Breast MRI Background Parenchymal Enhancement Categorization Using Deep Learning: Outperforming the Radiologist.

BACKGROUND: Background parenchymal enhancement (BPE) is assessed on breast MRI reports as mandated by the Breast Imaging Reporting and Data System (BI-RADS) but is prone to inter and intrareader variation. Semiautomated and fully automated BPE assessment tools have been developed but none has surpassed radiologist BPE designations. PURPOSE: To develop a deep learning model for automated BPE classification and to compare its performance with current standard-of-care radiology report BPE designations. STUDY TYPE: Retrospective. POPULATION: Consecutive high-risk patients (i.e. >20% lifetime risk of breast cancer) who underwent contrast-enhanced screening breast MRI from October 2013 to January 2019. The study included 5224 breast MRIs, divided into 3998 training, 444 validation, and 782 testing exams. On radiology reports, 1286 exams were categorized as high BPE (i.e., marked or moderate) and 3938 as low BPE (i.e., mild or minimal). FIELD STRENGTH/SEQUENCE: A 1.5 T or 3 T system; one precontrast and three postcontrast phases of fat-saturated T1-weighted dynamic contrast-enhanced imaging. ASSESSMENT: Breast MRIs were used to develop two deep learning models (Slab artificial intelligence (AI); maximum intensity projection [MIP] AI) for BPE categorization using radiology report BPE labels. Models were tested on a heldout test sets using radiology report BPE and three-reader averaged consensus as the reference standards. STATISTICAL TESTS: Model performance was assessed using receiver operating characteristic curve analysis. Associations between high BPE and BI-RADS assessments were evaluated using McNemar's chi-square test (α* = 0.025). RESULTS: The Slab AI model significantly outperformed the MIP AI model across the full test set (area under the curve of 0.84 vs. 0.79) using the radiology report reference standard. Using three-reader consensus BPE labels reference standard, our AI model significantly outperformed radiology report BPE labels. Finally, the AI model was significantly more likely than the radiologist to assign "high BPE" to suspicious breast MRIs and significantly less likely than the radiologist to assign "high BPE" to negative breast MRIs. DATA CONCLUSION: Fully automated BPE assessments for breast MRIs could be more accurate than BPE assessments from radiology reports. LEVEL OF EVIDENCE: 4 TECHNICAL EFFICACY STAGE: 3.

Biopsy Marker Standardization: What's in a Name?

OBJECTIVE. The purposes of this study were to compile mammographic images in various projections showing commercially available breast biopsy site markers and to provide a standardized nomenclature and marker guide to improve physician communication and patient care. MATERIALS AND METHODS. A retrospective review of all breast biopsy markers encountered at one institution was conducted from January 2012 to January 2018. Markers placed at the facility and those placed at outside institutions with the patient subsequently referred to the study institution were included. Additional drawings and photographs and biopsy marker information were compiled from manufacturers and the literature. Intrinsic properties, features, pitfalls, and biopsy marker mimics were recorded from the institution's experience and the literature. RESULTS. Thirty-eight different biopsy marker shapes available from six manufacturers were identified, and mammograms of 37 were compiled and organized by manufacturer. Nomenclature was compiled on the basis of the manufacturer names of each marker. Potential pitfalls and mimics were identified. Manufacturer-reported marker material composition and carrier properties were summarized, including decreased marker migration, enhanced ultrasound visibility, and varying MRI susceptibility. CONCLUSION. Variability in the appearance and nomenclature of breast biopsy site markers may contribute to misinterpretation, miscommunication, and possibly removal of the incorrect lesion. A comprehensive guide to breast biopsy marker nomenclature is clinically useful, and standardization is necessary.

Traumatic Abruptio Placenta Scale (TAPS): a proposed grading system of computed tomography evaluation of placental abruption in the trauma patient.

Placental abruption (PA) is one of the worst possible manifestations of injury in the pregnant trauma patient with ultrasound as the current initial imaging examination of choice, despite its known limitations in placental evaluation. Pregnant patients who undergo computed tomography (CT) for evaluation of potential maternal injuries provide an additional source of imaging for placental evaluation; however, few studies have delineated normal and abnormal placental appearance, therefore resulting in insufficient placental assessments on pregnant trauma patients. Retrospective literature analysis was performed to provide a structured descriptive classification of normal and abnormal placental appearance on CT. By offering a structured system of placental appearance, radiologists will become more familiar with normal variations of the placenta as well as be able to recognize areas of abnormality, furthermore assisting in clinical management efficiency.

