Multiple parameters from ultrafast dynamic contrast-enhanced magnetic resonance imaging to discriminate between benign and malignant breast lesions: Comparison with apparent diffusion coefficient

Benign and Malignant Breast Lesions: Differentiation Using Microstructural Metrics Derived from Time-Dependent Diffusion MRI

Purpose To investigate the diagnostic performance of microstructural metrics from time-dependent diffusion MRI (Td-dMRI) in distinguishing between benign and malignant breast lesions. Materials and Methods This prospective study (ClinicalTrials.gov identifier: NCT05373628) enrolled participants with breast lesions confirmed with US, mammography, or both from January 2022 to June 2023. Participants underwent oscillating and pulsed gradient encoded Td-dMRI and conventional diffusion-weighted imaging (DWI). Td-dMRI data were fitted using the imaging microstructural parameters using limited spectrally edited diffusion model. Lesions were classified as benign or malignant based on pathology. Diagnostic performances of Td-dMRI metrics and apparent diffusion coefficients (ADCs) from DWI in distinguishing between benign and malignant tumors were assessed using receiver operating characteristic analysis and compared using the DeLong test. Results The study included 102 female participants (mean age: 48 years ± 12 [SD]) with 105 breast lesions (three participants had two lesions), including 31 benign and 74 malignant lesions. The cell diameter, cell density, and intracellular volume fraction from Td-dMRI were higher and the ADC was lower in malignant lesions compared with benign lesions (P < .001 to P = .001). Among microstructural metrics from Td-dMRI, the cell density had the highest area under the receiver operating characteristic curve, which was higher than that of the ADC (0.93 [95% CI: 0.88, 0.98] vs 0.79 [95% CI: 0.70, 0.88], P = .03). Conclusion A single microstructural metric derived from Td-dMRI, cell density, had higher performance than conventional ADC in distinguishing benign and malignant breast lesions. Keywords: MR-Diffusion Weighted Imaging, Breast Clinical trial registration no. NCT05373628 Supplemental material is available for this article. © RSNA, 2025.

Breast tumour classification in DCE-MRI via cross-attention and discriminant correlation analysis enhanced feature fusion

AIM

Dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) has proven to be highly sensitive in diagnosing breast tumours, due to the kinetic and volumetric features inherent in it. To utilise the kinetics-related and volume-related information, this paper aims to develop and validate a classification for differentiating benign and malignant breast tumours based on DCE-MRI, though fusing deep features and cross-attention-encoded radiomics features using discriminant correlation analysis (DCA).

MATERIALS AND METHODS

Classification experiments were conducted on a dataset comprising 261 individuals who underwent DCE-MRI including those with multiple tumours, resulting in 137 benign and 163 malignant tumours. To improve the strength of correlation between features and reduce features' redundancy, a novel fusion method that fuses deep features and encoded radiomics features based on DCA (eFF-DCA) is proposed. The eFF-DCA includes three components: (1) a feature extraction module to capture kinetic information across phases, (2) a radiomics feature encoding module employing a cross-attention mechanism to enhance inter-phase feature correlation, and (3) a DCA-based fusion module that transforms features to maximise intra-class correlation while minimising inter-class redundancy, facilitating effective classification.

RESULTS

The proposed eFF-DCA method achieved an accuracy of 90.9% and an area under the receiver operating characteristic curve of 0.942, outperforming methods using single-modal features.

CONCLUSION

The proposed eFF-DCA utilises DCE-MRI kinetic-related and volume-related features to improve breast tumour diagnosis accuracy, but non-end-to-end design limits multimodal fusion. Future research should explore unified end-to-end deep learning architectures that enable seamless multimodal feature fusion and joint optimisation of feature extraction and classification.

