Artificial intelligence for ultrasound microflow imaging in breast cancer diagnosis.

PURPOSE: To develop and evaluate artificial intelligence (AI) algorithms for ultrasound (US) microflow imaging (MFI) in breast cancer diagnosis. MATERIALS AND METHODS: We retrospectively collected a dataset consisting of 516 breast lesions (364 benign and 152 malignant) in 471 women who underwent B-mode US and MFI. The internal dataset was split into training (n = 410) and test datasets (n = 106) for developing AI algorithms from deep convolutional neural networks from MFI. AI algorithms were trained to provide malignancy risk (0-100%). The developed AI algorithms were further validated with an independent external dataset of 264 lesions (229 benign and 35 malignant). The diagnostic performance of B-mode US, AI algorithms, or their combinations was evaluated by calculating the area under the receiver operating characteristic curve (AUROC). RESULTS: The AUROC of the developed three AI algorithms (0.955-0.966) was higher than that of B-mode US (0.842, P < 0.0001). The AUROC of the AI algorithms on the external validation dataset (0.892-0.920) was similar to that of the test dataset. Among the AI algorithms, no significant difference was found in all performance metrics combined with or without B-mode US. Combined B-mode US and AI algorithms had a higher AUROC (0.963-0.972) than that of B-mode US (P < 0.0001). Combining B-mode US and AI algorithms significantly decreased the false-positive rate of BI-RADS category 4A lesions from 87% to 13% (P < 0.0001). CONCLUSION: AI-based MFI diagnosed breast cancers with better performance than B-mode US, eliminating 74% of false-positive diagnoses in BI-RADS category 4A lesions.

Analysis of background echotexture on automated breast ultrasound using BI-RADS and modified classification: Association with clinical features and mammographic density.

PURPOSE: To analyze BE on ABUS using BI-RADS and a modified classification in association with mammographic density and clinical features. METHODS: Menopausal status, parity, and family history of breast cancer were collected for 496 women who underwent ABUS and mammography. Three radiologists independently reviewed all ABUS BE and mammographic density. Statistical analyses including kappa statistics (κ) for interobserver agreement, Fisher's exact test, and univariate and multivariate multinomial logistic regression were performed. RESULTS: BE distribution between the two classifications and between each classification and mammographic density were associated (P < 0.001). BI-RADS homogeneous-fibroglandular (76.8%) and modified heterogeneous BE (71.3%, 75.7%, and 87.5% of mild, moderate, and marked heterogeneous background echotexture, respectively) tended to be dense. BE was correlated between BI-RADS homogeneous-fat and modified homogeneous background (95.1%) and between BI-RADS homogeneous-fibroglandular or heterogeneous (90.6%) and modified heterogeneous (86.9%) (P < 0.001). In multinomial logistic regression, age < 50 years was independently associated with heterogeneous BE (OR, 8.89, P = 0.003, in BI-RADS; OR, 3.74; P = 0.020 in modified classification). CONCLUSION: BI-RADS homogeneous-fat and modified homogeneous BE on ABUS was likely to be mammographically fatty. However, BI-RADS homogeneous-fibroglandular or heterogeneous BE might be classified as any modified BE. Younger age was independently associated with heterogeneous BE.

Consistency of Artificial Intelligence (AI)-based Diagnostic Support Software in Short-term Digital Mammography Reimaging After Core Needle Biopsy.

To evaluate the consistency in the performance of Artificial Intelligence (AI)-based diagnostic support software in short-term digital mammography reimaging after core needle biopsy. Of 276 women who underwent short-term (<3 mo) serial digital mammograms followed by breast cancer surgery from Jan. to Dec. 2017, 550 breasts were included. All core needle biopsies for breast lesions were performed between serial exams. All mammography images were analyzed using a commercially available AI-based software providing an abnormality score (0-100). Demographic data for age, interval between serial exams, biopsy, and final diagnosis were compiled. Mammograms were reviewed for mammographic density and finding. Statistical analysis was performed to evaluate the distribution of variables according to biopsy and to test the interaction effects of variables with the difference in AI-based score according to biopsy. AI-based score of 550 exams (benign or normal in 263 and malignant in 287) showed significant difference between malignant and benign/normal exams (0.48 vs. 91.97 in first exam and 0.62 vs. 87.13 in second exam, P<0.0001). In comparison of serial exams, no significant difference was found in AI-based score. AI-based score difference between serial exams was significantly different according to biopsy performed or not (-0.25 vs. 0.07, P = 0.035). In linear regression analysis, there was no significant interaction effect of all clinical and mammographic characteristics with mammographic examinations performed after biopsy or not. The results from AI-based diagnostic support software for digital mammography was relatively consistent in short-term reimaging even after core needle biopsy.

