AI analytics can be used as imaging biomarkers for predicting invasive upgrade of ductal carcinoma in situ.

OBJECTIVES: To evaluate whether the quantitative abnormality scores provided by artificial intelligence (AI)-based computer-aided detection/diagnosis (CAD) for mammography interpretation can be used to predict invasive upgrade in ductal carcinoma in situ (DCIS) diagnosed on percutaneous biopsy. METHODS: Four hundred forty DCIS in 420 women (mean age, 52.8 years) diagnosed via percutaneous biopsy from January 2015 to December 2019 were included. Mammographic characteristics were assessed based on imaging features (mammographically occult, mass/asymmetry/distortion, calcifications only, and combined mass/asymmetry/distortion with calcifications) and BI-RADS assessments. Routine pre-biopsy 4-view digital mammograms were analyzed using AI-CAD to obtain abnormality scores (AI-CAD score, ranging 0-100%). Multivariable logistic regression was performed to identify independent predictive mammographic variables after adjusting for clinicopathological variables. A subgroup analysis was performed with mammographically detected DCIS. RESULTS: Of the 440 DCIS, 117 (26.6%) were upgraded to invasive cancer. Three hundred forty-one (77.5%) DCIS were detected on mammography. The multivariable analysis showed that combined features (odds ratio (OR): 2.225, p = 0.033), BI-RADS 4c or 5 assessments (OR: 2.473, p = 0.023 and OR: 5.190, p < 0.001, respectively), higher AI-CAD score (OR: 1.009, p = 0.007), AI-CAD score ≥ 50% (OR: 1.960, p = 0.017), and AI-CAD score ≥ 75% (OR: 2.306, p = 0.009) were independent predictors of invasive upgrade. In mammographically detected DCIS, combined features (OR: 2.194, p = 0.035), and higher AI-CAD score (OR: 1.008, p = 0.047) were significant predictors of invasive upgrade. CONCLUSION: The AI-CAD score was an independent predictor of invasive upgrade for DCIS. Higher AI-CAD scores, especially in the highest quartile of ≥ 75%, can be used as an objective imaging biomarker to predict invasive upgrade in DCIS diagnosed with percutaneous biopsy. CRITICAL RELEVANCE STATEMENT: Noninvasive imaging features including the quantitative results of AI-CAD for mammography interpretation were independent predictors of invasive upgrade in lesions initially diagnosed as ductal carcinoma in situ via percutaneous biopsy and therefore may help decide the direction of surgery before treatment. KEY POINTS: • Predicting ductal carcinoma in situ upgrade is important, yet there is a lack of conclusive non-invasive biomarkers. • AI-CAD scores-raw numbers, ≥ 50%, and ≥ 75%-predicted ductal carcinoma in situ upgrade independently. • Quantitative AI-CAD results may help predict ductal carcinoma in situ upgrade and guide patient management.

Unenhanced Breast MRI With Diffusion-Weighted Imaging for Breast Cancer Detection: Effects of Training on Performance and Agreement of Subspecialty Radiologists.

OBJECTIVE: To investigate whether reader training improves the performance and agreement of radiologists in interpreting unenhanced breast magnetic resonance imaging (MRI) scans using diffusion-weighted imaging (DWI). MATERIALS AND METHODS: A study of 96 breasts (35 cancers, 24 benign, and 37 negative) in 48 asymptomatic women was performed between June 2019 and October 2020. High-resolution DWI with b-values of 0, 800, and 1200 sec/mm² was performed using a 3.0-T system. Sixteen breast radiologists independently reviewed the DWI, apparent diffusion coefficient maps, and T1-weighted MRI scans and recorded the Breast Imaging Reporting and Data System (BI-RADS) category for each breast. After a 2-h training session and a 5-month washout period, they re-evaluated the BI-RADS categories. A BI-RADS category of 4 (lesions with at least two suspicious criteria) or 5 (more than two suspicious criteria) was considered positive. The per-breast diagnostic performance of each reader was compared between the first and second reviews. Inter-reader agreement was evaluated using a multi-rater κ analysis and intraclass correlation coefficient (ICC). RESULTS: Before training, the mean sensitivity, specificity, and accuracy of the 16 readers were 70.7% (95% confidence interval [CI]: 59.4-79.9), 90.8% (95% CI: 85.6-94.2), and 83.5% (95% CI: 78.6-87.4), respectively. After training, significant improvements in specificity (95.2%; 95% CI: 90.8-97.5; P = 0.001) and accuracy (85.9%; 95% CI: 80.9-89.8; P = 0.01) were observed, but no difference in sensitivity (69.8%; 95% CI: 58.1-79.4; P = 0.58) was observed. Regarding inter-reader agreement, the κ values were 0.57 (95% CI: 0.52-0.63) before training and 0.68 (95% CI: 0.62-0.74) after training, with a difference of 0.11 (95% CI: 0.02-0.18; P = 0.01). The ICC was 0.73 (95% CI: 0.69-0.74) before training and 0.79 (95% CI: 0.76-0.80) after training (P = 0.002). CONCLUSION: Brief reader training improved the performance and agreement of interpretations by breast radiologists using unenhanced MRI with DWI.