Artificial Intelligence Decision Support for Triple-Negative Breast Cancers on Ultrasound.

OBJECTIVE: To assess performance of an artificial intelligence (AI) decision support software in assessing and recommending biopsy of triple-negative breast cancers (TNBCs) on US. METHODS: Retrospective institutional review board-approved review identified patients diagnosed with TNBC after US-guided biopsy between 2009 and 2019. Artificial intelligence output for TNBCs on diagnostic US included lesion features (shape, orientation) and likelihood of malignancy category (benign, probably benign, suspicious, and probably malignant). Artificial intelligence true positive was defined as suspicious or probably malignant and AI false negative (FN) as benign or probably benign. Artificial intelligence and radiologist lesion feature agreement, AI and radiologist sensitivity and FN rate (FNR), and features associated with AI FNs were determined using Wilcoxon rank-sum test, Fisher's exact test, chi-square test of independence, and kappa statistics. RESULTS: The study included 332 patients with 345 TNBCs. Artificial intelligence and radiologists demonstrated moderate agreement for lesion shape and orientation (k = 0.48 and k = 0.47, each P <.001). On the set of examinations using 6 earlier diagnostic US, radiologists recommended biopsy of 339/345 lesions (sensitivity 98.3%, FNR 1.7%), and AI recommended biopsy of 333/345 lesions (sensitivity 96.5%, FNR 3.5%), including 6/6 radiologist FNs. On the set of examinations using immediate prebiopsy diagnostic US, AI recommended biopsy of 331/345 lesions (sensitivity 95.9%, FNR 4.1%). Artificial intelligence FNs were more frequently oval (q < 0.001), parallel (q < 0.001), circumscribed (q = 0.04), and complex cystic and solid (q = 0.006). CONCLUSION: Artificial intelligence accurately recommended biopsies for 96% to 97% of TNBCs on US and may assist radiologists in classifying these lesions, which often demonstrate benign sonographic features.

Cross-site Validation of AI Segmentation and Harmonization in Breast MRI.

This work aims to perform a cross-site validation of automated segmentation for breast cancers in MRI and to compare the performance to radiologists. A three-dimensional (3D) U-Net was trained to segment cancers in dynamic contrast-enhanced axial MRIs using a large dataset from Site 1 (n = 15,266; 449 malignant and 14,817 benign). Performance was validated on site-specific test data from this and two additional sites, and common publicly available testing data. Four radiologists from each of the three clinical sites provided two-dimensional (2D) segmentations as ground truth. Segmentation performance did not differ between the network and radiologists on the test data from Sites 1 and 2 or the common public data (median Dice score Site 1, network 0.86 vs. radiologist 0.85, n = 114; Site 2, 0.91 vs. 0.91, n = 50; common: 0.93 vs. 0.90). For Site 3, an affine input layer was fine-tuned using segmentation labels, resulting in comparable performance between the network and radiologist (0.88 vs. 0.89, n = 42). Radiologist performance differed on the common test data, and the network numerically outperformed 11 of the 12 radiologists (median Dice: 0.85-0.94, n = 20). In conclusion, a deep network with a novel supervised harmonization technique matches radiologists' performance in MRI tumor segmentation across clinical sites. We make code and weights publicly available to promote reproducible AI in radiology.

Preoperative localization of breast lesions: Comparing digital breast tomosynthesis-guided radioactive seed localization versus standard 2D stereotactic radioactive seed localization.

PURPOSE: To compare single seed digital breast tomosynthesis-guided radioseed localization (DBT-L) to standard 2D stereotactic-guided radioseed localization (SGL) of the breast. METHODS: A retrospective review of a large tertiary cancer center's database yielded 68 women who underwent preoperative DBT-L from March 2019-December 2019 and a matched cohort of 65 women who underwent SGL during the same period. The electronic medical record and radiology were reviewed for patient characteristics including breast density, exam technique, pre- and post-operative pathology, exam duration, and radiation dose to the patient. To compare margin outcomes between the groups, the chi-square test of independence was used; to compare continuous outcomes such as exam duration and total dose, the Wilcoxon rank sum test was used. RESULTS: DBT-L and SGL localization targets included biopsy marker (62/68, 91% vs 55/65, 85%), distortion (4/68, 6% vs 2/65, <3%), focal asymmetry (1/68 and 1/65, < 2% for both), calcifications (1/68, <2% vs 4/65, 6%), and anatomic landmarks (0% vs 3/65, 5%). 72% and 71% of localizations were performed for malignant pathology in the DBT-L and SGL groups, respectively. The median duration of DBT-L was 8.3 min vs 10.3 min for SGL, representing statistically significant time savings (p = 0.003). The median total organ dose of DBT-L was 8.6 mGy vs 10.4 mGy for SGL, representing statistically significant dose savings (p = 0.018). The incidence of positive margins at surgery was not statistically different between the groups (p = 0.26). CONCLUSION: DBT-L demonstrates both time and dose savings for the patient compared to SGL without compromising surgical outcome.