Depiction of breast cancers on digital mammograms by artificial intelligence-based computer-assisted diagnosis according to cancer characteristics.

OBJECTIVE: To evaluate how breast cancers are depicted by artificial intelligence-based computer-assisted diagnosis (AI-CAD) according to clinical, radiological, and pathological factors. MATERIALS AND METHODS: From January 2017 to December 2017, 896 patients diagnosed with 930 breast cancers were enrolled in this retrospective study. Commercial AI-CAD was applied to digital mammograms and abnormality scores were obtained. We evaluated the abnormality score according to clinical, radiological, and pathological characteristics. False-negative results were defined by abnormality scores less than 10. RESULTS: The median abnormality score of 930 breasts was 87.4 (range 0-99). The false-negative rate of AI-CAD was 19.4% (180/930). Cancers with an abnormality score of more than 90 showed a high proportion of palpable lesions, BI-RADS 4c and 5 lesions, cancers presenting as mass with or without microcalcifications and invasive cancers compared with low-scored cancers (all p < 0.001). False-negative cancers were more likely to develop in asymptomatic patients and extremely dense breasts and to be diagnosed as occult breast cancers and DCIS compared to detected cancers. CONCLUSION: Breast cancers depicted with high abnormality scores by AI-CAD are associated with higher BI-RADS category, invasive pathology, and higher cancer stage. KEY POINTS: • High-scored cancers by AI-CAD included a high proportion of BI-RADS 4c and 5 lesions, masses with or without microcalcifications, and cancers with invasive pathology. • Among invasive cancers, cancers with higher T and N stage and HER2-enriched subtype were depicted with higher abnormality scores by AI-CAD. • Cancers missed by AI-CAD tended to be in asymptomatic patients and extremely dense breasts and to be diagnosed as occult breast cancers by radiologists.

Added value of abbreviated breast magnetic resonance imaging for assessing suspicious microcalcification on screening mammography-a prospective study.

OBJECTIVES: To investigate the added diagnostic value of abbreviated breast magnetic resonance imaging (MRI) for suspicious microcalcifications on screening mammography. METHODS: This prospective study included 80 patients with suspicious calcifications on screening mammography who underwent abbreviated MRI before undergoing breast biopsy between August 2017 and September 2020. The abbreviated protocol included one pre-contrast and the first post-contrast T1-weighted series. MRI examinations were interpreted as either positive or negative based on the visibility of any significant enhancement. The positive predictive value (PPV) was compared before and after the MRI. RESULTS: Of the 80 suspicious microcalcifications, 33.8% (27/80) were malignant and 66.2% (53/80) were false positives. Abbreviated MRI revealed 33 positive enhancement lesions, and 25 and two lesions showed true-positive and false-negative findings, respectively. Abbreviated MRI increased PPV from 33.8 (27 of 80 cases; 95% CI: 26.2%, 40.8%) to 75.8% (25 of 33 cases; 95% CI: 62.1%, 85.7%). A total of 85% (45 of 53) false-positive diagnoses were reduced after abbreviated MRI assessment. CONCLUSIONS: Abbreviated MRI added significant diagnostic value in patients with suspicious microcalcifications on screening mammography, as demonstrated by a significant increase in PPV with a potential reduction in unnecessary biopsy. KEY POINTS: • Abbreviated breast magnetic resonance imaging increased the positive predictive value of suspicious microcalcifications on screening mammography from 33.8 (27/80 cases) to 75.8% (25/33 cases) (p < .01). • Abbreviated magnetic resonance imaging helped avoid unnecessary benign biopsies in 85% (45/53 cases) of lesions without missing invasive cancer.

Application of Artificial Intelligence Computer-Assisted Diagnosis Originally Developed for Thyroid Nodules to Breast Lesions on Ultrasound.