Learnability of Thyroid Nodule Assessment on Ultrasonography: Using a Big Data Set.

OBJECTIVE: The aims of the work described here were to evaluate the learnability of thyroid nodule assessment on ultrasonography (US) using a big data set of US images and to evaluate the diagnostic utilities of artificial intelligence computer-aided diagnosis (AI-CAD) used by readers with varying experience to differentiate benign and malignant thyroid nodules. METHODS: Six college freshmen independently studied the "learning set" composed of images of 13,560 thyroid nodules, and their diagnostic performance was evaluated after their daily learning sessions using the "test set" composed of images of 282 thyroid nodules. The diagnostic performance of two residents and an experienced radiologist was evaluated using the same "test set." After an initial diagnosis, all readers once again evaluated the "test set" with the assistance of AI-CAD. RESULTS: Diagnostic performance of almost all students increased after the learning program. Although the mean areas under the receiver operating characteristic curves (AUROCs) of residents and the experienced radiologist were significantly higher than those of students, the AUROCs of five of the six students did not differ significantly compared with that of the one resident. With the assistance of AI-CAD, sensitivity significantly increased in three students, specificity in one student, accuracy in four students and AUROC in four students. Diagnostic performance of the two residents and the experienced radiologist was better with the assistance of AI-CAD. CONCLUSION: A self-learning method using a big data set of US images has potential as an ancillary tool alongside traditional training methods. With the assistance of AI-CAD, the diagnostic performance of readers with varying experience in thyroid imaging could be further improved.

Machine learning­based radiomics models for prediction of locoregional recurrence in patients with breast cancer.

Locoregional recurrence (LRR) is the predominant pattern of relapse after definitive breast cancer treatment. The present study aimed to develop machine learning (ML)-based radiomics models to predict LRR in patients with breast cancer by using preoperative magnetic resonance imaging (MRI) data. Data from patients with localized breast cancer that underwent preoperative MRI between January 2013 and December 2017 were collected. Propensity score matching (PSM) was performed to adjust for clinical factors between patients with and without LRR. Radiomics features were obtained from T2-weighted with and without fat-suppressed MRI and contrast-enhanced T1-weighted with fat-suppressed MRI. In the present study five ML models were designed, three base models (support vector machine, random forest, and logistic regression) and two ensemble models (voting model and stacking model) composed of the three base models, and the performance of each base model was compared with the stacking model. After PSM, 28 patients with LRR and 86 patients without LRR were included. Of these 114 patients, 80 patients were randomly selected to train the models, and the remaining 34 patients were used to evaluate the performance of the trained models. In total, 5,064 features were obtained from each patient, and 47-51 features were selected by applying variance threshold and least absolute shrinkage and selection operator. The stacking model demonstrated superior performance in area under the receiver operating characteristic curve (AUC), with an AUC of 0.78 compared to a range of 0.61 to 0.70 for the other models. An external validation study to investigate the efficacy of the stacking model of the present study was initiated and is still ongoing (Korean Radiation Oncology Group 2206).

Corrigendum: Diffusion-Weighted Magnetic Resonance Imaging for Preoperative Evaluation of Patients With Breast Cancer: Protocol of a Prospective, Multicenter, Observational Cohort Study.

