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.

BACKGROUND: Accurate diagnosis of breast lesions and discrimination of axillary lymph node (ALN) metastases largely depend on radiologist experience. PURPOSE: To develop a deep learning-based whole-process system (DLWPS) for segmentation and diagnosis of breast lesions and discrimination of ALN metastasis. STUDY TYPE: Retrospective. POPULATION: 1760 breast patients, who were divided into training and validation sets (1110 patients), internal (476 patients), and external (174 patients) test sets. FIELD STRENGTH/SEQUENCE: 3.0T/dynamic contrast-enhanced (DCE)-MRI sequence. ASSESSMENT: 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. STATISTICAL TESTS: Dice similarity coefficient (DI), area under receiver operating characteristic curve (AUC), accuracy, sensitivity, specificity, and kappa value. RESULTS: 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. DATA CONCLUSION: The performance of DLWPS indicates that it can promote radiologists in the judgment of breast lesions and ALN metastasis and nonmetastasis. LEVEL OF EVIDENCE: 4 TECHNICAL EFFICACY STAGE: 3.

The value of Demetics ultrasound-assisted diagnosis system in diagnosis of breast lesions and in assessment Ki-67 status of breast cancer.

PURPOSE: This study aims to explore the diagnostic efficiency of the Demetics for breast lesions and assessment of Ki-67 status. MATERIAL: This retrospective study included 291 patients. Three combined methods (method 1: upgraded BI-RADS when Demetics classified the breast lesion as malignant; method 2: downgraded BI-RADS when Demetics classified the breast lesion as benign; method 3: BI-RADS was upgraded or downgraded according to Demetrics' diagnosis) were used to compare the diagnostic efficiency of two radiologists with different seniority before and after using Demetics. The correlation between the visual heatmap by Demetics and the Ki-67 expression level of breast cancer was explored. RESULTS: The sensitivity, specificity, and area under curve (AUC) of diagnosis by Demetics, junior radiologist and senior radiologist were 89.5%, 83.1%, 0.863; 76.9%, 82.4%, 0.797 and 81.1%, 89.9%, 0.855, respectively. Method 1 was the best for senior radiologist, which increased AUC from 0.855 to 0.884. For junior radiologist, Method 3 was the best method, improving sensitivity (88.8% vs. 76.9%) and specificity (87.2% vs. 82.4%). Demetics paid more attention to the peripheral area of breast cancer with high expression of Ki-67. CONCLUSION: Demetics has shown good diagnostic efficiency in the assisted diagnosis of breast lesions and is expected to further distinguish Ki-67 status of breast cancer.

Head-to-head comparison of perfluorobutane contrast-enhanced US and multiparametric MRI for breast cancer: a prospective, multicenter study.

BACKGROUND: Multiparametric magnetic resonance imaging (MP-MRI) has high sensitivity for diagnosing breast cancers but cannot always be used as a routine diagnostic tool. The present study aimed to evaluate whether the diagnostic performance of perfluorobutane (PFB) contrast-enhanced ultrasound (CEUS) is similar to that of MP-MRI in breast cancer and whether combining the two methods would enhance diagnostic efficiency. PATIENTS AND METHODS: This was a head-to-head, prospective, multicenter study. Patients with breast lesions diagnosed by US as Breast Imaging Reporting and Data System (BI-RADS) categories 3, 4, and 5 underwent both PFB-CEUS and MP-MRI scans. On-site operators and three reviewers categorized the BI-RADS of all lesions on two images. Logistic-bootstrap 1000-sample analysis and cross-validation were used to construct PFB-CEUS, MP-MRI, and hybrid (PFB-CEUS + MP-MRI) models to distinguish breast lesions. RESULTS: In total, 179 women with 186 breast lesions were evaluated from 17 centers in China. The area under the receiver operating characteristic curve (AUC) for the PFB-CEUS model to diagnose breast cancer (0.89; 95% confidence interval [CI] 0.74, 0.97) was similar to that of the MP-MRI model (0.89; 95% CI 0.73, 0.97) (P = 0.85). The AUC of the hybrid model (0.92, 95% CI 0.77, 0.98) did not show a statistical advantage over the PFB-CEUS and MP-MRI models (P = 0.29 and 0.40, respectively). However, 90.3% false-positive and 66.7% false-negative results of PFB-CEUS radiologists and 90.5% false-positive and 42.8% false-negative results of MP-MRI radiologists could be corrected by the hybrid model. Three dynamic nomograms of PFB-CEUS, MP-MRI and hybrid models to diagnose breast cancer are freely available online. CONCLUSIONS: PFB-CEUS can be used in the differential diagnosis of breast cancer with comparable performance to MP-MRI and with less time consumption. Using PFB-CEUS and MP-MRI as joint diagnostics could further strengthen the diagnostic ability. Trial registration Clinicaltrials.gov; NCT04657328. Registered 26 September 2020. IRB number 2020-300 was approved in Chinese PLA General Hospital. Every patient signed a written informed consent form in each center.