Dynamic contrast-enhanced magnetic resonance imaging parameters combined with diffusion-weighted imaging for discriminating malignant lesions, molecular subtypes, and pathological grades in invasive ductal carcinoma patients

Dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) parameters or diffusion-weighted imaging (DWI) findings provide prognostic information on breast cancer. However, the accuracy of a single MRI technique is unsatisfactory. This study intended to explore the combination of DWI and DCE-MRI parameters in discriminating molecular subtypes in invasive ductal carcinoma (IDC) patients. Eighty-two IDC patients who underwent breast DWI and DCE-MRI examinations were retrospectively analyzed. Eighty-six patients with benign masses were retrieved as benign controls. The combination of ADC value, Ktrans, Kep, Ve, and iAUC had a good ability to discriminate IDC patients (vs. benign controls) with an area under the curve (AUC) [95% confidence interval (CI)] of 0.961 (0.935-0.987). A nomogram-based prediction model with the above combination showed a good predictive value for IDC probability. The combination of ADC value, Ktrans, Kep, and iAUC also had a certain ability to discriminate pathological grade III (vs. I or II) [AUC (95% CI): 0.698 (0.572-0.825)] in IDC patients. Notably, ADC value (P=0.010) and Kep (P=0.043) differed in IDC patients with different molecular subtypes. Besides, ADC value was increased (P<0.001), but Ktrans (P=0.037) and Kep (P=0.004) were decreased in IDC patients with Lumina A (vs. other molecular subtypes). The combination of ADC value, Ktrans, Kep, had an acceptable ability to discriminate Luminal A (vs. other molecular subtypes) [AUC (95% CI): 0.845 (0.748-0.941)] in IDC patients. DWI combined with DCE-MRI parameters discriminates IDC from benign masses; it also identifies Luminal A and pathological grade III in IDC patients.

T1 Mapping-Derived Parameters in Breast Lesions: Diagnostic Accuracy and Correlation with Pathologic Features

RATIONALE AND OBJECTIVES

To evaluate the diagnostic potential of T1 mapping-derived parameters for distinguishing between benign and malignant breast tumors and their associations with pathologic prognostic indicators in invasive breast cancer.

MATERIALS AND METHODS

Patients who underwent breast surgery and quantitative magnetic resonance imaging (MRI), including apparent diffusion coefficient (ADC) and T1 mapping, between August 2023 and March 2024 were prospectively included. T1 parameters, including lesion T1 values before and after contrast agent injection (T10, T1c), reduction in T1 value (ΔT1), ratio of reduction (ΔT1%), extracellular volume fractions (ECVs), and ADC values were compared between benign and malignant breast lesions. The classification effect was evaluated via receiver operating characteristic (ROC) curves, and the correlation between MRI parameters and each prognostic indicator in invasive ductal carcinoma (IDC) was analyzed via Spearman correlation.

RESULTS

The ROC curves revealed that the area under the curve (AUC) of the ECV was slightly larger than that of the ADC (0.90 [95% CI: 0.84-0.95] vs 0.89 [95% CI: 0.83-0.94]). The combined diagnostic model of all parameters had the highest AUC (0.95 [95% CI: 0.90-0.98]). In IDC, ECV was positively correlated with the expression of estrogen receptor (r = 0.449, P < .001) and progesterone receptor (r = 0.433, P < .001) and negatively correlated with Ki-67 protein expression (r = -0.407, P < .001). No correlation was found between the ADC values and prognostic indicators.

CONCLUSION

T1 parameters can effectively differentiate benign and malignant breast lesions and have potential utility in predicting tumor invasiveness.

Ultrafast Breast MRI: A Narrative Review

Breast magnetic resonance imaging (MRI) is considered the most effective method for detecting breast cancer due to its high sensitivity. Yet multiple factors limit its widespread use, including high direct and indirect costs, a prolonged acquisition time with consequent patient discomfort, and a lack of trained radiologists. During the last decade, new strategies have been followed to increase the availability of breast MRI, including the omission of non-essential sequences to generate abbreviated MRI protocols (AB-MRIs) aimed at reducing the acquisition time with the potential of improving the patient's experience and accommodating a higher number of MRI examinations per day. An alternative method is ultrafast MRI (UF-MRI), a novel technique that gathers kinetic data within the first minute after contrast injection, offering high temporal resolution. This enables the analysis of early contrast wash-in curves, showing promising outcomes. In this study, we reviewed the role of UF-MRI in breast imaging and detailed how the integration of this new approach with radiomics and mathematical models might further improve diagnostic accuracy and even have a prognostic role, a fundamental characteristic in the modern scenarios of personalized medicine. In addition, possible clinical applications and advantages of UF-MRI will be discussed.