This corrects the article on p. 292 in vol. 26, PMID: 37272245.

Diffusion-Weighted Magnetic Resonance Imaging for Preoperative Evaluation of Patients With Breast Cancer: Protocol of a Prospective, Multicenter, Observational Cohort Study.

PURPOSE: Detection of multifocal, multicentric, and contralateral breast cancers in patients affects surgical management. Dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) can identify additional foci that were initially undetected by conventional imaging. However, its use is limited owing to low specificity and high false-positive rate. Multiparametric MRI (DCE-MRI + diffusion-weighted [DW] MRI) can increase the specificity. We aimed to describe the protocols of our prospective, multicenter, observational cohort studies designed to compare the diagnostic performance of DCE-MRI and multiparametric MRI for the diagnosis of multifocal, multicentric cancer and contralateral breast cancer in patients with newly diagnosed breast cancer. METHODS: Two studies comparing the performance of DCE-MRI and multiparametric MRI for the diagnosis of multifocal, multicentric cancer (NCT04656639) and contralateral breast cancer (NCT05307757) will be conducted. For trial NCT04656639, 580 females with invasive breast cancer candidates for breast conservation surgery whose DCE-MRI showed additional suspicious lesions (breast imaging reporting and data system [BI-RADS] category ≥ 4) on DCE-MRI in the ipsilateral breast will be enrolled. For trial NCT05307757, 1098 females with invasive breast cancer whose DCE-MRI showed contralateral lesions (BI-RADS category ≥ 3 or higher on DCE-MRI) will be enrolled. Participants will undergo 3.0-T DCE-MRI and DW-MRI. The diagnostic performance of DCE-MRI and multiparametric MRI will be compared. The receiver operating characteristic curve, sensitivity, specificity, positive predictive value, and characteristics of the detected cancers will be analyzed. The primary outcome is the difference in the receiver operating characteristic curve between DCE-MRI and multiparametric MRI interpretation. Enrollment completion is expected in 2024, and study results are expected to be presented in 2026. DISCUSSION: This prospective, multicenter study will compare the performance of DCE-MRI versus multiparametric MRI for the preoperative evaluation of multifocal, multicentric, and contralateral breast cancer and is currently in the patient enrollment phase. TRIAL REGISTRATION: ClinicalTrials.gov Identifier: NCT04656639, NCT05307757. Registered on April 1 2022.

Changes in Automated Mammographic Breast Density Can Predict Pathological Response After Neoadjuvant Chemotherapy in Breast Cancer.

OBJECTIVE: Mammographic density is an independent risk factor for breast cancer that can change after neoadjuvant chemotherapy (NCT). This study aimed to evaluate percent changes in volumetric breast density (ΔVbd%) before and after NCT measured automatically and determine its value as a predictive marker of pathological response to NCT. MATERIALS AND METHODS: A total of 357 patients with breast cancer treated between January 2014 and December 2016 were included. An automated volumetric breast density (Vbd) measurement method was used to calculate Vbd on mammography before and after NCT. Patients were divided into three groups according to ΔVbd%, calculated as follows: Vbd (post-NCT - pre-NCT)/pre-NCT Vbd × 100 (%). The stable, decreased, and increased groups were defined as -20% ≤ ΔVbd% ≤ 20%, ΔVbd% < -20%, and ΔVbd% > 20%, respectively. Pathological complete response (pCR) was considered to be achieved after NCT if there was no evidence of invasive carcinoma in the breast or metastatic tumors in the axillary and regional lymph nodes on surgical pathology. The association between ΔVbd% grouping and pCR was analyzed using univariable and multivariable logistic regression analyses. RESULTS: The interval between the pre-NCT and post-NCT mammograms ranged from 79 to 250 days (median, 170 days). In the multivariable analysis, ΔVbd% grouping (odds ratio for pCR of 0.420 [95% confidence interval, 0.195-0.905; P = 0.027] for the decreased group compared with the stable group), N stage at diagnosis, histologic grade, and breast cancer subtype were significantly associated with pCR. This tendency was more evident in the luminal B-like and triple-negative subtypes. CONCLUSION: ΔVbd% was associated with pCR in breast cancer after NCT, with the decreased group showing a lower rate of pCR than the stable group. Automated measurement of ΔVbd% may help predict the NCT response and prognosis in breast cancer.