Decorin inhibits the formation of hard nodules after microwave ablation by inhibiting the TGF-ß1/SMAD and MAPK signaling pathways: in a Bama miniature pig model of mammary gland hyperplasia.

BACKGROUND: Benign breast lesions are often associated with hard nodule formation after microwave ablation (MWA), which persists for a long time and causes problems in patients. The aim of this study was to evaluate the efficacy of decorin in the treatment of hard nodule formation and its potential mechanism of action. METHODS: Using a Bama miniature pig model of mammary gland hyperplasia, immunohistochemistry, Masson's trichrome and western blotting were firstly applied to compare the extent of fibrosis and activation of key members of the TGF-ß1/SMAD and MAPK signaling pathways of hard nodule in the control and MWA groups, and then the extent of fibrosis and expression of signaling pathways in hard nodule were examined after application of decorin. RESULTS: The results showed that the MWA group had increased levels of TGF-ß1, p-SMAD2/3, p-ERK1/2, and collagen I proteins and increased fibrosis at 2 weeks, 4 weeks, and 3 months after MWA. After decorin treatment, the expression levels of each protein were significantly downregulated, and the degree of fibrosis was reduced at 2 weeks, 4 weeks, and 3 months after MWA compared with the MWA group. CONCLUSION: In conclusion, these results suggest that activation of TGF-ß1 may play an important role in hard nodule formation and that decorin may reduce hard nodule formation after MWA in a model of mammary gland hyperplasia by inhibiting the TGF-ß1/SMAD and MAPK signaling pathways.

Application of advanced diffusion models from diffusion weighted imaging in a large cohort study of breast lesions.

BACKGROUND: To evaluate multiple parameters in multiple b-value diffusion-weighted imaging (DWI) in characterizing breast lesions and predicting prognostic factors and molecular subtypes. METHODS: In total, 504 patients who underwent 3-T magnetic resonance imaging (MRI) with T1-weighted dynamic contrast-enhanced (DCE) sequences, T2-weighted sequences and multiple b-value (7 values, from 0 to 3000 s/mm2) DWI were recruited. The average values of 13 parameters in 6 models were calculated and recorded. The pathological diagnosis of breast lesions was based on the latest World Health Organization (WHO) classification. RESULTS: Twelve parameters exhibited statistical significance in differentiating benign and malignant lesions. alpha demonstrated the highest sensitivity (89.5%), while sigma demonstrated the highest specificity (77.7%). The stretched-exponential model (SEM) demonstrated the highest sensitivity (90.8%), while the biexponential model demonstrated the highest specificity (80.8%). The highest AUC (0.882, 95% CI, 0.852-0.912) was achieved when all 13 parameters were combined. Prognostic factors were correlated with different parameters, but the correlation was relatively weak. Among the 6 parameters with significant differences among molecular subtypes of breast cancer, the Luminal A group and Luminal B (HER2 negative) group had relatively low values, and the HER2-enriched group and TNBC group had relatively high values. CONCLUSIONS: All 13 parameters, independent or combined, provide valuable information in distinguishing malignant from benign breast lesions. These new parameters have limited meaning for predicting prognostic factors and molecular subtypes of malignant breast tumors.

Reducing the number of unnecessary biopsies for mammographic BI-RADS 4 lesions through a deep transfer learning method.