A Machine Learning Model for Predicting the HER2 Positive Expression of Breast Cancer Based on Clinicopathological and Imaging Features

RATIONALE AND OBJECTIVES

To develop a machine learning (ML) model based on clinicopathological and imaging features to predict the Human Epidermal Growth Factor Receptor 2 (HER2) positive expression (HER2-p) of breast cancer (BC), and to compare its performance with that of a logistic regression (LR) model.

MATERIALS AND METHODS

A total of 2541 consecutive female patients with pathologically confirmed primary breast lesions were enrolled in this study. Based on chronological order, 2034 patients treated between January 2018 and December 2022 were designated as the retrospective development cohort, while 507 patients treated between January 2023 and May 2024 were designated as the prospective validation cohort. The patients were randomly divided into a train cohort (n=1628) and a test cohort (n=406) in an 8:2 ratio within the development cohort. Pretreatment mammography (MG) and breast MRI data, along with clinicopathological features, were recorded. Extreme Gradient Boosting (XGBoost) in combination with Artificial Neural Network (ANN) and multivariate LR analyses were employed to extract features associated with HER2 positivity in BC and to develop an ANN model (using XGBoost features) and an LR model, respectively. The predictive value was assessed using a receiver operating characteristic (ROC) curve.

RESULTS

Following the application of Recursive Feature Elimination with Cross-Validation (RFE-CV) for feature dimensionality reduction, the XGBoost algorithm identified tumor size, suspicious calcifications, Ki-67 index, spiculation, and minimum apparent diffusion coefficient (minimum ADC) as key feature subsets indicative of HER2-p in BC. The constructed ANN model consistently outperformed the LR model, achieving the area under the curve (AUC) of 0.853 (95% CI: 0.837-0.872) in the train cohort, 0.821 (95% CI: 0.798-0.853) in the test cohort, and 0.809 (95% CI: 0.776-0.841) in the validation cohort.

CONCLUSION

The ANN model, built using the significant feature subsets identified by the XGBoost algorithm with RFE-CV, demonstrates potential in predicting HER2-p in BC.

Differentiation between adrenocortical carcinoma and lipid-poor adrenal adenoma using a multiparametric MRI-based diagnostic algorithm

PURPOSE

The purpose of this study was to evaluate the capabilities of multiparametric magnetic resonance imaging (MRI) in differentiating between lipid-poor adrenal adenoma (LPAA) and adrenocortical carcinoma (ACC).

MATERIALS AND METHODS

Patients of two centers who underwent surgical resection of LPAA or ACC after multiparametric MRI were retrospectively included. A training cohort was used to build a diagnostic algorithm obtained through recursive partitioning based on multiparametric MRI variables, including apparent diffusion coefficient and chemical shift signal ratio (i.e., tumor signal intensity index). The diagnostic performances of the multiparametric MRI-based algorithm were evaluated using a validation cohort, alone first and then in association with adrenal tumor size using a cut-off of 4 cm. Performances of the diagnostic algorithm for the diagnosis of ACC vs. LPAA were calculated using pathology as the reference standard.

RESULTS

Fifty-four patients (27 with LPAA and 27 with ACC; 37 women; mean age, 48.5 ± 13.3 [standard deviation (SD)] years) were used as the training cohort and 61 patients (24 with LPAA and 37 with ACC; 47 women; mean age, 49 ± 11.7 [SD] years) were used as the validation cohort. In the validation cohort, the diagnostic algorithm yielded best accuracy for the diagnosis of ACC vs. LPAA (75%; 46/61; 95% CI: 55-88) when used without lesion size. Best sensitivity was obtained with the association of the diagnostic algorithm with tumor size (96%; 23/24; 95% CI: 80-99). Best specificity was obtained with the diagnostic algorithm used alone (76%; 28/37; 95% CI: 60-87).

CONCLUSION

A multiparametric MRI-based diagnostic algorithm that includes apparent diffusion coefficient and tumor signal intensity index helps discriminate between ACC and LPAA with high degrees of specificity and accuracy. The association of the multiparametric MRI-based diagnostic algorithm with adrenal lesion size helps maximize the sensitivity of multiparametric MRI for the diagnosis of ACC.