Diagnosis of thyroid micronodules on ultrasound using a deep convolutional neural network.

To assess the performance of deep convolutional neural network (CNN) to discriminate malignant and benign thyroid nodules < 10 mm in size and compare the diagnostic performance of CNN with those of radiologists. Computer-aided diagnosis was implemented with CNN and trained using ultrasound (US) images of 13,560 nodules ≥ 10 mm in size. Between March 2016 and February 2018, US images of nodules < 10 mm were retrospectively collected at the same institution. All nodules were confirmed as malignant or benign from aspirate cytology or surgical histology. Diagnostic performances of CNN and radiologists were assessed and compared for area under curve (AUC), sensitivity, specificity, accuracy, positive predictive value, and negative predictive value. Subgroup analyses were performed based on nodule size with a cut-off value of 5 mm. Categorization performances of CNN and radiologists were also compared. A total of 370 nodules from 362 consecutive patients were assessed. CNN showed higher negative predictive value (35.3% vs. 22.6%, P = 0.048) and AUC (0.66 vs. 0.57, P = 0.04) than radiologists. CNN also showed better categorization performance than radiologists. In the subgroup of nodules ≤ 5 mm, CNN showed higher AUC (0.63 vs. 0.51, P = 0.08) and specificity (68.2% vs. 9.1%, P < 0.001) than radiologists. Convolutional neural network trained with thyroid nodules ≥ 10 mm in size showed overall better diagnostic performance than radiologists in the diagnosis and categorization of thyroid nodules < 10 mm, especially in nodules ≤ 5 mm.

Radiomics Analysis of Gray-Scale Ultrasonographic Images of Papillary Thyroid Carcinoma > 1 cm: Potential Biomarker for the Prediction of Lymph Node Metastasis.

Purpose: This study aimed to investigate radiomics analysis of ultrasonographic images to develop a potential biomarker for predicting lymph node metastasis in papillary thyroid carcinoma (PTC) patients. Materials and Methods: This study included 431 PTC patients from August 2013 to May 2014 and classified them into the training and validation sets. A total of 730 radiomics features, including texture matrices of gray-level co-occurrence matrix and gray-level run-length matrix and single-level discrete two-dimensional wavelet transform and other functions, were obtained. The least absolute shrinkage and selection operator method was used for selecting the most predictive features in the training data set. Results: Lymph node metastasis was associated with the radiomics score (p < 0.001). It was also associated with other clinical variables such as young age (p = 0.007) and large tumor size (p = 0.007). The area under the receiver operating characteristic curve was 0.687 (95% confidence interval: 0.616-0.759) for the training set and 0.650 (95% confidence interval: 0.575-0.726) for the validation set. Conclusion: This study showed the potential of ultrasonography-based radiomics to predict cervical lymph node metastasis in patients with PTC; thus, ultrasonography-based radiomics can act as a biomarker for PTC.

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.

AI-CAD for differentiating lesions presenting as calcifications only on mammography: outcome analysis incorporating the ACR BI-RADS descriptors for calcifications.