BACKGROUND: In clinical practice, reducing unnecessary biopsies for mammographic BI-RADS 4 lesions is crucial. The objective of this study was to explore the potential value of deep transfer learning (DTL) based on the different fine-tuning strategies for Inception V3 to reduce the number of unnecessary biopsies that residents need to perform for mammographic BI-RADS 4 lesions. METHODS: A total of 1980 patients with breast lesions were included, including 1473 benign lesions (185 women with bilateral breast lesions), and 692 malignant lesions collected and confirmed by clinical pathology or biopsy. The breast mammography images were randomly divided into three subsets, a training set, testing set, and validation set 1, at a ratio of 8:1:1. We constructed a DTL model for the classification of breast lesions based on Inception V3 and attempted to improve its performance with 11 fine-tuning strategies. The mammography images from 362 patients with pathologically confirmed BI-RADS 4 breast lesions were employed as validation set 2. Two images from each lesion were tested, and trials were categorized as correct if the judgement (≥ 1 image) was correct. We used precision (Pr), recall rate (Rc), F1 score (F1), and the area under the receiver operating characteristic curve (AUROC) as the performance metrics of the DTL model with validation set 2. RESULTS: The S5 model achieved the best fit for the data. The Pr, Rc, F1 and AUROC of S5 were 0.90, 0.90, 0.90, and 0.86, respectively, for Category 4. The proportions of lesions downgraded by S5 were 90.73%, 84.76%, and 80.19% for categories 4 A, 4B, and 4 C, respectively. The overall proportion of BI-RADS 4 lesions downgraded by S5 was 85.91%. There was no significant difference between the classification results of the S5 model and pathological diagnosis (P = 0.110). CONCLUSION: The S5 model we proposed here can be used as an effective approach for reducing the number of unnecessary biopsies that residents need to conduct for mammographic BI-RADS 4 lesions and may have other important clinical uses.

Superb Microvascular Imaging for the Differentiation of Benign and Malignant Breast Lesions: A System Review and Meta-Analysis.

OBJECTIVE: To evaluate the diagnostic performance of SMI in the diagnosis of benign and malignant breast lesions. METHODS: A systematic search of PubMed, EMBASE, Cochrane, OVID, SCI, and SCOPUS was performed to find relevant studies which applied SMI to differentiate benign and malignant breast lesions. All the studies were published before October 10, 2022. Only studies published in English were collected. Quality Assessment of Diagnostic Accuracy Studies-2 (QUADAS-2) tool was applied to assess the quality of the included studies. Summary receiver operating characteristic (SROC) modeling was also performed to the diagnostic performance of SMI in the diagnosis of benign and malignant breast lesions. Subgroup analyses and meta-regression were performed to find out the heterogeneity. RESULTS: Twenty studies which include a total of 2873 lesions (1748 benign and 1125 malignant) in 2740 patients were evaluated in this meta-analysis. The summary sensitivity and specificity were 0.82 (95% confidence interval [CI]: 0.76-0.86), 0.70 (95% CI: 0.64-0.76) for SMI vascular degree, 0.77 (95% CI: 0.67-0.84), 0.79 (95% CI: 0.75-0.83) for SMI vascular distribution, 0.78 (95% CI: 0.70-0.84), 0.75 (95% CI: 0.69-0.80) for SMI vascular morphology, 0.81 (95% CI: 0.72-0.87), 0.80 (95% CI: 0.75-0.85) SMI penetration vessel. For SMI overall vascular features, the summary sensitivity and summary specificity were 0.74 (95% CI: 0.61-0.84) and 0.80 (95% CI: 0.76-0.84). The result of subgroup analysis and meta-analysis showed malignant rate and country might be the cause of heterogeneity of diagnostic accuracy of vascular grade and morphology. CONCLUSION: SMI vascular features have high sensitivity and specificity in the differentiation of benign and malignant lesions. Future international multicenter studies in various regions with large sample size are required to confirm these findings.

Artificial intelligence for non-mass breast lesions detection and classification on ultrasound images: a comparative study.

BACKGROUND: This retrospective study aims to validate the effectiveness of artificial intelligence (AI) to detect and classify non-mass breast lesions (NMLs) on ultrasound (US) images. METHODS: A total of 228 patients with NMLs and 596 volunteers without breast lesions on US images were enrolled in the study from January 2020 to December 2022. The pathological results served as the gold standard for NMLs. Two AI models were developed to accurately detect and classify NMLs on US images, including DenseNet121_448 and MobileNet_448. To evaluate and compare the diagnostic performance of AI models, the area under the curve (AUC), accuracy, specificity and sensitivity was employed. RESULTS: A total of 228 NMLs patients confirmed by postoperative pathology with 870 US images and 596 volunteers with 1003 US images were enrolled. In the detection experiment, the MobileNet_448 achieved the good performance in the testing set, with the AUC, accuracy, sensitivity, and specificity were 0.999 (95%CI: 0.997-1.000),96.5%,96.9% and 96.1%, respectively. It was no statistically significant compared to DenseNet121_448. In the classification experiment, the MobileNet_448 model achieved the highest diagnostic performance in the testing set, with the AUC, accuracy, sensitivity, and specificity were 0.837 (95%CI: 0.990-1.000), 70.5%, 80.3% and 74.6%, respectively. CONCLUSIONS: This study suggests that the AI models, particularly MobileNet_448, can effectively detect and classify NMLs in US images. This technique has the potential to improve early diagnostic accuracy for NMLs.

