Intra- and peritumoral radiomics on assessment of breast cancer molecular subtypes based on mammography and MRI

Predicting axillary lymph node metastasis in breast cancer patients: A radiomics-based multicenter approach with interpretability analysis

PURPOSE

To develop a MRI-based radiomics model, integrating the intratumoral and peritumoral imaging information to predict axillary lymph node metastasis (ALNM) in patients with breast cancer and to elucidate the model's decision-making process via interpretable algorithms.

METHODS

This study included 376 patients from three institutions who underwent contrast-enhanced breast MRI between 2021 and 2023. We used multiple machine learning algorithms to combine peritumoral, intratumoral, and radiological characteristics with the building of radiological, radiomics, and combined models. The model's performance was compared based on the area under the curve (AUC) obtained from the receiver operating characteristic analysis and interpretable machine learning techniques to analyze the operating mechanism of the model.

RESULTS

The radiomics model, incorporating features from both intratumoral tissue and the 3 mm peritumoral region and utilizing the backpropagation neural network (BPNN) algorithm, demonstrated superior diagnostic efficacy, achieving an AUC of 0.820. The AUC of the combination of the RAD score, clinical T stage, and spiculated margin was as high as 0.855. Furthermore, we conducted SHapley Additive exPlanations (SHAP) analysis to evaluate the contributions of RAD score, clinical T stage, and spiculated margin in ALNM status prediction.

CONCLUSIONS

The interpretable radiomics model we propose can better predict the ALNM status of breast cancer and help inform clinical treatment decisions.

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

RATIONALE AND OBJECTIVES

Accurate assessment of lymphovascular invasion (LVI) in invasive breast cancer (IBC) plays a pivotal role in tailoring personalized treatment plans. This study aimed to investigate habitats-based spatial distributions to quantitatively measure tumor heterogeneity on multiparametric magnetic resonance imaging (MRI) scans and assess their predictive capability for LVI in patients with IBC.

MATERIALS AND METHODS

In this retrospective cohort study, we consecutively enrolled 241 women diagnosed with IBC between July 2020 and July 2023 and who had 1.5 T/T1-weighted images, fat-suppressed T2-weighted images, and dynamic contrast-enhanced MRI. Habitats-based spatial distributions were derived from the gross tumor volume (GTV) and gross tumor volume plus peritumoral volume (GPTV). GTV_habitats and GPTV_habitats were generated through sub-region segmentation, and their performances were compared. Subsequently, a combined nomogram was developed by integrating relevant spatial distributions with the identified MR morphological characteristics. Diagnostic performance was compared using receiver operating characteristic curve analysis and decision curve analysis. Statistical significance was set at p < 0.05.

RESULTS

GPTV_habitats exhibited superior performance compared to GTV_habitats. Consequently, the GPTV_habitats, diffusion-weighted imaging rim signs, and peritumoral edema were integrated to formulate the combined nomogram. This combined nomogram outperformed individual MR morphological characteristics and the GPTV_habitats index, achieving area under the curve values of 0.903 (0.847 -0.959), 0.770 (0.689 -0.852), and 0.843 (0.776 -0.910) in the training set and 0.931 (0.863 -0.999), 0.747 (0.613 -0.880), and 0.849 (0.759 -0.938) in the validation set.

CONCLUSION

The combined nomogram incorporating the GPTV_habitats and identified MR morphological characteristics can effectively predict LVI in patients with IBC.

Molecular subtypes of breast cancer identified by dynamically enhanced MRI radiomics: the delayed phase cannot be ignored

OBJECTIVES

To compare the diagnostic performance of intratumoral and peritumoral features from different contrast phases of breast dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) by building radiomics models for differentiating molecular subtypes of breast cancer.