OBJECTIVES: To evaluate how AI-CAD triages calcifications and to compare its performance to an experienced breast radiologist. METHODS: Among routine mammography performed between June 2016 and May 2018, 535 lesions detected as calcifications only on mammography in 500 women (mean age, 48.8 years) that were additionally interpreted with additional magnification views were included in this study. One dedicated breast radiologist retrospectively reviewed the magnification mammograms to assess morphology, distribution, and final assessment category according to ACR BI-RADS. AI-CAD analyzed routine mammograms providing AI-CAD marks and corresponding AI-CAD scores (ranging from 0 to 100%), for which values ≥ 10% were considered positive. Ground truth in terms of malignancy or benignity was confirmed with a histopathologic diagnosis or at least 1 year of imaging follow - up. RESULTS: Of the 535 calcifications, 215 (40.2%) were malignant. Calcifications with positive AI-CAD scores showed significantly higher PPVs compared to calcifications with negative scores for all morphology (all p < 0.05). PPVs were significantly higher in calcifications with positive AI-CAD scores compared to those with negative scores for BI-RADS 3, 4a, or 4b assessments (all p < 0.05). AI-CAD and the experienced radiologist did not show significant difference in diagnostic performance; sensitivity 92.1% vs 95.4% (p = 0.125), specificity 71.9% vs 72.5% (p = 0.842), and accuracy 80.0% vs 81.7% (p = 0.413). CONCLUSION: Among calcifications with same morphology or BI-RADS assessment, those with positive AI-CAD scores had significantly higher PPVs. AI-CAD showed similar diagnostic performances to the experienced radiologist for calcifications detected on mammography. KEY POINTS: • Among calcifications with same morphology or BI-RADS assessment, those with positive AI-CAD scores had significantly higher PPVs. • AI-CAD showed similar diagnostic performance to an experienced radiologist in assessing lesions detected as calcifications only on mammography. • Among malignant calcifications, calcifications with positive AI-CAD scores showed higher rates of invasive cancers than calcifications with negative scores (all p > 0.05).

Corrigendum: Diffusion-Weighted Magnetic Resonance Imaging for Breast Cancer Screening in High-Risk Women: Design and Imaging Protocol of a Prospective Multicenter Study in Korea.

This corrects the article on p. 218 in vol. 24, PMID: 33913277.

Sarcopenia increases the risk of major organ or vessel invasion in patients with papillary thyroid cancer.

While sarcopenia is associated with poor overall survival and cancer-specific survival in solid cancer patients, the impact of sarcopenia on clinicopathologic features that can influence conventional papillary thyroid cancer (PTC) prognosis remains unclear. To investigate the impact of sarcopenia on aggressive clinicopathologic features in PTC patients, prospectively collected data on 305 patients who underwent surgery for PTC with preoperative staging ultrasonography and bioelectrical impedance analysis were retrospectively analyzed. Nine sarcopenia patients with preoperative sarcopenia showed more patients aged 55 or older (p = 0.022), higher male proportion (p < 0.001), lower body-mass index (p = 0.015), higher incidence of major organ or vessel invasion (p = 0.001), higher T stage (p = 0.002), higher TNM stage (p = 0.007), and more tumor recurrence (p = 0.023) compared to the non-sarcopenia patients. Unadjusted and adjusted logistic regression analyses showed that sarcopenia (odds ratio (OR) 9.936, 95% confidence interval (CI) 2.052-48.111, p = 0.004), tumor size (OR 1.048, 95% CI 1.005-1.093, p = 0.027), and tumor multiplicity (OR 3.323, 95% CI 1.048-10.534, p = 0.041) significantly increased the risk of T4 cancer. Sarcopenia patients showed significantly lower disease-free survival probability compared to non-sarcopenia patients. Therefore, preoperative sarcopenia in PTC patients should raise clinical suspicion for a more locally advanced disease and direct appropriate management and careful follow-up.

Feasibility study using multifocal Doppler twinkling artifacts to detect suspicious microcalcifications in ex vivo specimens of breast cancer on US.

Multifocal Doppler twinkling artifact (MDTA) imaging has shown high detection rates of microcalcifications in phantom studies. We aimed to evaluate its performance in detecting suspicious microcalcifications in comparison with mammography by using ex vivo breast cancer specimens. We prospectively included ten women with breast cancer that presented with calcifications on mammography. Both digital specimen mammography and MDTA imaging were performed for ex vivo breast cancer specimens on the day of surgery. Five breast radiologists marked cells that included suspicious microcalcifications (referred to as 'positive cell') on specimen mammographic images using a grid of 5-mm cells. Cells that were marked by at least three readers were considered as 'consensus-positive'. Matched color Doppler twinkling artifact (CDTA) signals were compared between reconstructed US-MDTA projection images and mammographic images. The median detection rate for each case was 74.7% for positive cells and 96.7% for consensus-positive cells. Of the 10 cases, 90% showed a detection rate of ≥ 80%, with 50% of cases showing a 100% detection rate for consensus-positive cells. The proposed MDTA imaging method showed high performance for detecting suspicious microcalcifications in ex vivo breast cancer specimens, and may be a feasible approach for detecting suspicious breast microcalcifications with US.

