which contributed to appropriate axillary treatment and might avoid unnecessary ALND.

Prospective clinical research of radiomics and deep learning in oncology: A translational review

Radiomics and deep learning (DL) hold transformative promise and substantial and significant advances in oncology; however, most methods have been tested in retrospective or simulated settings. There is considerable interest in the biomarker validation, clinical utility, and methodological robustness of these studies and their deployment in real-world settings. This review summarizes the characteristics of studies, the level of prospective validation, and the overview of research on different clinical endpoints. The discussion of methodological robustness shows the potential for independent external replication of prospectively reported results. These in-depth analyses further describe the barriers limiting the translation of radiomics and DL into primary care options and provide specific recommendations regarding clinical deployment. Finally, we propose solutions for integrating novel approaches into the treatment environment to unravel the critical process of translating AI models into the clinical routine and explore strategies to improve personalized medicine.