A Radiomics Study: Classification of Breast Lesions by Textural Features from Mammography Images.

This study investigates the feasibility of using texture radiomics features extracted from mammography images to distinguish between benign and malignant breast lesions and to classify benign lesions into different categories and determine the best machine learning (ML) model to perform the tasks. Six hundred and twenty-two breast lesions from 200 retrospective patient data were segmented and analysed. Three hundred fifty radiomics features were extracted using the Standardized Environment for Radiomics Analysis (SERA) library, one of the radiomics implementations endorsed by the Image Biomarker Standardisation Initiative (IBSI). The radiomics features and selected patient characteristics were used to train selected machine learning models to classify the breast lesions. A fivefold cross-validation was used to evaluate the performance of the ML models and the top 10 most important features were identified. The random forest (RF) ensemble gave the highest accuracy (89.3%) and positive predictive value (66%) and likelihood ratio of 13.5 in categorising benign and malignant lesions. For the classification of benign lesions, the RF model again gave the highest likelihood ratio of 3.4 compared to the other models. Morphological and textural radiomics features were identified as the top 10 most important features from the random forest models. Patient age was also identified as one of the significant features in the RF model. We concluded that machine learning models trained against texture-based radiomics features and patient features give reasonable performance in differentiating benign versus malignant breast lesions. Our study also demonstrated that the radiomics-based machine learning models were able to emulate the visual assessment of mammography lesions, typically used by radiologists, leading to a better understanding of how the machine learning model arrive at their decision.

Stereotactic vs open biopsy for non-palpable breast lesions. Their impact on short term quality of life.

Stereotactic biopsy and open biopsy represent useful diagnostic tools of breast lesions. However, both affect the Quality of Life (QoL) in various degrees. We conducted this prospective longitudinal comparative study in order, first to access the impact of these techniques on short-term QoL and second to compare and critically discuss our results with those of literature review. Group A (58 patients) underwent vacuum-assisted stereotactic biopsy and Group B (46 patients) underwent open biopsy. The Health-Related Quality of Life (HRQol) was estimated using the European Quality of Life scale (EuroQol) and the SF-36 (The 36-Item Short-Form Health Status Survey) questionnaires. The stereotactic breast biopsy seems to be more accepted from the patients as it affects quality of life to a lesser extent than open breast biopsy. This difference is mainly attributable to a reduction of physical discomfort and pain.

Detection and classification of breast lesions using multiple information on contrast-enhanced mammography by a multiprocess deep-learning system: A multicenter study.

Objective: Accurate detection and classification of breast lesions in early stage is crucial to timely formulate effective treatments for patients. We aim to develop a fully automatic system to detect and classify breast lesions using multiple contrast-enhanced mammography (CEM) images. Methods: In this study, a total of 1,903 females who underwent CEM examination from three hospitals were enrolled as the training set, internal testing set, pooled external testing set and prospective testing set. Here we developed a CEM-based multiprocess detection and classification system (MDCS) to perform the task of detection and classification of breast lesions. In this system, we introduced an innovative auxiliary feature fusion (AFF) algorithm that could intelligently incorporates multiple types of information from CEM images. The average free-response receiver operating characteristic score (AFROC-Score) was presented to validate system's detection performance, and the performance of classification was evaluated by area under the receiver operating characteristic curve (AUC). Furthermore, we assessed the diagnostic value of MDCS through visual analysis of disputed cases, comparing its performance and efficiency with that of radiologists and exploring whether it could augment radiologists' performance. Results: On the pooled external and prospective testing sets, MDCS always maintained a high standalone performance, with AFROC-Scores of 0.953 and 0.963 for detection task, and AUCs for classification were 0.909 [95% confidence interval (95% CI): 0.822-0.996] and 0.912 (95% CI: 0.840-0.985), respectively. It also achieved higher sensitivity than all senior radiologists and higher specificity than all junior radiologists on pooled external and prospective testing sets. Moreover, MDCS performed superior diagnostic efficiency with an average reading time of 5 seconds, compared to the radiologists' average reading time of 3.2 min. The average performance of all radiologists was also improved to varying degrees with MDCS assistance. Conclusions: MDCS demonstrated excellent performance in the detection and classification of breast lesions, and greatly enhanced the overall performance of radiologists.