METHODS

This retrospective study included 377 patients with pathologically confirmed breast cancer. Patients were divided into training set (n = 202), validation set (n = 87) and test set (n = 88). The intratumoral volume of interest (VOI) and peritumoral VOI were delineated on primary breast cancers at three different DCE-MRI contrast phases: early, peak, and delayed. Radiomics features were extracted from each phase. After feature standardization, the training set was filtered by variance analysis, correlation analysis, and least absolute shrinkage and selection (LASSO). Using the extracted features, a logistic regression model based on each tumor subtype (Luminal A, Luminal B, HER2-enriched, triple-negative) was established. Ten models based on intratumoral or/plus peritumoral features from three different phases were developed for each differentiation.

RESULTS

Radiomics features extracted from delayed phase DCE-MRI demonstrated dominant diagnostic performance over features from other phases. However, the differences were not statistically significant. In the full fusion model for differentiating different molecular subtypes, the most frequently screened features were those from the delayed phase. According to the Shapley additive explanation (SHAP) method, the most important features were also identified from the delayed phase.

CONCLUSIONS

The intratumoral and peritumoral radiomics features from the delayed phase of DCE-MRI can provide additional information for preoperative molecular typing. The delayed phase of DCE-MRI cannot be ignored.

CRITICAL RELEVANCE STATEMENT

Radiomics features extracted and radiomics models constructed from the delayed phase of DCE-MRI played a crucial role in molecular subtype classification, although no significant difference was observed in the test cohort.

KEY POINTS

The molecular subtype of breast cancer provides a basis for setting treatment strategy and prognosis. The delayed-phase radiomics model outperformed that of early-/peak-phases, but no differently than other phases or combinations. Both intra- and peritumoral radiomics features offer valuable insights for molecular typing.

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

RATIONALE AND OBJECTIVES

This study aimed to create a nomogram model that combines clinical factors with radiomics analysis of both intra- and peritumoral regions extracted from preoperative digital breast tomosynthesis (DBT) images, in order to develop a reliable method for predicting the lymphovascular invasion (LVI) status in invasive breast cancer (IBC) patients.

MATERIALS AND METHODS

A total of 178 patients were randomly split into a training dataset (N = 124) and a validation dataset (N = 54). Comprehensive clinical data, encompassing DBT features, were gathered for all cases. Radiomics features were extracted and selected from intra- and peritumoral region to establish radiomics signature (Radscore). To construct the clinical model and nomogram model, univariate and multivariate logistic regression analyses were utilized to identify independent risk factors. To assess and validate these models, various analytical methods were employed, including receiver operating characteristic (ROC) curve analysis, calibration curve analysis, decision curve analysis (DCA), net reclassification improvement (NRI), and integrated discriminatory improvement (IDI).

RESULTS

The clinical model is constructed based on two independent risk factors: tumor margin and the DBT-reported lymph node metastasis (DBT_reported_LNM). Incorporating Radscore_Combine (utilizing both intra- and peritumoral radiomics features), tumor margin, and DBT_reported_LNM into the nomogram achieved a reliable predictive performance, with area under the curve (AUC) values of 0.906 and 0.905 in both datasets, respectively. The significant improvement demonstrated by the NRI and IDI indicates that the Radscore_Combine could be a valuable biomarker for effectively predicting the status of LVI.

CONCLUSION

The nomogram demonstrated a reliable ability to predict LVI in IBC patients.

MOB-CBAM: A dual-channel attention-based deep learning generalizable model for breast cancer molecular subtypes prediction using mammograms

BACKGROUND AND OBJECTIVE

Deep Learning models have emerged as a significant tool in generating efficient solutions for complex problems including cancer detection, as they can analyze large amounts of data with high efficiency and performance. Recent medical studies highlight the significance of molecular subtype detection in breast cancer, aiding the development of personalized treatment plans as different subtypes of cancer respond better to different therapies.

METHODS

In this work, we propose a novel lightweight dual-channel attention-based deep learning model MOB-CBAM that utilizes the backbone of MobileNet-V3 architecture with a Convolutional Block Attention Module to make highly accurate and precise predictions about breast cancer. We used the CMMD mammogram dataset to evaluate the proposed model in our study. Nine distinct data subsets were created from the original dataset to perform coarse and fine-grained predictions, enabling it to identify masses, calcifications, benign, malignant tumors and molecular subtypes of cancer, including Luminal A, Luminal B, HER-2 Positive, and Triple Negative. The pipeline incorporates several image pre-processing techniques, including filtering, enhancement, and normalization, for enhancing the model's generalization ability.