US, Mammography, and Histopathologic Evaluation to Identify Low Nuclear Grade Ductal Carcinoma in Situ.

Background Low nuclear grade ductal carcinoma in situ (DCIS) identified at biopsy can be upgraded to intermediate to high nuclear grade DCIS at surgery. Methods that confirm low nuclear grade are needed to consider nonsurgical approaches for these patients. Purpose To develop a preoperative model to identify low nuclear grade DCIS and to evaluate factors associated with low nuclear grade DCIS at biopsy that was not upgraded to intermediate to high nuclear grade DCIS at surgery. Materials and Methods In this retrospective study, 470 women (median age, 50 years; interquartile range, 44-58 years) with 477 pure DCIS lesions at surgical histopathologic evaluation were included (January 2010 to December 2015). Patients were divided into the training set (n = 330) or validation set (n = 147) to develop a preoperative model to identify low nuclear grade DCIS. Features at US (mass, nonmass) and at mammography (morphologic characteristics, distribution of microcalcification) were reviewed. The upgrade rate of low nuclear grade DCIS was calculated, and multivariable regression was used to evaluate factors for associations with low nuclear grade DCIS that was not upgraded later. Results A preoperative model that included lesions manifesting as a mass at US without microcalcification and no comedonecrosis at biopsy was used to identify low nuclear grade DCIS, with a high area under the receiver operating characteristic curve of 0.97 (95% CI: 0.94, 1.00) in the validation set. The upgrade rate of low nuclear grade DCIS at biopsy was 38.8% (50 of 129). Ki-67 positivity (odds ratio, 0.04; 95% CI: 0.0003, 0.43; P = .005) was inversely associated with constant low nuclear grade DCIS. Conclusion The upgrade rate of low nuclear grade ductal carcinoma in situ (DCIS) at biopsy to intermediate to high nuclear grade DCIS at surgery occurred in more than a third of patients; low nuclear grade DCIS at final histopathologic evaluation could be identified if the mass was viewed at US without microcalcifications and had no comedonecrosis at histopathologic evaluation of biopsy. © RSNA, 2022 Online supplemental material is available for this article. See also the editorial by Rahbar in this issue. An earlier incorrect version appeared online. This article was corrected on April 14, 2022.

Cancer yield and imaging features of probably benign calcifications at digital magnification view.

OBJECTIVES: To investigate the malignancy rate of probably benign calcifications assessed by digital magnification view and imaging and clinical features associated with malignancy. METHODS: This retrospective study included consecutive women with digital magnification views assessed as probably benign for calcifications without other associated mammographic findings from March 2009 to January 2014. Initial studies rendering a probably benign assessment were analyzed, with biopsy or 4-year imaging follow-up. Fisher's exact test and univariable logistic regression were performed. Cancer yields were calculated. RESULTS: A total of 458 lesions in 422 patients were finally included. The overall cancer yield was 2.2% (10 of 458, invasive cancer [n = 4] and DCIS [n = 6]). Calcification distribution (OR = 23.80, p = .041), calcification morphology (OR = 10.84, p = .005), increased calcifications (OR = 29.40, p = .001), and having a concurrent newly diagnosed breast cancer or high-risk lesion (OR = 10.24, p = .001) were associated with malignancy. Cancer yields did not significantly differ between grouped punctate calcifications vs. calcifications with other features (1.2% [2 of 162] vs. 2.7% [8 of 296], p = .506). The cancer yield was 1.6% (7 of 437) in women without newly diagnosed breast cancer or high-risk lesions. CONCLUSION: The cancer yield of probably benign calcifications assessed by digital magnification view was below the 2% threshold for grouped punctate calcifications and for women without newly diagnosed breast cancer or high-risk lesions. Calcification distribution, morphology, increase in calcifications, and the presence of newly diagnosed breast cancer/high-risk lesion were associated with malignancy. KEY POINTS: • Among 458 probably benign calcifications assessed by digital magnification view, the overall cancer yield was 2.2% (10 of 458). • The cancer yield was below the 2% threshold for grouped punctate calcifications (1.2%, 2 of 162) and in women without newly diagnosed breast cancer or high-risk lesions (1.6%, 7 of 437). • Calcification distribution, morphology, increase in calcifications, and the presence of newly diagnosed breast cancer/high-risk lesion were associated with malignancy (all p < .05).