Nuclear morphology in breast lesions: refining its assessment to improve diagnostic concordance.

AIMS: Although evaluation of nuclear morphology is important for the diagnosis and categorisation of breast lesions, the criteria used to assess nuclear atypia rely upon the subjective evaluation of several features that may result in inter- and intraobserver variation. This study aims to refine the definitions of cytonuclear features in various breast lesions. METHODS AND RESULTS: ImageJ was used to assess the nuclear morphological features including nuclear diameter, axis length, perimeter, area, circularity and roundness in 160 breast lesions comprising ductal carcinoma in situ (DCIS), invasive breast carcinoma of no special type (IBC-NST), tubular carcinoma, usual ductal hyperplasia (UDH), columnar cell change (CCC) and flat epithelial atypia (FEA). Reference cells included normal epithelial cells, red blood cells (RBCs) and lymphocytes. Reference cells showed size differences not only between normal epithelial cells and RBCs but also between RBCs in varied-sized blood vessels. Nottingham grade nuclear pleomorphism scores 1 and 3 cut-offs in IBC-NST, compared to normal epithelial cells, were < ×1.2 and > ×1.4 that of mean maximum Feret's diameter and < ×1.6 and > ×2.4 that of mean nuclear area, respectively. Nuclear morphometrics were significantly different in low-grade IBC-NST versus tubular carcinoma, low-grade DCIS versus UDH and CCC versus FEA. No differences in the nuclear features between grade-matched DCIS and IBC-NST were identified. CONCLUSION: This study provides a guide for the assessment of nuclear atypia in breast lesions, refines the comparison with reference cells and highlights the potential diagnostic value of image analysis tools in the era of digital pathology.

Collagen fiber features and COL1A1: are they associated with elastic parameters in breast lesions, and can COL1A1 predict axillary lymph node metastasis?

BACKGROUND: This study aimed to explore whether collagen fiber features and collagen type I alpha 1 (COL1A1) are related to the stiffness of breast lesions and whether COL1A1 can predict axillary lymph node metastasis (LNM). METHODS: Ninety-four patients with breast lesions were consecutively enrolled in the study. Amongst the 94 lesions, 30 were benign, and 64 were malignant (25 were accompanied by axillary lymph node metastasis). Ultrasound (US) and shear wave elastography (SWE) were performed for each breast lesion before surgery. Sirius red and immunohistochemical staining were used to examine the shape and arrangement of collagen fibers and COL1A1 expression in the included tissue samples. We analyzed the correlation between the staining results and SWE parameters and investigated the effectiveness of COL1A1 expression levels in predicting axillary LNM. RESULTS: The optimal cut-off values for Emax, Emean, and Eratio for diagnosing the benign and malignant groups, were 58.70 kPa, 52.50 kPa, and 3.05, respectively. The optimal cutoff for predicting axillary LNM were 107.5 kPa, 85.15 kPa, and 3.90, respectively. Herein, the collagen fiber shape and arrangement features in breast lesions were classified into three categories. One-way analysis of variance (ANOVA) showed that Emax, Emean, and Eratio differed between categories 0, 1, and 2 (P < 0.05). Meanwhile, elasticity parameters were positively correlated with collagen categories and COL1A1 expression. The COL1A1 expression level > 0.145 was considered the cut-off value, and its efficacy in benign and malignant breast lesions was 0.808, with a sensitivity of 66% and a specificity of 90%. Furthermore, when the COL1A1 expression level > 0.150 was considered the cut-off, its efficacy in predicting axillary LNM was 0.796, with sensitivity and specificity of 96% and 59%, respectively. CONCLUSIONS: The collagen fiber features and expression levels of COL1A1 positively correlated with the elastic parameters of breast lesions. The expression of COL1A1 may help diagnose benign and malignant breast lesions and predict axillary LNM.

Differentiating Benign and Malignant Breast Lesions in Diffusion Kurtosis MRI: Does the Averaging Procedure Matter?