RESULTS

While identifying benign versus malignant tumors, i.e., coarse-grained classification, the MOB-CBAM model produced exceptional results with 99 % accuracy, precision, recall, and F1-score values of 0.99 and MCC of 0.98. In terms of fine-grained classification, the MOB-CBAM model has proven to be highly efficient in accurately identifying mass with (benign/malignant) and calcification with (benign/malignant) classification tasks with an impressive accuracy rate of 98 %. We have also cross-validated the efficiency of the proposed MOB-CBAM deep learning architecture on two datasets: MIAS and CBIS-DDSM. On the MIAS dataset, an accuracy of 97 % was reported for the task of classifying benign, malignant, and normal images, while on the CBIS-DDSM dataset, an accuracy of 98 % was achieved for the classification of mass with either benign or malignant, and calcification with benign and malignant tumors.

CONCLUSION

This study presents lightweight MOB-CBAM, a novel deep learning framework, to address breast cancer diagnosis and subtype prediction. The model's innovative incorporation of the CBAM enhances precise predictions. The extensive evaluation of the CMMD dataset and cross-validation on other datasets affirm the model's efficacy.

Machine learning based on optimal VOI of multi-sequence MR images to predict lymphovascular invasion in invasive breast cancer

Objectives

Lymphovascular invasion serves as a crucial prognostic indicator in invasive breast cancer, influencing treatment decisions. We aimed to develop a machine learning model utilizing optimal volumes of interest extracted from multisequence magnetic resonance images to predict lymphovascular invasion in patients with invasive breast cancer.

Materials and methods

This study comprised 191 patients postoperatively diagnosed with invasive breast cancer through multi-sequence magnetic resonance imaging. Independent predictors were identified through univariate and multivariate logistic regression analyses, culminating in the construction of a clinical model. Radiomic features were extracted from multi-sequence magnetic resonance imaging images across various volume of interest scales (-2 mm, entire, +2 mm, +4 mm, and +6 mm). Subsequently, various radiomic models were developed using machine learning model algorithms, including logistic regression, support vector machine, k-nearest neighbor, gradient boosting machine, classification and regression tree, and random forest. A hybrid model was then formulated, amalgamating optimal radiomic and clinical models.

Results

The area under the curve of the clinical model was 0.757. Among the radiomic models, the most efficient diagnosis was achieved by the k-nearest neighbor-based radiomics-volume of interest (+2 mm), resulting in an area under the curve of 0.780. The hybrid model, integrating the k-nearest neighbor-based radiomics-volume of interest (+2 mm), and the clinical model surpassed the individual clinical and radiomics models, exhibiting a superior area under the curve of 0.864.

Conclusion

Utilizing a hybrid approach integrating clinical data and multi-sequence magnetic resonance imaging-derived radiomics models based on the multiscale tumor region volume of interest (+2 mm) proved effective in determining lymphovascular invasion status in patients with invasive breast cancer. This innovative methodology may offer valuable insights for treatment planning and disease management.

Ultrasound Radiomics Features to Identify Patients With Triple-Negative Breast Cancer: A Retrospective, Single-Center Study

OBJECTIVES

Patients with triple-negative breast cancer (TNBC) exhibit a fast tumor growth rate and poor survival outcomes. In this study, we aimed to develop and compare intelligent algorithms using ultrasound radiomics features in addition to clinical variables to identify patients with TNBC prior to histopathologic diagnosis.

METHODS

We used single-center, retrospective data of patients who underwent ultrasound before histopathologic verification and subsequent neoadjuvant systemic treatment (NAST). We developed a logistic regression with an elastic net penalty algorithm using pretreatment ultrasound radiomics features in addition to patient and tumor variables to identify patients with TNBC. Findings were compared to the histopathologic evaluation of the biopsy specimen. The main outcome measure was the area under the curve (AUC).