As thyroid and breast cancer have several US findings in common, we applied an artificial intelligence computer-assisted diagnosis (AI-CAD) software originally developed for thyroid nodules to breast lesions on ultrasound (US) and evaluated its diagnostic performance. From January 2017 to December 2017, 1042 breast lesions (mean size 20.2 ± 11.8 mm) of 1001 patients (mean age 45.9 ± 12.9 years) who underwent US-guided core-needle biopsy were included. An AI-CAD software that was previously trained and validated with thyroid nodules using the convolutional neural network was applied to breast nodules. There were 665 benign breast lesions (63.0%) and 391 breast cancers (37.0%). The area under the receiver operating characteristic curve (AUROC) of AI-CAD to differentiate breast lesions was 0.678 (95% confidence interval: 0.649, 0.707). After fine-tuning AI-CAD with 1084 separate breast lesions, the diagnostic performance of AI-CAD markedly improved (AUC 0.841). This was significantly higher than that of radiologists when the cutoff category was BI-RADS 4a (AUC 0.621, P < 0.001), but lower when the cutoff category was BI-RADS 4b (AUC 0.908, P < 0.001). When applied to breast lesions, the diagnostic performance of an AI-CAD software that had been developed for differentiating malignant and benign thyroid nodules was not bad. However, an organ-specific approach guarantees better diagnostic performance despite the similar US features of thyroid and breast malignancies.

Texture analysis using machine learning-based 3-T magnetic resonance imaging for predicting recurrence in breast cancer patients treated with neoadjuvant chemotherapy.

OBJECTIVES: To determine whether texture analysis for magnetic resonance imaging (MRI) can predict recurrence in patients with breast cancer treated with neoadjuvant chemotherapy (NAC). METHODS: This retrospective study included 130 women who received NAC and underwent subsequent surgery for breast cancer between January 2012 and August 2017. We assessed common features, including standard morphologic MRI features and clinicopathologic features. We used a  commercial software and analyzed texture features from pretreatment and midtreatment MRI. A random forest (RF) method was performed to build a model for predicting recurrence. The diagnostic performance of this model for predicting recurrence was assessed and compared with those of five other machine learning classifiers using the Wald test. RESULTS: Of the 130 women, 21 (16.2%) developed recurrence at a median follow-up of 35.4 months. The RF classifier with common features including clinicopathologic and morphologic MRI features showed the lowest diagnostic performance (area under the receiver operating characteristic curve [AUC], 0.83). The texture analysis with the RF method showed the highest diagnostic performances for pretreatment T2-weighted images and midtreatment DWI and ADC maps showed better diagnostic performance than that of an analysis of common features (AUC, 0.94 vs. 0.83, p < 0.05). The RF model based on all sequences showed a better diagnostic performance for predicting recurrence than did the five other machine learning classifiers. CONCLUSIONS: Texture analysis using an RF model for pretreatment and midtreatment MRI may provide valuable prognostic information for predicting recurrence in patients with breast cancer treated with NAC and surgery. KEY POINTS: • RF model-based texture analysis showed a superior diagnostic performance than traditional MRI and clinicopathologic features (AUC, 0.94 vs.0.83, p < 0.05) for predicting recurrence in breast cancer after NAC. • Texture analysis using RF classifier showed the highest diagnostic performances (AUC, 0.94) for pretreatment T2-weighted images and midtreatment DWI and ADC maps. • RF model showed a better diagnostic performance for predicting recurrence than did the five other machine learning classifiers.

Fully automated measurements of volumetric breast density adapted for BIRADS 5th edition: a comparison with visual assessment.

BACKGROUND: Since the 5th edition of BI-RADS was released, prior studies have compared BI-RADS and quantitative fully automated volumetric assessment, but with software packages that were not recalibrated according to the 5th edition. PURPOSE: To investigate mammographic density assessment of automated volumetric measurements recalibrated according to the BI-RADS 5th edition compared with visual assessment. MATERIAL AND METHODS: A total of 4000 full-field digital mammographic examinations were reviewed by three radiologists for the BI-RADS 5th edition density category by consensus after individual assessments. Volumetric density data obtained using Quantra and Volpara software were collected. The comparison of visual and volumetric density assessments was performed in total and according to the presence of cancer. RESULTS: Among 4000 examinations, 129 were mammograms of breast cancer. Compared to visual assessment, volumetric measurements showed higher category B (40.6% vs. 19.8%) in Quantra, and higher category D (40.4% vs. 14.7%) and lower category A (0.2% vs. 5.0%) in Volpara (P < 0.0001). All volumetric data showed a difference according to visually assessed categories and were correlated between the two volumetric measurements (P < 0.0001). The group with cancer showed a lower proportion of fatty breast than that without cancer: 17.8% vs. 46.9% for Quantra (P < 0.0001) and 9.3% vs. 21.5% for Volpara (P = 0.003). Both measurements showed significantly higher mean density data in the group with cancer than without cancer (P < 0.005 for all). CONCLUSION: Automated volumetric measurements adapted for the BI-RADS 5th edition showed different but correlated results with visual assessment and each other. Recalibration of volumetric measurement has not completely reflected the visual assessment.