BACKGROUND: Diffusion kurtosis imaging (DKI) is used to differentiate between benign and malignant breast lesions. DKI fits are performed either on voxel-by-voxel basis or using volume-averaged signal. PURPOSE: Investigate and compare DKI parameters' diagnostic performance using voxel-by-voxel and volume-averaged signal fit approach. STUDY TYPE: Retrospective. STUDY POPULATION: A total of 104 patients, aged 24.1-86.4 years. FIELD STRENGTH/SEQUENCE: A 3 T Spin-echo planar diffusion-weighted sequence with b-values: 50 s/mm2 , 750 s/mm2 , and 1500 s/mm2 . Dynamic contrast enhanced (DCE) sequence. ASSESSMENT: Lesions were manually segmented by M.P. under supervision of S.O. (2 and 5 years of experience in breast MRI). DKI fits were performed on voxel-by-voxel basis and with volume-averaged signal. Diagnostic performance of DKI parameters D K (kurtosis corrected diffusion coefficient) and kurtosis K was compared between both approaches. STATISTICAL TESTS: Receiver operating characteristics analysis and area under the curve (AUC) values were computed. Wilcoxon rank sum and Students t-test tested DKI parameters for significant (P <0.05) difference between benign and malignant lesions. DeLong test was used to test the DKI parameter performance for significant fit approach dependency. Correlation between parameters of the two approaches was determined by Pearson correlation coefficient. RESULTS: DKI parameters were significantly different between benign and malignant lesions for both fit approaches. Median benign vs. malignant values for voxel-by-voxel and volume-averaged approach were 2.00 vs. 1.28 ( D K in µm2 /msec), 2.03 vs. 1.26 ( D K in µm2 /msec), 0.54 vs. 0.90 ( K ), 0.55 vs. 0.99 ( K ). AUC for voxel-by-voxel and volume-averaged fit were 0.9494 and 0.9508 ( D K ); 0.9175 and 0.9298 ( K ). For both, AUC did not differ significantly (P = 0.20). Correlation of values between the two approaches was very high (r = 0.99 for D K and r = 0.97 for K ). DATA CONCLUSION: Voxel-by-voxel and volume-averaged signal fit approach are equally well suited for differentiating between benign and malignant breast lesions in DKI. EVIDENCE LEVEL: 3 TECHNICAL EFFICACY: Stage 3.

Detection and classification of breast lesions with You Only Look Once version 5.

Aim: To explore the ability of You Only Look Once version 5 (YOLOv5) to detect and classify breast lesions on dynamic contrast-enhanced MRI. Methods: Four YOLOv5 submodels were examined. A total of 2124 and 2226 images of benign and malignant lesions were obtained, respectively. Precision, recall rate and mean average precision were used to evaluate model performance. Results: The precision (0.916) and mean average precision _0.5 (0.894) of YOLOv5s were higher than those of YOLOv5m (0.832, 0.794), YOLOv5l (0.843, 0.803) and YOLOv5x (0.854, 0.821). In the validation set, YOLOv5s required 1.1 ms to detect lesions per image. Conclusion: YOLOv5s was the fastest and had the highest precision among the four YOLOv5 submodels for the detection and classification of breast lesions on dynamic contrast-enhanced MRI. It has a greater clinical application value.

You Only Look Once version 5 (YOLOv5) is the latest YOLO series, which may be a useful tool for detecting and classifying breast lesions on dynamic contrast-enhanced MRI (DCE-MRI) and help clinicians make a rapid, accurate diagnosis and provide treatment. Data were retrospectively collected from a single-center study. The performances of the four submodels (YOLOv5s, YOLOv5m, YOLOv5l and YOLOv5x) were compared. The diagnostic performances of YOLOv5s were comparable with some convolutional neural network models for breast lesion identification in breast ultrasonography and mammography. This study may provide novel insights into the detection and classification of breast lesions on DCE-MRI. Thus, a sufficiently large series of data and high-quality DCE-MRIs are warranted. Owing to its applications in artificial intelligence-assisted imaging diagnosis, this method has promising prospects.

The role of the IAC Yokohama System for Reporting Breast Fine Needle Aspiration Biopsy and the ACR Breast Imaging-Reporting and Data System in the evaluation of breast lesions.

BACKGROUND: Stratification of breast lesions for appropriate management is achieved through an integration of clinical examination, imaging, and fine needle aspiration biopsy (FNAB). The current study aimed to evaluate the combined effectiveness of the widely used Breast Imaging-Reporting and Data System (BI-RADS) with the recently proposed International Academy of Cytology (IAC) Yokohama System for Reporting Breast Fine Needle Aspiration Biopsy Cytopathology. METHODS: A retrospective analysis was done on all breast FNABs from 2016 through 2020. The cases were categorised according to the IAC Yokohama System. Histopathological correlation of the BI-RADS and IAC Yokohama System was performed. The rate of malignancy (ROM) for each category of the BI-RADS and IAC Yokohama System was calculated. RESULTS: The ROM values for categories I to V were 38%, 0.6%, 21.9%, 100%, and 97%, respectively. The sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and accuracy of FNAB with category III assumed as malignant were 98.9%, 85%, 76.1%, 99.3%, and 89.5%, respectively. With category III assumed as benign, these indices were 90.8%, 98.9%, 97.5%, 95.7%, and 96.2%, respectively. The sensitivity, specificity, PPV, NPV and accuracy of BI-RADS were 91.5%, 81.9%, 72%, 95%, and 85.1%, respectively. CONCLUSIONS: FNAB is still an indispensable test in the evaluation of breast lesions. The utilisation of the IAC Yokohama reporting system for breast cytology in conjunction with ACR BI-RADS aids in better stratification of lesions.