RESULTS

We included 1161 patients, 813 in the development set and 348 in the validation set. Median age was 50.1 years and 24.4% (283 of 1161) had TNBC. The integrative model using radiomics and clinical information showed significantly better performance in identifying TNBC compared to the radiomics model (AUC: 0.71, 95% confidence interval [CI]: 0.65-0.76 versus 0.64, 95% CI: 0.57-0.71, P = .004). The five most important variables were cN status, shape surface volume ratio (SA:V), gray level co-occurrence matrix (GLCM) correlation, gray level dependence matrix (GLDM) dependence nonuniformity normalized, and age. Patients with TNBC were more often categorized as BI-RADS 4 than BI-RADS 5 compared to non-TNBC patients (P = .002).

CONCLUSION

A machine learning algorithm showed promising potential to identify patients with TNBC using ultrasound radiomics features and clinical information prior to histopathologic evaluation.

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

BACKGROUND

Radiomics has been applied for assessing lymphovascular invasion (LVI) in patients with breast cancer. However, associations between features from peritumoral regions and the LVI status were not investigated.

PURPOSE

To investigate the value of intra- and peritumoral radiomics for assessing LVI, and to develop a nomogram to assist in making treatment decisions.

STUDY TYPE

Retrospective.

POPULATION

Three hundred and sixteen patients were enrolled from two centers and divided into training (N = 165), internal validation (N = 83), and external validation (N = 68) cohorts.

FIELD STRENGTH/SEQUENCE

1.5 T and 3.0 T/dynamic contrast-enhanced (DCE) and diffusion-weighted imaging (DWI).

ASSESSMENT

Radiomics features were extracted and selected based on intra- and peritumoral breast regions in two magnetic resonance imaging (MRI) sequences to create the multiparametric MRI combined radiomics signature (RS-DCE plus DWI). The clinical model was built with MRI-axillary lymph nodes (MRI ALN), MRI-reported peritumoral edema (MPE), and apparent diffusion coefficient (ADC). The nomogram was constructed with RS-DCE plus DWI, MRI ALN, MPE, and ADC.

STATISTICAL TESTS

Intra- and interclass correlation coefficient analysis, Mann-Whitney U test, and least absolute shrinkage and selection operator regression were used for feature selection. Receiver operating characteristic and decision curve analyses were applied to compare performance of the RS-DCE plus DWI, clinical model, and nomogram.

RESULTS

A total of 10 features were found to be associated with LVI, 3 from intra- and 7 from peritumoral areas. The nomogram showed good performance in the training (AUCs, nomogram vs. clinical model vs. RS-DCE plus DWI, 0.884 vs. 0.695 vs. 0.870), internal validation (AUCs, nomogram vs. clinical model vs. RS-DCE plus DWI, 0.813 vs. 0.695 vs. 0.794), and external validation (AUCs, nomogram vs. clinical model vs. RS-DCE plus DWI, 0.862 vs. 0.601 vs. 0.849) cohorts.

DATA CONCLUSION

The constructed preoperative nomogram might effectively assess LVI.

LEVEL OF EVIDENCE

3 TECHNICAL EFFICACY: Stage 2.

A Mammography-Based Radiomic Nomogram for Predicting Malignancy in Breast Suspicious Microcalcifications

RATIONALE AND OBJECTIVES

Preoperative accurate identification of benign and malignant breast lesions is vital for patients to achieve individualized treatment. This study aimed to develop and validate a mammography-based radiomic nomogram for predicting malignant risk of breast suspicious microcalcifications (MCs).

MATERIALS AND METHODS

496 patients with histologically confirmed breast suspicious MCs were randomly divided into the training set (n = 346) and validation set (n = 150). Radiomics features was extracted from the craniocaudal and mediolateral oblique images. Least absolute shrinkage and selection operator algorithm were used to select radiomics features, then radiomics score (Rad-score) was calculated. Univariate analysis was used to identify malignant MCs-related clinical independent risk factors. Multivariate logistic regression was used to establish a clinical-radiomics model by incorporating Rad-score and clinic factors. A nomogram was developed to visualize the clinical-radiomics model. The receiver operating characteristic curve, calibration curve and decision curve analysis (DCA) were used to evaluate the performance of the nomogram.

