Multitask Deep Learning-Based Whole-Process System for Automatic Diagnosis of Breast Lesions and Axillary Lymph Node Metastasis Discrimination from Dynamic Contrast-Enhanced-MRI: A Multicenter Study.

Accurate diagnosis of breast lesions and discrimination of axillary lymph node (ALN) metastases largely depend on radiologist experience. To develop a deep learning-based whole-process system (DLWPS) for segmentation and diagnosis of breast lesions and discrimination of ALN metastasis. Retrospective. 1760 breast patients, who were divided into training and validation sets (1110 patients), internal (476 patients), and external (174 patients) test sets. 3.0T/dynamic contrast-enhanced (DCE)-MRI sequence. DLWPS was developed using segmentation and classification models. The DLWPS-based segmentation model was developed by the U-Net framework, which combined the attention module and the edge feature extraction module. The average score of the output scores of three networks was used as the result of the DLWPS-based classification model. Moreover, the radiologists' diagnosis without and with the DLWPS-assistance was explored. To reveal the underlying biological basis of DLWPS, genetic analysis was performed based on RNA-sequencing data. Dice similarity coefficient (DI), area under receiver operating characteristic curve (AUC), accuracy, sensitivity, specificity, and kappa value. The segmentation model reached a DI of 0.828 and 0.813 in the internal and external test sets, respectively. Within the breast lesions diagnosis, the DLWPS achieved AUCs of 0.973 in internal test set and 0.936 in external test set. For ALN metastasis discrimination, the DLWPS achieved AUCs of 0.927 in internal test set and 0.917 in external test set. The agreement of radiologists improved with the DLWPS-assistance from 0.547 to 0.794, and from 0.848 to 0.892 in breast lesions diagnosis and ALN metastasis discrimination, respectively. Additionally, 10 breast cancers with ALN metastasis were associated with pathways of aerobic electron transport chain and cytoplasmic translation. The performance of DLWPS indicates that it can promote radiologists in the judgment of breast lesions and ALN metastasis and nonmetastasis. 4 TECHNICAL EFFICACY STAGE: 3.

Zhou H ，Hua Z ，Gao J ，Lin F ，Chen Y ，Zhang S ，Zheng T ，Wang Z ，Shao H ，Li W ，Liu F ，Li Q ，Chen J ，Wang X ，Zhao F ，Qu N ，Xie H ，Ma H ，Zhang H ，Mao N ... -  《-》

Attention-based Deep Learning for the Preoperative Differentiation of Axillary Lymph Node Metastasis in Breast Cancer on DCE-MRI.

Previous studies have explored the potential on radiomics features of primary breast cancer tumor to identify axillary lymph node (ALN) metastasis. However, the value of deep learning (DL) to identify ALN metastasis remains unclear. To investigate the potential of the proposed attention-based DL model for the preoperative differentiation of ALN metastasis in breast cancer on dynamic contrast-enhanced MRI (DCE-MRI). Retrospective. A total of 941 breast cancer patients who underwent DCE-MRI before surgery were included in the training (742 patients), internal test (83 patients), and external test (116 patients) cohorts. A 3.0 T MR scanner, DCE-MRI sequence. A DL model containing a 3D deep residual network (ResNet) architecture and a convolutional block attention module, named RCNet, was proposed for ALN metastasis identification. Three RCNet models were established based on the tumor, ALN, and combined tumor-ALN regions on the images. The performance of these models was compared with ResNet models, radiomics models, the Memorial Sloan-Kettering Cancer Center (MSKCC) model, and three radiologists (W.L., H.S., and F. L.). Dice similarity coefficient for breast tumor and ALN segmentation. Accuracy, sensitivity, specificity, intercorrelation and intracorrelation coefficients, area under the curve (AUC), and Delong test for ALN classification. The optimal RCNet model, that is, RCNet-tumor+ALN , achieved an AUC of 0.907, an accuracy of 0.831, a sensitivity of 0.824, and a specificity of 0.837 in the internal test cohort, as well as an AUC of 0.852, an accuracy of 0.828, a sensitivity of 0.792, and a specificity of 0.853 in the external test cohort. Additionally, with the assistance of RCNet-tumor+ALN , the radiologists' performance was improved (external test cohort, P < 0.05). DCE-MRI-based RCNet model could provide a noninvasive auxiliary tool to identify ALN metastasis preoperatively in breast cancer, which may assist radiologists in conducting more accurate evaluation of ALN status. 3 TECHNICAL EFFICACY: Stage 2.

Gao J ，Zhong X ，Li W ，Li Q ，Shao H ，Wang Z ，Dai Y ，Ma H ，Shi Y ，Zhang H ，Duan S ，Zhang K ，Yang P ，Zhao F ，Zhang H ，Xie H ，Mao N ... -  《-》

Multi-modality radiomics model predicts axillary lymph node metastasis of breast cancer using MRI and mammography.