Ultrasound-guided 8-Gauge vacuum-assisted excision for selected B3 breast lesions: a preliminary experience.

PURPOSE: Nowadays, surgical excision is no longer justified for all B3 lesions and a minimally-invasive therapeutic treatment has been encouraged. The aim of this study was to evaluate the feasibility and the therapeutic efficacy of ultrasound-guided vacuum-assisted excision (US-VAE) for the treatment of selected breast lesions of uncertain malignant potential (B3). MATERIAL AND METHODS: From July 2018 to December 2019, 11/48 breast lesions classified as B3 after ultrasound-guided core needle biopsy were treated with US-VAE in our Institution. Inclusion criteria were: B3 nodules ultrasonographically detectable for which VAE is recommended by international guidelines2, size ranging between 5 and 25 mm, circumscribed margins, and lesion position at least 5 mm from the skin and the nipple. A radiological follow-up to evaluate the completeness of excision, the presence of post-procedural hematoma or of residual disease/recurrence was performed after 10 and 30 days and 6 and 12 months. 12-month ultrasound was considered the gold standard. All patients were asked to complete a satisfaction survey and a full assessment of the costs of US-VAE was performed. RESULTS: Complete excision was achieved in 81.8% of US-VAE. No lesions were upgraded to carcinoma and no patients had to undergo surgery. No complications occurred during or after US-VAE. All patients were satisfied with the procedure and the cosmetic result (100%). US-VAE cost approximately 422 Euros per procedure. CONCLUSION: US-VAE has proven to be an optimal tool for the therapeutic excision of selected B3 lesions, with high success rate, good patient compliance and considerable money savings compared to surgery. This technique has the potential to reduce unnecessary surgery and healthcare costs.

TGF-ß1 and its signal molecules: are they correlated with the elasticity characteristics of breast lesions?

BACKGROUND: Shear wave elastography can evaluate tissue stiffness. Previous studies showed that the elasticity characteristics of breast lesions were related to the components of extracellular matrix which was regulated by transforming growth factor beta 1(TGF-ß1) directly or indirectly. However, the correlation of the expression level of TGF-ß1, its signal molecules and elasticity characteristics of breast lesions have rarely been reported. The purpose of this study was to investigate the correlation between the expression level of TGF-ß1, its signal molecules, and the elasticity characteristics of breast lesions. METHODS: 135 breast lesions in 130 patients were included. Elasticity parameters, including elasticity modulus, the elasticity ratio, the "stiff rim sign", were recorded before biopsy and surgical excision. The expression levels of TGF-ß1 and its signal molecules, including Smad2/3, Erk1/2, p38 mitogen-activated protein kinase (MAPK), c-Jun N-terminal kinase 2 (JNK2), phosphoinositide 3-kinase (PI3K), and protein kinase B (PKB/AKT) were detected by immunohistochemistry. The diagnostic performance of the expression level of those molecules and their correlation with the elasticity characteristics were analyzed. RESULTS: Elasticity parameters and the expression levels of TGF- ß1 and its signal molecules of benign lesions were lower than those of malignant lesions (P<0.0001). The expression levels of TGF- ß1 and its signal molecules were correlated with elasticity parameters. The expression levels of TGF- ß1 and its signal molecules in lesions with "stiff rim sign" were higher than those without "stiff rim sign" (P<0.05). And the expression levels of Smad2/3, Erk1/2, p38 MAPK, JNK2, PI3K and AKT were correlated with that of TGF- ß1. The area under the curve for receiver operator characteristic curve of TGF-ß1 and its signal molecules in the differentiation of malignant and benign breast lesions ranged from 0.920-0.960. CONCLUSIONS: The expression levels of TGF-ß1, its signal molecules of breast lesions showed good diagnostic performance and were correlated with the elasticity parameters. The expression levels of signal molecules were correlated with that of TGF- ß1, which speculated that TGF- ß1 might play an important role in the regulation of breast lesion elasticity parameters and multiple signal molecule expressions.