RESULTS

The Rad-score was consisted of 29 optimal radiomics features. We developed a nomogram by incorporating Rad-score, menopause status, MCs morphology and distribution, the area under the curve value of the combined model was 0.926(95% confidence interval [CI]: 0.878-0.975) for the validation set. The calibration curves and DCA indicated the combined model had favorable calibration and clinical utility.

CONCLUSION

The combined model could be considered as a potential imaging marker to predict malignant risk of breast suspicious MCs.

Radiomics signatures for predicting the Ki-67 level and HER-2 status based on bone metastasis from primary breast cancer

This study explores the potential of radiomics to predict the proliferation marker protein Ki-67 levels and human epidermal growth factor receptor 2 (HER-2) status based on MRI images of patients with spinal metastasis from primary breast cancer. A total of 110 patients with pathologically confirmed spinal metastases from primary breast cancer were enrolled between Dec. 2017 and Dec. 2021. All patients underwent T1-weighted contrast-enhanced MRI scans. The PyRadiomics package was used to extract features from the MRI images based on the intraclass correlation coefficient and least absolute shrinkage and selection operator. The most predictive features were used to develop the radiomics signature. The Chi-Square test, Fisher's exact test, Student's t-test, and Mann-Whitney U test were used to evaluate the clinical and pathological characteristics between the high- and low-level Ki-67 groups and the HER-2 positive/negative groups. The radiomics models were compared using receiver operating characteristic curve analysis. The area under the receiver operating characteristic curve (AUC), sensitivity (SEN), and specificity (SPE) were generated as comparison metrics. From the spinal MRI scans, five and two features were identified as the most predictive for the Ki-67 level and HER-2 status, respectively. The developed radiomics signatures generated good prediction performance for the Ki-67 level in the training (AUC = 0.812, 95% CI: 0.710-0.914, SEN = 0.667, SPE = 0.846) and validation (AUC = 0.799, 95% CI: 0.652-0.947, SEN = 0.722, SPE = 0.833) cohorts. Good prediction performance for the HER-2 status was also achieved in the training (AUC = 0.796, 95% CI: 0.686-0.906, SEN = 0.720, SPE = 0.776) and validation (AUC = 0.705, 95% CI: 0.506-0.904, SEN = 0.733, SPE = 0.762) cohorts. The results of this study provide a better understanding of the potential clinical implications of spinal MRI-based radiomics on the prediction of Ki-67 levels and HER-2 status in breast cancer.

Identification of Luminal A breast cancer by using deep learning analysis based on multi-modal images

Purpose

To evaluate the diagnostic performance of a deep learning model based on multi-modal images in identifying molecular subtype of breast cancer.

Materials and methods

A total of 158 breast cancer patients (170 lesions, median age, 50.8 ± 11.0 years), including 78 Luminal A subtype and 92 non-Luminal A subtype lesions, were retrospectively analyzed and divided into a training set (n = 100), test set (n = 45), and validation set (n = 25). Mammography (MG) and magnetic resonance imaging (MRI) images were used. Five single-mode models, i.e., MG, T2-weighted imaging (T2WI), diffusion weighting imaging (DWI), axial apparent dispersion coefficient (ADC), and dynamic contrast-enhanced MRI (DCE-MRI), were selected. The deep learning network ResNet50 was used as the basic feature extraction and classification network to construct the molecular subtype identification model. The receiver operating characteristic curve were used to evaluate the prediction efficiency of each model.

Results

The accuracy, sensitivity and specificity of a multi-modal tool for identifying Luminal A subtype were 0.711, 0.889, and 0.593, respectively, and the area under the curve (AUC) was 0.802 (95% CI, 0.657- 0.906); the accuracy, sensitivity, and AUC were higher than those of any single-modal model, but the specificity was slightly lower than that of DCE-MRI model. The AUC value of MG, T2WI, DWI, ADC, and DCE-MRI model was 0.593 (95%CI, 0.436-0.737), 0.700 (95%CI, 0.545-0.827), 0.564 (95%CI, 0.408-0.711), 0.679 (95%CI, 0.523-0.810), and 0.553 (95%CI, 0.398-0.702), respectively.