Ultrafast dynamic contrast-enhanced breast MRI: association with pathologic complete response in neoadjuvant treatment of breast cancer.

OBJECTIVES: The purpose of this study was to investigate whether pretreatment kinetic features from ultrafast DCE-MRI are associated with pathological complete response (pCR) in patients with invasive breast cancer and according to immunohistochemistry (IHC) subtype. METHODS: Between August 2018 and June 2019, 256 consecutive breast cancer patients (mean age, 50.2 years; range, 25-86 years) who underwent both ultrafast and conventional DCE-MRI and surgery following neoadjuvant chemotherapy were included. DCE-MRI kinetic features were obtained from pretreatment MRI data. Time-to-enhancement, maximal slope (MS), and volumes at U1 and U2 (U1, time point at which the lesion starts to enhance; U2, subsequent time point after U1) were derived from ultrafast MRI. Logistic regression analysis was performed to identify factors associated with pCR. RESULTS: Overall, 41.4% of all patients achieved pCR. None of the kinetic features was associated with pCR when including all cancers. Among ultrafast DCE-MRI kinetic features, a lower MS (OR, 0.982; p = 0.040) was associated with pCR at univariable analysis in hormone receptor (HR)-positive cancers. In triple-negative cancers, a higher volume ratio U1/U2 was associated with pCR at univariable (OR, 11.787; p = 0.006) and multivariable analysis (OR, 14.811; p = 0.005). Among conventional DCE-MRI kinetic features, a lower peak enhancement (OR, 0.993; p = 0.031) and a lower percentage of washout (OR, 0.904; p = 0.039) was associated with pCR only in HR-positive cancers at univariable analysis. CONCLUSIONS: A higher volume ratio of U1/U2 derived from ultrafast DCE-MRI was independently associated with pCR in triple-negative invasive breast cancer. KEY POINTS: • The ratio of tumor volumes obtained at the first (U1) and second time points (U2) of enhancement was independently associated with pCR in triple-negative invasive breast cancers. • Ultrafast MRI has the potential to improve accuracy in predicting treatment response and personalizing therapy.

Automatic Detection and Segmentation of Breast Cancer on MRI Using Mask R-CNN Trained on Non-Fat-Sat Images and Tested on Fat-Sat Images.

RATIONALE AND OBJECTIVES: Computer-aided methods have been widely applied to diagnose lesions on breast magnetic resonance imaging (MRI). The first step was to identify abnormal areas. A deep learning Mask Regional Convolutional Neural Network (R-CNN) was implemented to search the entire set of images and detect suspicious lesions. MATERIALS AND METHODS: Two DCE-MRI datasets were used, 241 patients acquired using non-fat-sat sequence for training, and 98 patients acquired using fat-sat sequence for testing. All patients have confirmed unilateral mass cancers. The tumor was segmented using fuzzy c-means clustering algorithm to serve as the ground truth. Mask R-CNN was implemented with ResNet-101 as the backbone. The neural network output the bounding boxes and the segmented tumor for evaluation using the Dice Similarity Coefficient (DSC). The detection performance, and the trade-off between sensitivity and specificity, was analyzed using free response receiver operating characteristic. RESULTS: When the precontrast and subtraction image of both breasts were used as input, the false positive from the heart and normal parenchymal enhancements could be minimized. The training set had 1469 positive slices (containing lesion) and 9135 negative slices. In 10-fold cross-validation, the mean accuracy = 0.86 and DSC = 0.82. The testing dataset had 1568 positive and 7264 negative slices, with accuracy = 0.75 and DSC = 0.79. When the obtained per-slice results were combined, 240 of 241 (99.5%) lesions in the training and 98 of 98 (100%) lesions in the testing datasets were identified. CONCLUSION: Deep learning using Mask R-CNN provided a feasible method to search breast MRI, localize, and segment lesions. This may be integrated with other artificial intelligence algorithms to develop a fully automatic breast MRI diagnostic system.