Preoperative Localization of Breast MRI Lesions: MRI-guided Marker Placement With Radioactive Seed Localization as an Alternative to MRI-guided Wire Localization.

OBJECTIVE: Preoperative MRI-guided wire localization (MWL) presents challenges to both the physician and patient. In this study, we examined the efficiency and outcome of MRI-guided marker placement followed by mammographic-guided radioactive seed localization (MMP/RSL) as an alternative localization method. The primary outcome parameter was pathology upon excision. The secondary outcome parameters were total procedure time and clinical indication for localization. METHODS: A retrospective review of a large tertiary cancer center's breast imaging database was performed. Records of 21 patients with MMP/RSL (24 markers) from August 2013 to January 2019 were compared with 34 patients receiving MWL (48 wires) from January 2016 to January 2019. Multiple factors, including age, prelocalization pathology, postsurgical pathology, concordance, re-excision rates, and total procedure time required for each technique, were compared. Univariate and descriptive statistical analyses were performed. RESULTS: Mean patient age in years (MMP/RSL = 54.1 ± 13.1, MWL = 55.1 ± 10.8, P = 0.389), time in MR scanner in minutes (MMP/RSL = 31.7 ± 12.0, MWL = 35.8 ± 13.1, P = 0.678), and postsurgical pathology malignancy rates (MMP/RSL = 71.4%, MWL = 65.7%, P = 0.7715) were similar without statistically significant differences. As expected, the mean total procedure time was slightly longer without a statistically significant difference (47.3 ± 19.8 min versus 35.8 ± 13.1 min, P = 0.922) for the MMP/RSL group. All patients in both groups underwent successful localization with 100% radiologic-pathology concordance. Re-excision rates were lower for the MMP/RSL group (9.5%) versus the MWL group (16.7%); however, they were not found to be statistically significant (P = 0.7104). CONCLUSION: MMP/RSL is a feasible alternative to MWL and may alleviate many challenges presented by MWL. Further studies are needed.

Radiologist-Patient Communication: Current Practices and Barriers to Communication in Breast Imaging.

PURPOSE: The aim of this study was to assess variability in radiologist-patient communication practices and barriers to communication among members of the Society of Breast Imaging (SBI). METHODS: A 36-item questionnaire developed by the SBI Patient Care and Delivery Task Force was distributed electronically to SBI members to evaluate patient communication, education, and screening practices. Data from 14 items investigating patient communication (eg, practices, comfort, barriers to communication) were analyzed and compared with demographic variables using χ2 or independent t tests as appropriate. RESULTS: Ninety-three percent of radiologists reported that they directly communicate abnormal results of diagnostic mammographic examinations that require biopsy and malignant or high-risk biopsy results that require surgery. Radiologists (66%) and technologists (57%) often provide normal or negative diagnostic mammographic results. Most respondents were completely comfortable discussing the need for additional imaging, recommending biopsy, and discussing biopsy results directly with patients, and 71% rated their communication skills as excellent. Radiologists who spend less time in breast imaging reported only average communication skills. The most frequent barriers to communication were that practices were not set up for direct communication (loss of revenue) and discomfort with angry patients. CONCLUSIONS: Although variation in breast imaging communication practices exists among radiologists and practice types, the majority of radiologists directly communicate the most distressing results to patients, such as those regarding abnormal diagnostic mammographic findings requiring biopsies and abnormal biopsy results leading to cancer diagnoses and surgery. The majority of radiologists are completely comfortable with these conversations, but all feel that enhancing communication with patients will lead to greater patient satisfaction.

Sonographic evaluation of deep endometriosis: protocol for a US radiology practice.

Endometriosis is a common condition with significant morbidity, including pain and subfertility, which is often subject to a delay in diagnosis. Ultrasound has been successfully utilized, mostly outside North America, to preoperatively stage deep endometriosis, but in these international settings, imaging is typically performed solely by expert radiologists and gynecologists. We outline a method for detailed sonographic survey of the lower abdomen and pelvis to ensure optimum detection and communication of disease extent that is geared to radiologists practicing ultrasound in the United States, with the use of diagnostic medical sonographers.