Conclusion

The combination of deep learning and multi-modal imaging is of great significance for diagnosing breast cancer subtypes and selecting personalized treatment plans for doctors.

Using Radiomics to Differentiate Brain Metastases From Lung Cancer Versus Breast Cancer, Including Predicting Epidermal Growth Factor Receptor and human Epidermal Growth Factor Receptor 2 Status

OBJECTIVE

We evaluated the feasibility of using multiregional radiomics to identify brain metastasis (BM) originating from lung adenocarcinoma (LA) and breast cancer (BC) and assess the epidermal growth factor receptor (EGFR) mutation and human epidermal growth factor receptor 2 (HER2) status.

METHODS

Our experiment included 160 patients with BM originating from LA (n = 70), BC (n = 67), and other tumor types (n = 23), between November 2017 and December 2021. All patients underwent contrast-enhanced T1- and T2-weighted magnetic resonance imaging (MRI) scans. A total of 1967 quantitative MRI features were calculated from the tumoral active area and peritumoral edema area and selected using least absolute shrinkage and selection operator regression with 5-fold cross-validation. We constructed radiomic signatures (RSs) based on the most predictive features for preoperative assessment of the metastatic origins, EGFR mutation, and HER2 status. Prediction performance of the constructed RSs was evaluated based on the receiver operating characteristic curve analysis.

RESULTS

The developed multiregion RSs generated good area under the receiver operating characteristic curve (AUC) for identifying the LA and BC origin in the training (AUCs, RS-LA vs RS-BC, 0.767 vs 0.898) and validation (AUCs, RS-LA vs RS-BC, 0.778 and 0.843) cohort and for predicting the EGFR and HER2 status in the training (AUCs, RS-EGFR vs RS-HER2, 0.837 vs 0.894) and validation (AUCs, RS-EGFR vs RS-HER2, 0.729 vs 0.784) cohorts.

CONCLUSIONS

Our results revealed associations between brain MRI-based radiomics and their metastatic origins, EGFR mutations, and HER2 status. The developed multiregion combined RSs may be considered noninvasive predictive markers for planning early treatment for BM patients.

Improving the malignancy prediction of breast cancer based on the integration of radiomics features from dual-view mammography and clinical parameters

Radiomics has been a promising imaging biomarker for many malignant diseases. We developed a novel radiomics strategy that incorporating radiomics features extracted from dual-view mammograms and clinical parameters for identifying benign and malignant breast lesions, and validated whether the radiomics assessment could improve the accurate diagnosis of breast cancer. A total of 380 patients (mean age, 52 ± 7 years) with 621 breast lesions utilizing mammograms on craniocaudal (CC) and mediolateral oblique (MLO) views were randomly allocated into the training (n = 486) and testing (n = 135) sets in this retrospective study. A total of 1184 and 2368 radiomics features were extracted from single-position region of interest (ROI) and position-paired ROI, separately. Clinical parameters were then combined for better prediction. Recursive feature elimination and least absolute shrinkage and selection operator methods were applied to select optimal predictive features. Random forest was used to conduct the predictive model. Intraclass correlation coefficient test was used to assess repeatability and reproducibility of features. After preprocessing, 467 radiomics features and clinical parameters remained in the single-view and dual-view models. The performance and significance of models were quantified by the area under the curve (AUC), sensitivity, specificity, and accuracy. The correlation analysis between variables was evaluated using the correlation ratio and Pearson correlation coefficient. The model using a combination of dual-view radiomics and clinical parameters achieved a favorable performance (AUC: 0.804, 95% CI: 0.668-0.916), outperformed single-view model and model without clinical parameters. Incorporating with radiomics features of dual-view (CC&MLO) mammogram, age, breast density, and type of suspicious lesions can provide a noninvasive approach to evaluate the malignancy of breast lesions and facilitate clinical decision-making.