Texture Analysis with 3.0-T MRI for Association of Response to Neoadjuvant Chemotherapy in Breast Cancer.

Background Previous studies have suggested that texture analysis is a promising tool in the diagnosis, characterization, and assessment of treatment response in various cancer types. Therefore, application of texture analysis may be helpful for early prediction of pathologic response in breast cancer. Purpose To investigate whether texture analysis of features from MRI is associated with pathologic complete response (pCR) to neoadjuvant chemotherapy (NAC) in breast cancer. Materials and Methods This retrospective study included 136 women (mean age, 47.9 years; range, 31-70 years) who underwent NAC and subsequent surgery for breast cancer between January 2012 and August 2017. Patients were monitored with 3.0-T MRI before (pretreatment) and after (midtreatment) three or four cycles of NAC. Texture analysis was performed at pre- and midtreatment T2-weighted MRI, contrast material-enhanced T1-weighted MRI, diffusion-weighted MRI, and apparent diffusion coefficient (ADC) mapping by using commercial software. A random forest method was applied to build a predictive model for classifying those with pCR with use of texture parameters. Diagnostic performance for predicting pCR was assessed and compared with that of six other machine learning classifiers (adaptive boosting, decision tree, k-nearest neighbor, linear support vector machine, naive Bayes, and linear discriminant analysis) by using the Wald test and DeLong method. Results Forty of the 136 patients (29%) achieved pCR after NAC. In the prediction of pCR, the random forest classifier showed the lowest diagnostic performance with pretreatment ADC (area under the receiver operating characteristic curve [AUC], 0.53; 95% confidence interval: 0.44, 0.61) and the highest diagnostic performance with midtreatment contrast-enhanced T1-weighted MRI (AUC, 0.82; 95% confidence interval: 0.74, 0.88) among pre- and midtreatment T2-weighted MRI, contrast-enhanced T1-weighted MRI, diffusion-weighted MRI, and ADC mapping. Conclusion Texture parameters using a random forest method of contrast-enhanced T1-weighted MRI at midtreatment of neoadjuvant chemotherapy were valuable and associated with pathologic complete response in breast cancer. © RSNA, 2019 Online supplemental material is available for this article.

Dynamic contrast-enhanced and diffusion-weighted MRI of invasive breast cancer for the prediction of sentinel lymph node status.

BACKGROUND: Although sentinel lymph node biopsy (SLNB) is the current standard for identifying lymph metastasis in breast cancer patients, there are complications of SLNB. PURPOSE: To evaluate preoperative dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) and diffusion-weighted imaging (DWI) of invasive breast cancer for predicting sentinel lymph node metastasis. STUDY TYPE: Retrospective. POPULATION: In all, 309 patients who underwent clinically node-negative invasive breast cancer surgery FIELD STRENGTH/SEQUENCE: 3.0T, DCE-MRI, DWI. ASSESSMENT: We collected clinicopathologic variables (age, histologic and nuclear grade, extensive intraductal carcinoma component, lymphovascular invasion, and immunohistochemical profiles) and preoperative MRI features (tumor size, background parenchymal enhancement, internal enhancement, adjacent vessel sign, whole-breast vascularity, initial enhancement pattern, kinetic curve types, quantitative kinetic parameters, tumoral apparent diffusion coefficient [ADC], peritumoral maximal ADC, and peritumoral-tumoral ADC ratio). STATISTICAL TESTS: Multivariate logistic regressions were performed to determine independent variables associated with SLN metastasis, and the area under the receiver operating characteristic curve (AUC) was analyzed for those variables. RESULTS: 41 (13.3%) of the patients showed SLN metastasis. With MRI, tumor size (odds ratio [OR], 1.11; 95% confidence interval [CI], 1.06-1.17), heterogeneous (OR, 5.33; 95% CI, 1.71-16.58), and rim (OR, 15.54; 95% CI, 2.12-113.72) enhancement and peritumoral-tumoral ADC ratio (OR, 72.79; 95% CI, 7.15-740.82) were independently associated with SLN metastasis. Clinicopathologic variables independently associated with SLN metastasis included age (OR, 0.96; 95% CI, 0.92-0.99) and CD31 (OR, 2.90; 95% CI, 1.04-8.92). The area under the curve (AUC) of MRI features (0.80; 95% CI, 0.73-0.87) was significantly higher than for clinicopathologic variables (0.68; 95% CI, 0.60-0.77; P = 0.048) and was barely below statistical significance for combined MRI features with clinicopathologic variables (0.84; 95% CI 0.78-0.90, P = 0.057). DATA CONCLUSION: Preoperative internal enhancement on DCE-MRI and peritumoral-tumoral ADC ratio on DWI might be useful for predicting SLN metastasis in patients with invasive breast cancer. LEVEL OF EVIDENCE: 3 Technical Efficacy: Stage 2 J. Magn. Reson. Imaging 2020;51:615-626.