The diagnostic performance of ultrafast MRI to differentiate benign from malignant breast lesions: a systematic review and meta-analysis

OBJECTIVES

To assess the diagnostic performance of ultrafast magnetic resonance imaging (UF-DCE MRI) in differentiating benign from malignant breast lesions.

MATERIALS AND METHODS

A comprehensive search was conducted until September 1, 2023, in Medline, Embase, and Cochrane databases. Clinical studies evaluating the diagnostic performance of UF-DCE MRI in breast lesion stratification were screened and included in the meta-analysis. Pooled summary estimates for sensitivity, specificity, diagnostic odds ratio (DOR), and hierarchic summary operating characteristics (SROC) curves were pooled under the random-effects model. Publication bias and heterogeneity between studies were calculated.

RESULTS

A final set of 16 studies analyzing 2090 lesions met the inclusion criteria and were incorporated into the meta-analysis. Using UF-DCE MRI kinetic parameters, the pooled sensitivity, specificity, DOR, and area under the curve (AUC) for differentiating benign from malignant breast lesions were 83% (95% CI 79-88%), 77% (95% CI 72-83%), 18.9 (95% CI 13.7-26.2), and 0.876 (95% CI 0.83-0.887), respectively. We found no significant difference in diagnostic accuracy between the two main UF-DCE MRI kinetic parameters, maximum slope (MS) and time to enhancement (TTE). DOR and SROC exhibited low heterogeneity across the included studies. No evidence of publication bias was identified (p = 0.585).

CONCLUSIONS

UF-DCE MRI as a stand-alone technique has high accuracy in discriminating benign from malignant breast lesions.

CLINICAL RELEVANCE STATEMENT

UF-DCE MRI has the potential to obtain kinetic information and stratify breast lesions accurately while decreasing scan times, which may offer significant benefit to patients.

KEY POINTS

• Ultrafast breast MRI is a novel technique which captures kinetic information with very high temporal resolution. • The kinetic parameters of ultrafast breast MRI demonstrate a high level of accuracy in distinguishing between benign and malignant breast lesions. • There is no significant difference in accuracy between maximum slope and time to enhancement kinetic parameters.

Ultrafast Dynamic Contrast-Enhanced MRI of the Breast: From Theory to Practice

The development of ultrafast dynamic contrast-enhanced (UF-DCE) MRI has occurred in tandem with fast MRI scan techniques, particularly view-sharing and compressed sensing. Understanding the strengths of each technique and optimizing the relevant parameters are essential to their implementation. UF-DCE MRI has now shifted from research protocols to becoming a part of clinical scan protocols for breast cancer. UF-DCE MRI is expected to compensate for the low specificity of abbreviated MRI by adding kinetic information from the upslope of the time-intensity curve. Because kinetic information from UF-DCE MRI is obtained from the shape and timing of the initial upslope, various new kinetic parameters have been proposed. These parameters may be associated with receptor status or prognostic markers for breast cancer. In addition to the diagnosis of malignant lesions, more emphasis has been placed on predicting and evaluating treatment response because hyper-vascularity is linked to the aggressiveness of breast cancers. In clinical practice, it is important to note that breast lesion images obtained from UF-DCE MRI are slightly different from those obtained by conventional DCE MRI in terms of morphology. A major benefit of using UF-DCE MRI is avoidance of the marked or moderate background parenchymal enhancement (BPE) that can obscure the target enhancing lesions. BPE is less prominent in the earlier phases of UF-DCE MRI, which offers better lesion-to-noise contrast. The excellent contrast of early-enhancing vessels provides a key to understanding the detailed pathological structure of tumor-associated vessels. UF-DCE MRI is normally accompanied by a large volume of image data for which automated/artificial intelligence-based processing is expected to be useful. In this review, both the theoretical and practical aspects of UF-DCE MRI are summarized. EVIDENCE LEVEL: 5 TECHNICAL EFFICACY: Stage 2.