Ultrasound-based radiomics model for predicting molecular biomarkers in breast cancer

Background

Breast cancer (BC) is the most common cancer in women and is highly heterogeneous. BC can be classified into four molecular subtypes based on the status of estrogen receptor (ER), progesterone receptor (PR), human epidermal growth factor receptor 2 (HER2) and proliferation marker protein Ki-67. However, they can only be obtained by biopsy or surgery, which is invasive. Radiomics can noninvasively predict molecular expression via extracting the image features. Nevertheless, there is a scarcity of data available regarding the prediction of molecular biomarker expression using ultrasound (US) images in BC.

Objectives

To investigate the prediction performance of US radiomics for the assessment of molecular profiling in BC.

Methods

A total of 342 patients with BC who underwent preoperative US examination between January 2013 and December 2021 were retrospectively included. They were confirmed by pathology and molecular subtype analysis of ER, PR, HER2 and Ki-67. The radiomics features were extracted and four molecular models were constructed through support vector machine (SVM). Pearson correlation coefficient heatmaps are employed to analyze the relationship between selected features and their predictive power on molecular expression. The receiver operating characteristic curve was used for the prediction performance of US radiomics in the assessment of molecular profiling.

Results

359 lesions with 129 ER- and 230 ER+, 163 PR- and 196 PR+, 265 HER2- and 94 HER2+, 114 Ki-67- and 245 Ki-67+ expression were included. 1314 features were extracted from each ultrasound image. And there was a significant difference of some specific radiomics features between the molecule positive and negative groups. Multiple features demonstrated significant association with molecular biomarkers. The area under curves (AUCs) were 0.917, 0.835, 0.771, and 0.896 in the training set, while 0.868, 0.811, 0.722, and 0.706 in the validation set to predict ER, PR, HER2, and Ki-67 expression respectively.

Conclusion

Ultrasound-based radiomics provides a promising method for predicting molecular biomarker expression of ER, PR, HER2, and Ki-67 in BC.

A Statistical Approach to Assess the Robustness of Radiomics Features in the Discrimination of Mammographic Lesions

Despite mammography (MG) being among the most widespread techniques in breast cancer screening, tumour detection and classification remain challenging tasks due to the high morphological variability of the lesions. The extraction of radiomics features has proved to be a promising approach in MG. However, radiomics features can suffer from dependency on factors such as acquisition protocol, segmentation accuracy, feature extraction and engineering methods, which prevent the implementation of robust and clinically reliable radiomics workflow in MG. In this study, the variability and robustness of radiomics features is investigated as a function of lesion segmentation in MG images from a public database. A statistical analysis is carried out to assess feature variability and a radiomics robustness score is introduced based on the significance of the statistical tests performed. The obtained results indicate that variability is observable not only as a function of the abnormality type (calcification and masses), but also among feature categories (first-order and second-order), image view (craniocaudal and medial lateral oblique), and the type of lesions (benign and malignant). Furthermore, through the proposed approach, it is possible to identify those radiomics characteristics with a higher discriminative power between benign and malignant lesions and a lower dependency on segmentation, thus suggesting the most appropriate choice of robust features to be used as inputs to automated classification algorithms.

Noninvasive assessment of breast cancer molecular subtypes on multiparametric MRI using convolutional neural network with transfer learning

BACKGROUND

To evaluate the performances of multiparametric MRI-based convolutional neural networks (CNNs) for the preoperative assessment of breast cancer molecular subtypes.

METHODS

A total of 136 patients with 136 pathologically confirmed invasive breast cancers were randomly divided into training, validation, and testing sets in this retrospective study. The CNN models were established based on contrast-enhanced T₁ -weighted imaging (T₁ C), Apparent diffusion coefficient (ADC), and T₂ -weighted imaging (T₂ W) using the training and validation sets. The performances of CNN models were evaluated on the testing set. The area under the receiver operating characteristic curve (AUC), sensitivity, specificity, and accuracy were calculated to assess the performance.

