A Deep Learning Decision Support Tool to Improve Risk Stratification and Reduce Unnecessary Biopsies in BI-RADS 4 Mammograms.

Purpose: To evaluate the performance of a biopsy decision support algorithmic model, the intelligent-augmented breast cancer risk calculator (iBRISK), on a multicenter patient dataset. Materials and Methods: iBRISK was previously developed by applying deep learning to clinical risk factors and mammographic descriptors from 9700 patient records at the primary institution and validated using another 1078 patients. All patients were seen from March 2006 to December 2016. In this multicenter study, iBRISK was further assessed on an independent, retrospective dataset (January 2015-June 2019) from three major health care institutions in Texas, with Breast Imaging Reporting and Data System (BI-RADS) category 4 lesions. Data were dichotomized and trichotomized to measure precision in risk stratification and probability of malignancy (POM) estimation. iBRISK score was also evaluated as a continuous predictor of malignancy, and cost savings analysis was performed. Results: The iBRISK model's accuracy was 89.5%, area under the receiver operating characteristic curve (AUC) was 0.93 (95% CI: 0.92, 0.95), sensitivity was 100%, and specificity was 81%. A total of 4209 women (median age, 56 years [IQR, 45-65 years]) were included in the multicenter dataset. Only two of 1228 patients (0.16%) in the "low" POM group had malignant lesions, while in the "high" POM group, the malignancy rate was 85.9%. iBRISK score as a continuous predictor of malignancy yielded an AUC of 0.97 (95% CI: 0.97, 0.98). Estimated potential cost savings were more than $420 million. Conclusion: iBRISK demonstrated high sensitivity in the malignancy prediction of BI-RADS 4 lesions. iBRISK may safely obviate biopsies in up to 50% of patients in low or moderate POM groups and reduce biopsy-associated costs.Keywords: Mammography, Breast, Oncology, Biopsy/Needle Aspiration, Radiomics, Precision Mammography, AI-augmented Biopsy Decision Support Tool, Breast Cancer Risk Calculator, BI-RADS 4 Mammography Risk Stratification, Overbiopsy Reduction, Probability of Malignancy (POM) Assessment, Biopsy-based Positive Predictive Value (PPV3) Supplemental material is available for this article. Published under a CC BY 4.0 license.See also the commentary by McDonald and Conant in this issue.

Optoacoustic Breast Imaging: Imaging-Pathology Correlation of Optoacoustic Features in Benign and Malignant Breast Masses.

OBJECTIVE: Optoacoustic ultrasound breast imaging is a fused anatomic and functional modality that shows morphologic features, as well as hemoglobin amount and relative oxygenation within and around breast masses. The purpose of this study is to investigate the positive predictive value (PPV) of optoacoustic ultrasound features in benign and malignant masses. SUBJECTS AND METHODS: In this study, 92 masses assessed as BI-RADS category 3, 4, or 5 in 94 subjects were imaged with optoacoustic ultrasound. Each mass was scored by seven blinded independent readers according to three internal features in the tumor interior and two external features in its boundary zone and periphery. Mean and median optoacoustic ultrasound scores were compared with histologic findings for biopsied masses and nonbiopsied BI-RADS category 3 masses, which were considered benign if they were stable at 12-month follow-up. Statistical significance was analyzed using a two-sided Wilcoxon rank sum test with a 0.05 significance level. RESULTS: Mean and median optoacoustic ultrasound scores for all individual internal and external features, as well as summed scores, were higher for malignant masses than for benign masses (p < 0.0001). High external scores, indicating increased hemoglobin and deoxygenation and abnormal vessel morphologic features in the tumor boundary zone and periphery, better distinguished benign from malignant masses than did high internal scores reflecting increased hemoglobin and deoxygenation within the tumor interior. CONCLUSION: High optoacoustic ultrasound scores, particularly those based on external features in the boundary zone and periphery of breast masses, have high PPVs for malignancy and, conversely, low optoacoustic ultrasound scores have low PPV for malignancy. The functional component of optoacoustic ultrasound may help to overcome some of the limitations of morphologic overlap in the distinction of benign and malignant masses.

Downgrading and Upgrading Gray-Scale Ultrasound BI-RADS Categories of Benign and Malignant Masses With Optoacoustics: A Pilot Study.

OBJECTIVE: False-positive findings remain challenging in breast imaging. This study investigates the incremental value of optoacoustic imaging in improving BI-RADS categorization of breast masses at ultrasound. SUBJECTS AND METHODS: The study device is an optoacoustic breast imaging device with a handheld duplex laser and internal gray-scale ultrasound probe, fusing functional and morphologic information (optoacoustic ultrasound). In this prospective multisite study, breast masses assessed as BI-RADS category 3, 4A, 4B, 4C, or 5 by site radiologists underwent both gray-scale ultrasound and optoacoustic imaging with the study device. Independent reader radiologists assessed internal gray-scale ultrasound and optoacoustic ultrasound features for each mass and assigned a BI-RADS category. The percentage of mass reads for which optoacoustic ultrasound resulted in a downgrade or upgrade of BI-RADS category relative to internal gray-scale ultrasound was determined. RESULTS: Of 94 total masses, 39 were biopsy-proven malignant, 44 were biopsy-proven benign, and 11 BI-RADS category 3 masses were stable at 12-month follow-up. The sensitivity of both optoacoustic ultrasound and internal gray-scale ultrasound was 97.1%. The specificity was 44.3% for optoacoustic ultrasound and 36.4% for internal gray-scale ultrasound. Using optoacoustic ultrasound, 41.7% of benign masses or BI-RADS category 3 masses that were stable at 12-month follow-up were downgraded to BI-RADS category 2 by independent readers; 36.6% of masses assigned BI-RADS category 4A were downgraded to BI-RADS category 3 or 2, and 10.1% assigned BI-RADS category 4B were downgraded to BI-RADS category 3 or 2. Using optoacoustic ultrasound, independent readers upgraded 75.0% of the malignant masses classified as category 4A, 4B, 4C, or 5, and 49.4% of the malignant masses were classified as category 4B, 4C, or 5. CONCLUSION: Optoacoustic ultrasound resulted in BI-RADS category downgrading of benign masses and upgrading of malignant masses compared with gray-scale ultrasound.