Prediction of axillary response by monitoring with ultrasound and MRI during and after neoadjuvant chemotherapy in breast cancer patients.

PURPOSE: To investigate whether monitoring with ultrasound and MR imaging before, during and after neoadjuvant chemotherapy (NAC) can predict axillary response in breast cancer patients. MATERIALS AND METHODS: A total of 131 breast cancer patients with clinically positive axillary lymph node (LN) who underwent NAC and subsequent surgery were enrolled. They had ultrasound and 3.0 T-MR examinations before, during and after NAC. After reviewing ultrasound and MR images, axillary LN features and tumour size (T size) were noted. According to LN status after surgery, imaging features and their diagnostic performances were analysed. RESULTS: Of the 131 patients, 60 (45.8%) had positive LNs after surgery. Pre-NAC T size at ultrasound and MR was different in positive LN status after surgery (p < 0.01). There were significant differences in mid- and post-NAC number, cortical thickness (CxT), T size and T size reduction at ultrasound and mid- and post-NAC CxT, hilum, T size and T size reduction, and post-NAC ratio of diameter at MR (p < 0.03). On multivariate analysis, pre-NAC MR T size (OR, 1.03), mid-NAC ultrasound T size (OR, 1.05) and CxT (OR, 1.53), and post-NAC MR T size (OR, 1.06) and CxT (OR, 1.64) were independently associated with positive LN (p < 0.004). Combined mid-NAC ultrasound T size and CxT showed the best diagnostic performance with AUC of 0.760. CONCLUSION: Monitoring ultrasound and MR axillary LNs and T size can be useful to predict axillary response to NAC in breast cancer patients. KEY POINTS: • Monitoring morphologic features of LNs is useful to predict axillary response. • Monitoring tumour size by imaging is useful to predict axillary response. • The axillary ultrasound during NAC showed the highest diagnostic performance.

A nomogram constructed using intraoperative ex vivo shear-wave elastography precisely predicts metastasis of sentinel lymph nodes in breast cancer.

OBJECTIVE: To develop a nomogram and validate its use for the intraoperative evaluation of nodal metastasis using shear-wave elastography (SWE) elasticity values and nodal size METHODS: We constructed a nomogram to predict metastasis using ex vivo SWE values and ultrasound features of 228 axillary LNs from fifty-five patients. We validated its use in an independent cohort comprising 80 patients. In the validation cohort, a total of 217 sentinel LNs were included. RESULTS: We developed the nomogram using the nodal size and elasticity values of the development cohort to predict LN metastasis; the area under the curve (AUC) was 0.856 (95% confidence interval (CI), 0.783-0.929). In the validation cohort, 15 (7%) LNs were metastatic, and 202 (93%) were non-metastatic. The mean stiffness (23.54 and 10.41 kPa, p = 0.005) and elasticity ratio (3.24 and 1.49, p = 0.028) were significantly higher in the metastatic LNs than those in the non-metastatic LNs. However, the mean size of the metastatic LNs was not significantly larger than that of the non-metastatic LNs (8.70 mm vs 7.20 mm, respectively; p = 0.123). The AUC was 0.791 (95% CI, 0.668-0.915) in the validation cohort, and the calibration plots of the nomogram showed good agreement. CONCLUSIONS: We developed a well-validated nomogram to predict LN metastasis. This nomogram, mainly based on ex vivo SWE values, can help evaluate nodal metastasis during surgery. KEY POINTS: • A nomogram was developed based on axillary LN size and ex vivo SWE values such as mean stiffness and elasticity ratio to easily predict axillary LN metastasis during breast cancer surgery. • The constructed nomogram presented high predictive performance of sentinel LN metastasis with an independent cohort. • This nomogram can reduce unnecessary intraoperative frozen section which increases the surgical time and costs in breast cancer patients.