Predictive value of MRI-based deep learning model for lymphovascular invasion status in node-negative invasive breast cancer

To retrospectively assess the effectiveness of deep learning (DL) model, based on breast magnetic resonance imaging (MRI), in predicting preoperative lymphovascular invasion (LVI) status in patients diagnosed with invasive breast cancer who have negative axillary lymph nodes (LNs). Data was gathered from 280 patients, including 148 with LVI-positive and 141 with LVI-negative lesions. These patients had undergone preoperative breast MRI and were histopathologically confirmed to have invasive breast cancer without axillary LN metastasis. The cohort was randomly split into training and validation groups in a 7:3 ratio. Radiomics features for each lesion were extracted from the first post-contrast dynamic contrast-enhanced (DCE)-MRI. The Least Absolute Shrinkage and Selection Operator (LASSO) regression method and logistic regression analyses were employed to identify significant radiomic features and clinicoradiological variables. These models were established using four machine learning (ML) algorithms and one DL algorithm. The predictive performance of the models (radiomics, clinicoradiological, and combination) was assessed through discrimination and compared using the DeLong test. Four clinicoradiological parameters and 10 radiomic features were selected by LASSO for model development. The Multilayer Perceptron (MLP) model, constructed using both radiomic and clinicoradiological features, demonstrated excellent performance in predicting LVI, achieving a high area under the curve (AUC) of 0.835 for validation. The DL model (MLP-radiomic) achieved the highest accuracy (AUC = 0.896), followed by DL model (MLP-combination) with an AUC of 0.835. Both DL models were significantly superior to the ML model (RF-clinical) with an AUC of 0.720. The DL model (MLP), which integrates radiomic features with clinicoradiological information, effectively aids in the preoperative determination of LVI status in patients with invasive breast cancer and negative axillary LNs. This is beneficial for making informed clinical decisions.

Bi-regional dynamic contrast-enhanced MRI for prediction of microvascular invasion in solitary BCLC stage A hepatocellular carcinoma

OBJECTIVES

To construct a combined model based on bi-regional quantitative dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI), as well as clinical-radiological (CR) features for predicting microvascular invasion (MVI) in solitary Barcelona Clinic Liver Cancer (BCLC) stage A hepatocellular carcinoma (HCC), and to assess its ability for stratifying the risk of recurrence after hepatectomy.

METHODS

Patients with solitary BCLC stage A HCC were prospective collected and randomly divided into training and validation sets. DCE perfusion parameters were obtained both in intra-tumoral region (ITR) and peritumoral region (PTR). Combined DCE perfusion parameters (CDCE) were constructed to predict MVI. The combined model incorporating CDCE and CR features was developed and evaluated. Kaplan-Meier method was used to investigate the prognostic significance of the model and the survival benefits of different hepatectomy approaches.

RESULTS

A total of 133 patients were included. Total blood flow in ITR and arterial fraction in PTR exhibited the best predictive performance for MVI with areas under the curve (AUCs) of 0.790 and 0.792, respectively. CDCE achieved AUCs of 0.868 (training set) and 0.857 (validation set). A combined model integrated with the α-fetoprotein, corona enhancement, two-trait predictor of venous invasion, and CDCE could improve the discrimination ability to AUCs of 0.966 (training set) and 0.937 (validation set). The combined model could stratify the prognosis of HCC patients. Anatomical resection was associated with a better prognosis in the high-risk group (p < 0.05).

CONCLUSION

The combined model integrating DCE perfusion parameters and CR features could be used for MVI prediction in HCC patients and assist clinical decision-making.

CRITICAL RELEVANCE STATEMENT

The combined model incorporating bi-regional DCE-MRI perfusion parameters and CR features predicted MVI preoperatively, which could stratify the risk of recurrence and aid in optimizing treatment strategies.

KEY POINTS

Microvascular invasion (MVI) is a significant predictor of prognosis for hepatocellular carcinoma (HCC). Quantitative DCE-MRI could predict MVI in solitary BCLC stage A HCC; the combined model improved performance. The combined model could help stratify the risk of recurrence and aid treatment planning.

Optimizing ultrafast dynamic contrast-enhanced MRI scan duration in the differentiation of benign and malignant breast lesions

OBJECTIVE

To determine the optimal scan duration for ultrafast DCE-MRI in effectively differentiating benign from malignant breast lesions.