Idiopathic Granulomatous Mastitis: Manifestations at Multimodality Imaging and Pitfalls.

Idiopathic granulomatous mastitis (IGM) is a rare benign inflammatory breast entity characterized by lobulocentric granulomas. IGM has a persistent or recurrent disease course and affects parous premenopausal women with a history of lactation. It has also been associated with hyperprolactinemia. The most common clinical sign is a palpable tender mass. However, the nonspecific manifestations and varied demographic features of this condition, as well as the other similar-appearing and superimposed breast entities, pose substantial diagnostic challenges. Entities with similar manifestations include inflammatory breast cancer (IBC), infective mastitis, foreign body injection granulomas, mammary duct ectasia, diabetic fibrous mastopathy, and systemic granulomatous processes. The strategy for imaging IGM depends on patient age, clinical manifestations, and risk factors. Targeted ultrasonography, mammography, and less commonly, magnetic resonance imaging have proven to be useful for imaging evaluation. Core-needle biopsy, with or without fine-needle aspiration for cytopathologic examination, and culture analysis are usually required to exclude IBC and other benign inflammatory breast processes. Patients with IGM have an excellent prognosis when they are appropriately treated with oral steroids or second-line immunosuppressive and prolactin-lowering medications. However, surgical excision may be an option for patients in whom medication therapy is unsuccessful. Imaging surveillance can be offered to patients with incidentally encountered IGM or mild symptoms. Clinical suspicion for this rare disease and the breast imager's prompt diagnosis can lead to an improved patient outcome. The purpose of this article is to review the imaging manifestations of IGM in a multimodality case-based format and to describe relevant clinical and imaging-based differential diagnoses. The associated pitfalls, epidemiologic and histopathologic factors, clinical manifestations, natural course, and management of IGM also are discussed. ©RSNA, 2018.

A Pivotal Study of Optoacoustic Imaging to Diagnose Benign and Malignant Breast Masses: A New Evaluation Tool for Radiologists.

Purpose To compare the diagnostic utility of an investigational optoacoustic imaging device that fuses laser optical imaging (OA) with grayscale ultrasonography (US) to grayscale US alone in differentiating benign and malignant breast masses. Materials and Methods This prospective, 16-site study of 2105 women (study period: 12/21/2012 to 9/9/2015) compared Breast Imaging Reporting and Data System (BI-RADS) categories assigned by seven blinded independent readers to benign and malignant breast masses using OA/US versus US alone. BI-RADS 3, 4, or 5 masses assessed at diagnostic US with biopsy-proven histologic findings and BI-RADS 3 masses stable at 12 months were eligible. Independent readers reviewed US images obtained with the OA/US device, assigned a probability of malignancy (POM) and BI-RADS category, and locked results. The same independent readers then reviewed OA/US images, scored OA features, and assigned OA/US POM and a BI-RADS category. Specificity and sensitivity were calculated for US and OA/US. Benign and malignant mass upgrade and downgrade rates, positive and negative predictive values, and positive and negative likelihood ratios were compared. Results Of 2105 consented subjects with 2191 masses, 100 subjects (103 masses) were analyzed separately as a training population and excluded. An additional 202 subjects (210 masses) were excluded due to technical failures or incomplete imaging, 72 subjects (78 masses) due to protocol deviations, and 41 subjects (43 masses) due to high-risk histologic results. Of 1690 subjects with 1757 masses (1079 [61.4%] benign and 678 [38.6%] malignant masses), OA/US downgraded 40.8% (3078/7535) of benign mass reads, with a specificity of 43.0% (3242/7538, 99% confidence interval [CI]: 40.4%, 45.7%) for OA/US versus 28.1% (2120/7543, 99% CI: 25.8%, 30.5%) for the internal US of the OA/US device. OA/US exceeded US in specificity by 14.9% (P < .0001; 99% CI: 12.9, 16.9%). Sensitivity for biopsied malignant masses was 96.0% (4553/4745, 99% CI: 94.5%, 97.0%) for OA/US and 98.6% (4680/4746, 99% CI: 97.8%, 99.1%) for US (P < .0001). The negative likelihood ratio of 0.094 for OA/US indicates a negative examination can reduce a maximum US-assigned pretest probability of 17.8% (low BI-RADS 4B) to a posttest probability of 2% (BI-RADS 3). Conclusion OA/US increases the specificity of breast mass assessment compared with the device internal grayscale US alone. Online supplemental material is available for this article. © RSNA, 2017.