Grayscale Ultrasound Radiomic Features and Shear-Wave Elastography Radiomic Features in Benign and Malignant Breast Masses.

PURPOSE: To identify and compare diagnostic performance of radiomic features between grayscale ultrasound (US) and shear-wave elastography (SWE) in breast masses. MATERIALS AND METHODS: We retrospectively collected 328 pathologically confirmed breast masses in 296 women who underwent grayscale US and SWE before biopsy or surgery. A representative SWE image of the mass displayed with a grayscale image in split-screen mode was selected. An ROI was delineated around the mass boundary on the grayscale image and copied and pasted to the SWE image by a dedicated breast radiologist for lesion segmentation. A total of 730 candidate radiomic features including first-order statistics and textural and wavelet features were extracted from each image. LASSO regression was used for data dimension reduction and feature selection. Univariate and multivariate logistic regression was performed to identify independent radiomic features, differentiating between benign and malignant masses with calculation of the AUC. RESULTS: Of 328 breast masses, 205 (62.5 %) were benign and 123 (37.5 %) were malignant. Following radiomic feature selection, 22 features from grayscale and 6 features from SWE remained. On univariate analysis, all 6 SWE radiomic features (P < 0.0001) and 21 of 22 grayscale radiomic features (P < 0.03) were significantly different between benign and malignant masses. After multivariate analysis, three grayscale radiomic features and two SWE radiomic features were independently associated with malignant breast masses. The AUC was 0.929 for grayscale US and 0.992 for SWE (P < 0.001). CONCLUSION: US radiomic features may have the potential to improve diagnostic performance for breast masses, but further investigation of independent and larger datasets is needed.

Correction to: Effect of training on ultrasonography (US) BI-RADS features for radiology residents: a multicenter study comparing performances after training.

The original version of this article, published on 07 January 2019, unfortunately contained a mistake.

Effect of training on ultrasonography (US) BI-RADS features for radiology residents: a multicenter study comparing performances after training.

OBJECTIVES: To evaluate the effect of training radiology residents on breast ultrasonography (US) according to the Breast Imaging Reporting And Data System (BI-RADS) and the factors that influence the training effect. METHODS: This multicenter, prospective study was approved by eight institutional review boards. From September 2013 to July 2014, 248 breast masses in 227 women were included for US image acquisition. Representative B-mode and video images of the breast masses were recorded, among which 54 cases were included in the education set and 66 in the test set. Sixty-one radiology residents scheduled for breast imaging training individually reviewed the test set, immediately before, 1 month after, and 6 months after training. Diagnostic performances and US descriptors of the residents were evaluated and compared against those of expert radiologists. RESULTS: Agreements between residents and experienced radiologists showed improvement after training, while agreements between post-training and post-6-month training descriptors did not show significant differences (all p > 0.05, respectively). Sensitivity, negative predictive value (NPV), and AUC were significantly improved for residents post-training and post-6-month training (all p < 0.05), while approximating the performances of expert radiologists except for AUC (0.836, 0.840, and 0.908, respectively, p < 0.05). Low levels of pre-training AUC, total number of breast US examinations, and the number of sessions per week that residents were involved in were factors influencing the improvement of AUC. CONCLUSION: Training using education material dedicated for breast US imaging effectively improved the diagnostic performances of radiology residents and agreements with experienced radiologists on US BI-RADS features. KEY POINTS: • Agreements on lesion descriptors between residents and experienced radiologists showed improvement after training, regardless of test point. • Sensitivity, NPV, and AUC were significantly improved for residents in post-training and post-6-month training (all p < 0.05). • Low levels of pre-training AUC, total number of breast US examinations, and the number of sessions per week that residents were involved in were factors influencing the improvement of AUC.