We aimed to develop a multi-modality model to predict axillary lymph node (ALN) metastasis by combining clinical predictors with radiomic features from magnetic resonance imaging (MRI) and mammography (MMG) in breast cancer. This model might potentially eliminate unnecessary axillary surgery in cases without ALN metastasis, thereby minimizing surgery-related complications. We retrospectively enrolled 485 breast cancer patients from two hospitals and extracted radiomics features from tumor and lymph node regions on MRI and MMG images. After feature selection, three random forest models were built using the retained features, respectively. Significant clinical factors were integrated with these radiomics models to construct a multi-modality model. The multi-modality model was compared to radiologists' diagnoses on axillary ultrasound and MRI. It was also used to assist radiologists in making a secondary diagnosis on MRI. The multi-modality model showed superior performance with AUCs of 0.964 in the training cohort, 0.916 in the internal validation cohort, and 0.892 in the external validation cohort. It surpassed single-modality models and radiologists' ALN diagnosis on MRI and axillary ultrasound in all validation cohorts. Additionally, the multi-modality model improved radiologists' MRI-based ALN diagnostic ability, increasing the average accuracy from 70.70 to 78.16% for radiologist A and from 75.42 to 81.38% for radiologist B. The multi-modality model can predict ALN metastasis of breast cancer accurately. Moreover, the artificial intelligence (AI) model also assisted the radiologists to improve their diagnostic ability on MRI. The multi-modality model based on both MRI and mammography images allows preoperative prediction of axillary lymph node metastasis in breast cancer patients. With the assistance of the model, the diagnostic efficacy of radiologists can be further improved. • We developed a novel multi-modality model that combines MRI and mammography radiomics with clinical factors to accurately predict axillary lymph node (ALN) metastasis, which has not been previously reported. • Our multi-modality model outperformed both the radiologists' ALN diagnosis based on MRI and axillary ultrasound, as well as single-modality radiomics models based on MRI or mammography. • The multi-modality model can serve as a potential decision support tool to improve the radiologists' ALN diagnosis on MRI.

Wang Q ，Lin Y ，Ding C ，Guan W ，Zhang X ，Jia J ，Zhou W ，Liu Z ，Bai G ... -  《-》

Multiregional dynamic contrast-enhanced MRI-based integrated system for predicting pathological complete response of axillary lymph node to neoadjuvant chemotherapy in breast cancer: multicentre study.

Li Z ，Gao J ，Zhou H ，Li X ，Zheng T ，Lin F ，Wang X ，Chu T ，Wang Q ，Wang S ，Cao K ，Liang Y ，Zhao F ，Xie H ，Xu C ，Zhang H ，Niu Q ，Ma H ，Mao N ... -  《EBioMedicine》

Differentiating axillary lymph node metastasis in invasive breast cancer patients: A comparison of radiomic signatures from multiparametric breast MR sequences.

The axillary lymph node status is critical for breast cancer staging and individualized treatment planning. To assess the effect of determining axillary lymph node (ALN) metastasis by breast MRI-derived radiomic signatures, and compare the discriminating abilities of different MR sequences. Retrospective. In all, 120 breast cancer patients, 59 with ALN metastasis and 61 without metastasis, all confirmed by pathology. 3 .0T scanner with T1 -weighted imaging, T2 -weighted imaging, diffusion-weighted imaging, and dynamic contrast-enhanced (DCE) sequences. Typical morphological and texture features of the segmented tumor were extracted from four sequences, ie, T1 WI, T2 WI, DWI, and the second postcontrast phase (CE2) of the dynamic contrast-enhanced sequences. Additional contrast enhancement kinetic features were extracted from all DCE sequences (one pre- and seven postcontrast phases). Linear discriminant analysis classifiers were built and compared when using features from an individual sequence or the combination of the sequences in differentiating the ALN metastasis status. Mann-Whitney U-test, Fisher's exact test, least absolute shrinkage selection operator (LASSO) regression, and receiver operating characteristic analysis were performed. The accuracy/AUC of the four sequences was 79%/0.87, 77%/0.85, 74%/0.79, and 79%/0.85 for the T1 WI, CE2, T2 WI, and DWI, respectively. When CE2 was augmented by adding kinetic features, the model achieved the highest performance (accuracy = 0.86 and AUC = 0.91). When all features from the four sequences and the kinetics were combined, it did not lead to a further increase in the performance (P = 0.48). Breast tumor's radiomic signatures from preoperative breast MRI sequences are associated with the ALN metastasis status, where CE2 phase and the contrast enhancement kinetic features lead to the highest classification effect. Level of Evidence 3 Technical Efficacy Stage 2 J. Magn. Reson. Imaging 2019;50:1125-1132.

Chai R ，Ma H ，Xu M ，Arefan D ，Cui X ，Liu Y ，Zhang L ，Wu S ，Xu K ... -  《-》