Investigation of Synthetic Relaxometry and Diffusion Measures in the Differentiation of Benign and Malignant Breast Lesions as Compared to BI-RADS.

BACKGROUND: Breast cancer is the most common malignant tumor in women and a quantitative contrast-free method is highly desirable for its diagnosis. PURPOSE: To investigate the performance of quantitative MRI in differentiating malignant from benign breast lesions and to compare with the Breast Imaging Reporting and Data System (BI-RADS). STUDY TYPE: Retrospective. SUBJECTS: Eighty patients (56 with malignant lesions and 24 with benign lesions). FIELD STRENGTH/SEQUENCE: Diffusion-weighted imaging (DWI) with a single-shot echo planar sequence and synthetic MRI with magnetic resonance image compilation (MAGiC) were performed at 3T. ASSESSMENT: T1 relaxation time (T1 ), T2 relaxation time (T2 ), and proton density (PD) from synthetic MRI and apparent diffusion coefficient (ADC) from DWI were analyzed by two radiologists (Reader A, Reader B). Univariable and multivariable models were developed to optimize differentiation between malignant and benign lesions and their performances compared to BI-RADS. STATISTICAL TESTS: The diagnostic performance was evaluated using multivariate logistic regression analysis and area under the receiver operating characteristic (ROC) curves (AUC). RESULTS: T2 , PD, and ADC values for malignant lesions were significantly lower than those in benign breast lesions for both radiologists (all P < 0.05). The combined T2 , PD, and ADC model had the best performance for differentiating malignant and benign lesions with AUC, sensitivity, specificity, positive predictive value, and negative predictive values of 0.904, 94.6%, 87.5%, 94.6%, and 87.5%, respectively. The corresponding results for BI-RADS were no AUC, 94.6%, 75.0%, 89.8%, and 85.7%, respectively. DATA CONCLUSION: The approach that combined synthetic MRI and DWI outperformed BI-RADS in the differential diagnosis of malignant and benign breast lesions and was achieved without contrast agents. This approach may serve as an alternative and effective strategy for the improvement of breast lesion differentiation. LEVEL OF EVIDENCE: 3. TECHNICAL EFFICACY STAGE: 3.

High-Risk Lesions Detected by MRI-Guided Core Biopsy: Upgrade Rates at Surgical Excision and Implications for Management.

OBJECTIVE. The purpose of our study was to evaluate the upgrade rates of high-risk lesions (HRLs) diagnosed by MRI-guided core biopsy and to assess which clinical and imaging characteristics are predictive of upgrade to malignancy. MATERIALS AND METHODS. A retrospective review was performed of all women who presented to an academic breast radiology center for MRI-guided biopsy between January 1, 2015, and November 30, 2018. Histopathologic results from each biopsy were extracted. HRLs-that is, atypical ductal hyperplasia (ADH), lobular carcinoma in situ (LCIS), atypical lobular hyperplasia (ALH), radial scar, papilloma, flat epithelial atypia (FEA), benign vascular lesion (BVL), and mucocelelike lesion-were included for analysis. Clinical history, imaging characteristics, surgical outcome, and follow-up data were recorded. Radiologic-pathologic correlation was performed. RESULTS. Of 810 MRI-guided biopsies, 189 cases (23.3%) met the inclusion criteria for HRLs. Of the 189 HRLs, 30 cases were excluded for the following reasons: 15 cases were lost to follow-up, six cases were in patients who received neoadjuvant chemotherapy after biopsy, two lesions that were not excised had less than 2 years of imaging follow-up, and seven lesions had radiologic-pathologic discordance at retrospective review. Of the 159 HRLs in our study cohort, 13 (8.2%) were upgraded to carcinoma. Surgical upgrade rates were high for ADH (22.5%, 9/40) and FEA (33.3%, 1/3); moderate for LCIS (6.3%, 3/48); and low for ALH (0.0%, 0/11), radial scar (0.0%, 0/28), papilloma (0.0%, 0/26), and BVL (0.0%, 0/3). Of the upgraded lesions, 69.2% (9/13) were upgraded to ductal carcinoma in situ (DCIS) or well-differentiated carcinoma. ADH lesions were significantly more likely to be upgraded than non-ADH lesions (p = .005). CONCLUSION. ADH diagnosed by MRI-guided core biopsy warrants surgical excision. The other HRLs, however, may be candidates for imaging follow-up rather than excision, especially after meticulous radiologic-pathologic correlation.