METHODS

The study prospectively recruited participants who underwent breast ultrafast DCE-MRI from September 2021 to March 2023. A 30-phase breast ultrafast DCE-MRI on a 3.0-T MRI system was conducted with a 4.5-s temporal resolution. Scan durations ranged from 40.5 s to 135.0 s, during which the analysis is performed at three-phase intervals, forming eight dynamic sets (scan duration [SD]40.5s: 40.5 s, SD54s: 54.0 s, SD67.5s: 67.5 s, SD81s: 81.0 s, SD94.5s: 94.5 s, SD108s: 108.0 s, SD121.5s: 121.5 s, and SD135s: 135.0 s). Two ultrafast DCE-MRI parameters, maximum slope (MS) and initial area under the curve in 60 s (iAUC), were calculated for each dynamic set and compared between benign and malignant lesions. Areas under the receiver operating characteristic curve (AUCs) were used to assess their diagnostic performance.

RESULTS

A total of 140 women (mean age, 47 ± 11 years) with 151 lesions were included. MS and iAUC from eight dynamic sets exhibited significant differences between benign and malignant lesions (all p < 0.05), except iAUC at SD40.5s. The AUC of MS (AUC = 0.804) and iAUC (AUC = 0.659) at SD67.5s were significantly higher than their values at SD40.5s (AUC = 0.606 and 0.516; corrected p < 0.05). No significant differences in AUCs for MS and iAUC were observed from SD67.5s to SD135s (all corrected p > 0.05).

CONCLUSIONS

Ultrafast DCE-MRI with a 67.5-s scan duration appears optimal for effectively differentiating malignant from benign breast lesions.

CRITICAL RELEVANCE STATEMENT

By evaluating scan durations (40.5-135 s) and analyzing two ultrafast DCE-MRI parameters, we found a scan duration of 67.5 s optimal for discriminating between these lesions and offering a balance between acquisition time and diagnostic efficacy.

KEY POINTS

Ultrafast DCE-MRI can effectively differentiate malignant from benign breast lesions. A minimum of 67.5-sec ultrafast DCE-MRI scan duration is required to differentiate benign and malignant lesions. Extending the scan duration beyond 67.5 s did not significantly improve diagnostic accuracy.

Multiparametric MRI model to predict molecular subtypes of breast cancer using Shapley additive explanations interpretability analysis

PURPOSE

The purpose of this study was to assess the predictive performance of multiparametric magnetic resonance imaging (MRI) for molecular subtypes and interpret features using SHapley Additive exPlanations (SHAP) analysis.

MATERIAL AND METHODS

Patients with breast cancer who underwent pre-treatment MRI (including ultrafast dynamic contrast-enhanced MRI, magnetic resonance spectroscopy, diffusion kurtosis imaging and intravoxel incoherent motion) were recruited between February 2019 and January 2022. Thirteen semantic and thirteen multiparametric features were collected and the key features were selected to develop machine-learning models for predicting molecular subtypes of breast cancers (luminal A, luminal B, triple-negative and HER2-enriched) by using stepwise logistic regression. Semantic model and multiparametric model were built and compared based on five machine-learning classifiers. Model decision-making was interpreted using SHAP analysis.

RESULTS

A total of 188 women (mean age, 53 ± 11 [standard deviation] years; age range: 25-75 years) were enrolled and further divided into training cohort (131 women) and validation cohort (57 women). XGBoost demonstrated good predictive performance among five machine-learning classifiers. Within the validation cohort, the areas under the receiver operating characteristic curves (AUCs) for the semantic models ranged from 0.693 (95% confidence interval [CI]: 0.478-0.839) for HER2-enriched subtype to 0.764 (95% CI: 0.681-0.908) for luminal A subtype, inferior to multiparametric models that yielded AUCs ranging from 0.771 (95% CI: 0.630-0.888) for HER2-enriched subtype to 0.857 (95% CI: 0.717-0.957) for triple-negative subtype. The AUCs between the semantic and the multiparametric models did not show significant differences (P range: 0.217-0.640). SHAP analysis revealed that lower iAUC, higher kurtosis, lower D*, and lower kurtosis were distinctive features for luminal A, luminal B, triple-negative breast cancer, and HER2-enriched subtypes, respectively.

CONCLUSION

Multiparametric MRI is superior to semantic models to effectively predict the molecular subtypes of breast cancer.