Performance of shear-wave elastography for breast masses using different region-of-interest (ROI) settings.

Background Various size and shape of region of interest (ROI) can be applied for shear-wave elastography (SWE). Purpose To investigate the diagnostic performance of SWE according to ROI settings for breast masses. Material and Methods To measure elasticity for 142 lesions, ROIs were set as follows: circular ROIs 1 mm (ROI-1), 2 mm (ROI-2), and 3 mm (ROI-3) in diameter placed over the stiffest part of the mass; freehand ROIs drawn by tracing the border of mass (ROI-M) and the area of peritumoral increased stiffness (ROI-MR); and circular ROIs placed within the mass (ROI-C) and to encompass the area of peritumoral increased stiffness (ROI-CR). Mean (Emean), maximum (Emax), and standard deviation (ESD) of elasticity values and their areas under the receiver operating characteristic (ROC) curve (AUCs) for diagnostic performance were compared. Results Means of Emean and ESD significantly differed between ROI-1, ROI-2, and ROI-3 ( P < 0.0001), whereas means of Emax did not ( P = 0.50). For ESD, ROI-1 (0.874) showed a lower AUC than ROI-2 (0.964) and ROI-3 (0.975) ( P < 0.002). The mean ESD was significantly different between ROI-M and ROI-MR and between ROI-C and ROI-CR ( P < 0.0001). The AUCs of ESD in ROI-M and ROI-C were significantly lower than in ROI-MR ( P = 0.041 and 0.015) and ROI-CR ( P = 0.007 and 0.004). Conclusion Shear-wave elasticity values and their diagnostic performance vary based on ROI settings and elasticity indices. Emax is recommended for the ROIs over the stiffest part of mass and an ROI encompassing the peritumoral area of increased stiffness is recommended for elastic heterogeneity of mass.

Evaluation of Screening US-detected Breast Masses by Combined Use of Elastography and Color Doppler US with B-Mode US in Women with Dense Breasts: A Multicenter Prospective Study.

Purpose To investigate the value of the combined use of elastography and color Doppler ultrasonography (US) with B-mode US for evaluation of screening US-detected breast masses in women with dense breasts. Materials and Methods This prospective, multicenter study included asymptomatic women with dense breasts who were referred for screening US between November 2013 and December 2014. Eligible women had a newly detected breast mass at conventional B-mode US screening, for which elastography and color Doppler US were performed. The following outcome measures were compared between B-mode US and the combination of B-mode US, elastography, and color Doppler US: area under the receiver operating characteristic curve (AUC), sensitivity, specificity, positive predictive value (PPV), and the number of false-positive findings at screening US. Results Among 1021 breast masses (mean size, 1.0 cm; range, 0.3-3.0 cm) in 1021 women (median age, 45 years), 68 were malignant (56 invasive). Addition of elastography and color Doppler US to B-mode US increased the AUC from 0.87 (95% confidence interval [CI]: 0.82, 0.91) to 0.96 (95% CI: 0.95, 0.98; P < .001); specificity from 27.0% (95% CI: 24.2%, 29.9%) to 76.4% (95% CI: 73.6%, 79.1%; P < .001) without loss in sensitivity (95% CI: -1.5%, 1.5%; P > .999); and PPV from 8.9% (95% CI: 7.0%, 11.2%) to 23.2% (95% CI: 18.5%, 28.5%; P < .001), while avoiding 67.7% (471 of 696) of unnecessary biopsies for nonmalignant lesions. Conclusion Addition of elastography and color Doppler US to B-mode US can increase the PPV of screening US in women with dense breasts while reducing the number of false-positive findings without missing cancers. © RSNA, 2017 Online supplemental material is available for this article.

Predictive Factors for Active Surveillance of Subcentimeter Thyroid Nodules with Highly Suspicious US Features.