RESULTS

For the separation of each subtype from other subtypes on the testing set, the T₁ C-based models yielded AUCs from 0.762 to 0.920; the ADC-based models yielded AUCs from 0.686 to 0.851; and the T₂ W-based models achieved AUCs from 0.639 to 0.697.

CONCLUSION

T₁ C-based models performed better than ADC-based models and T₂ W-based models in assessing the breast cancer molecular subtypes. The discriminating performances of our CNN models for triple negative and human epidermal growth factor receptor 2-enriched subtypes were better than that of luminal A and luminal B subtypes.

A deep learning approach with subregion partition in MRI image analysis for metastatic brain tumor

Purpose

To propose a deep learning network with subregion partition for predicting metastatic origins and EGFR/HER2 status in patients with brain metastasis.

Methods

We retrospectively enrolled 140 patients with clinico-pathologically confirmed brain metastasis originated from primary NSCLC (n = 60), breast cancer (BC, n = 60) and other tumor types (n = 20). All patients underwent contrast-enhanced brain MRI scans. The brain metastasis was subdivided into phenotypically consistent subregions using patient-level and population-level clustering. A residual network with a global average pooling layer (RN-GAP) was proposed to calculate deep learning-based features. Features from each subregion were selected with least absolute shrinkage and selection operator (LASSO) to build logistic regression models (LRs) for predicting primary tumor types (LR-NSCLC for the NSCLC origin and LR-BC for the BC origin), EGFR mutation status (LR-EGFR) and HER2 status (LR-HER2).

Results

The brain metastasis can be partitioned into a marginal subregion (S1) and an inner subregion (S2) in the MRI image. The developed models showed good predictive performance in the training (AUCs, LR-NSCLC vs. LR-BC vs. LR-EGFR vs. LR-HER2, 0.860 vs. 0.909 vs. 0.850 vs. 0.900) and validation (AUCs, LR-NSCLC vs. LR-BC vs. LR-EGFR vs. LR-HER2, 0.819 vs. 0.872 vs. 0.750 vs. 0.830) set.

Conclusion

Our proposed deep learning network with subregion partitions can accurately predict metastatic origins and EGFR/HER2 status of brain metastasis, and hence may have the potential to be non-invasive and preoperative new markers for guiding personalized treatment plans in patients with brain metastasis.

Potential of the Non-Contrast-Enhanced Chest CT Radiomics to Distinguish Molecular Subtypes of Breast Cancer: A Retrospective Study

Objectives

The molecular subtype plays an important role in breast cancer, which is the main reference to guide treatment and is closely related to prognosis. The objective of this study was to explore the potential of the non-contrast-enhanced chest CT-based radiomics to predict breast cancer molecular subtypes non-invasively.

Methods

A total of 300 breast cancer patients (153 luminal types and 147 non-luminal types) who underwent routine chest CT examination were included in the study, of which 220 cases belonged to the training set and 80 cases to the time-independent test set. Identification of the molecular subtypes is based on immunohistochemical staining of postoperative tissue samples. The region of interest (ROI) of breast masses was delineated on the continuous slices of CT images. Forty-two models to predict the luminal type of breast cancer were established by the combination of six feature screening methods and seven machine learning classifiers; 5-fold cross-validation (cv) was used for internal validation. Finally, the optimal model was selected for external validation on the independent test set. In addition, we also took advantage of SHapley Additive exPlanations (SHAP) values to make explanations of the machine learning model.

Results

During internal validation, the area under the curve (AUC) values for different models ranged from 0.599 to 0.842, and the accuracy ranged from 0.540 to 0.775. Eventually, the LASSO_SVM combination was selected as the final model, which included 9 radiomics features. The AUC, accuracy, sensitivity, and specificity of the model to distinguish luminal from the non-luminal type were 0.842 [95% CI: 0.728−0.957], 0.773, 0.818, and 0.773 in the training set and 0.757 [95% CI: 0.640–0.866], 0.713, 0.767, and 0.676 in the test set.

Conclusion

The radiomics based on chest CT may provide a new idea for the identification of breast cancer molecular subtypes.