PURPOSE: This study was designed to assess the outcomes of subcentimeter thyroid nodules with highly suspicious ultrasonography (US) features and to investigate the predictive factors associated with malignancy and aggressive biological behavior to determine appropriate candidate factors for active surveillance. METHODS: Between June 2011 and December 2013, 1866 subcentimeter thyroid nodules with highly suspicious US features that were subjected to US-guided fine needle aspiration and subsequent surgery or US follow-up of at least 2 years were evaluated. A multivariate logistic regression analysis was performed to identify independent clinical characteristics and US features associated with the malignancy rate and aggressive biological behavior. RESULTS: Of the 1866 subcentimeter thyroid nodules, 821 (44.0%) were benign and 1045 (56.0%) were malignant. Age younger than 45 years, presence of microcalcification, and a taller than wide shape on US were associated independently with malignancy in the subcentimeter thyroid nodules (P < 0.05). Of 1041 evaluated papillary microcarcinomas, a multivariate analysis revealed that male gender, presence of microcalcification, and a taller than wide on US were independently associated with lymph node metastasis and ATA intermediate risk (P < 0.01). CONCLUSIONS: Age younger than 45 years, male gender, and subcentimeter thyroid nodules exhibiting microcalcification, and a taller than wide shape on US might be not good candidate factors for active surveillance.

Comparison of Visual Assessment of Breast Density in BI-RADS 4th and 5th Editions With Automated Volumetric Measurement.

OBJECTIVE: The purpose of this study was to compare visual assessments of mammographic breast density by radiologists using BI-RADS 4th and 5th editions in correlation with automated volumetric breast density measurements. MATERIALS AND METHODS: A total of 337 consecutive full-field digital mammographic examinations with standard views were retrospectively assessed by two radiologists for mammographic breast density according to BI-RADS 4th and 5th editions. Fully automated measurement of the volume of fibroglandular tissue and total breast and percentage breast density was performed with a commercially available software program. Interobserver and intraobserver agreement was assessed with kappa statistics. The distributions of breast density categories for both editions of BI-RADS were compared and correlated with volumetric data. RESULTS: Interobserver agreement on breast density category was moderate to substantial (κ = 0.58-0.63) with use of BI-RADS 4th edition and substantial (κ = 0.63-0.66) with use of the 5th edition but without significant difference between the two editions. For intraobserver agreement between the two editions, the distributions of density category were significantly different (p < 0.0001), the proportions of dense breast increased, and the proportion of fatty breast decreased with use of the 5th edition compared with the 4th edition (p < 0.0001). All volumetric breast density data, including percentage breast density, were significantly different among density categories (p < 0.0001) and had significant correlation with visual assessment for both editions of BI-RADS (p < 0.01). CONCLUSION: Assessment using BI-RADS 5th edition revealed a higher proportion of dense breast than assessment using BI-RADS 4th edition. Nevertheless, automated volumetric density assessment had good correlation with visual assessment for both editions of BI-RADS.

Shear-Wave Elastography for Papillary Thyroid Carcinoma can Improve Prediction of Cervical Lymph Node Metastasis.

BACKGROUND: This study aimed to investigate whether the elasticity index of shear-wave elastography (SWE) can predict cervical lymph node (LN) metastasis of papillary thyroid carcinoma (PTC). METHODS: This retrospective study included 363 patients with a surgical diagnosis of PTC who underwent preoperative SWE evaluation. The elasticity indices of PTC (E mean, E max, E min, E ratio-p, and E ratio-m) and gray-scale ultrasound (US) parameters (extrathyroidal extension, multifocality, and cervical LN metastasis) were correlated with the pathologic staging parameters. The optimal cutoff values for the elasticity indices were determined for the prediction of cervical LN metastasis, and diagnostic performance was compared between gray-scale US and the combined application of gray-scale US and SWE. RESULTS: The findings showed E mean and E max to be associated with central LN metastasis (P = 0.037) and E min to be associated with lateral LN metastasis (P = 0.015). An E mean value higher than 124 kPa or an E max value higher than 138 kPa with suspicious gray-scale US findings improved the sensitivity and area under the curve (AUC) for predicting central LN metastasis (sensitivity, 45.4 and 44.6 % vs. 28 %, P < 0.001; AUC, 0.659 and 0.667 vs. 0.615, P = 0.011 and 0.019), whereas an E min value higher than 63 kPa with suspicious gray-scale US findings improved the sensitivity and AUC for predicting lateral LN metastasis (sensitivity, 95.8 vs. 75 %, P = 0.025; AUC, 0.924 vs. 0.871, P = 0.047). CONCLUSION: The quantitative elasticity index of PTC on preoperative SWE could be useful for predicting cervical LN metastasis.