Differences between human and machine perception in medical diagnosis.

Deep neural networks (DNNs) show promise in image-based medical diagnosis, but cannot be fully trusted since they can fail for reasons unrelated to underlying pathology. Humans are less likely to make such superficial mistakes, since they use features that are grounded on medical science. It is therefore important to know whether DNNs use different features than humans. Towards this end, we propose a framework for comparing human and machine perception in medical diagnosis. We frame the comparison in terms of perturbation robustness, and mitigate Simpson's paradox by performing a subgroup analysis. The framework is demonstrated with a case study in breast cancer screening, where we separately analyze microcalcifications and soft tissue lesions. While it is inconclusive whether humans and DNNs use different features to detect microcalcifications, we find that for soft tissue lesions, DNNs rely on high frequency components ignored by radiologists. Moreover, these features are located outside of the region of the images found most suspicious by radiologists. This difference between humans and machines was only visible through subgroup analysis, which highlights the importance of incorporating medical domain knowledge into the comparison.

Axillary Adenopathy after COVID-19 Vaccine: No Reason to Delay Screening Mammogram.

Online supplemental material is available for this article.

Breast MRI for Evaluation of Response to Neoadjuvant Therapy.

Neoadjuvant therapy is increasingly being used to treat early-stage triple-negative and human epidermal growth factor 2-overexpressing breast cancers, as well as locally advanced and inflammatory breast cancers. The rationales for neoadjuvant therapy are to shrink tumor size and potentially decrease the extent of surgery, to serve as an in vivo test of response to therapy, and to reveal prognostic information for the patient. MRI is the most accurate modality to demonstrate response to therapy and to help ensure accurate presurgical planning. Changes in lesion diameter, volume, and enhancement are used to predict complete response, partial response, or nonresponse to therapy. However, residual disease may be overestimated or underestimated at MRI. Fibrosis, necrotic tumors, and residual benign masses may be causes of overestimation of residual disease. Nonmass lesions, invasive lobular carcinoma, hormone receptor-positive tumors, nonconcentric shrinkage patterns, the use of antiangiogenic therapy, and late-enhancing foci may be causes of underestimation of residual disease. In patients with known axillary lymph node metastasis, neoadjuvant therapy may be followed by targeted axillary dissection to avoid the potential morbidity associated with an axillary lymph node dissection. Diffusion-weighted imaging, radiomics, machine learning, and deep learning methods are under investigation to improve MRI accuracy in predicting treatment response.©RSNA, 2021.

Abbreviated MR Imaging for Breast Cancer.

Breast MR imaging is the most sensitive imaging method for the detection of breast cancer and detects more aggressive malignancies than mammography and ultrasound examination. Despite these advantages, breast MR imaging has low use rates for breast cancer screening. Abbreviated breast MR imaging, in which a limited number of breast imaging sequences are obtained, has been proposed as a way to solve cost and patient tolerance issues while preserving the high cancer detection rate of breast MR imaging. This review discusses abbreviated breast MR imaging, including protocols, multicenter clinical trial results, clinical workflow implementation challenges, and future directions.

Deep Neural Networks Improve Radiologists' Performance in Breast Cancer Screening.

We present a deep convolutional neural network for breast cancer screening exam classification, trained, and evaluated on over 200000 exams (over 1000000 images). Our network achieves an AUC of 0.895 in predicting the presence of cancer in the breast, when tested on the screening population. We attribute the high accuracy to a few technical advances. 1) Our network's novel two-stage architecture and training procedure, which allows us to use a high-capacity patch-level network to learn from pixel-level labels alongside a network learning from macroscopic breast-level labels. 2) A custom ResNet-based network used as a building block of our model, whose balance of depth and width is optimized for high-resolution medical images. 3) Pretraining the network on screening BI-RADS classification, a related task with more noisy labels. 4) Combining multiple input views in an optimal way among a number of possible choices. To validate our model, we conducted a reader study with 14 readers, each reading 720 screening mammogram exams, and show that our model is as accurate as experienced radiologists when presented with the same data. We also show that a hybrid model, averaging the probability of malignancy predicted by a radiologist with a prediction of our neural network, is more accurate than either of the two separately. To further understand our results, we conduct a thorough analysis of our network's performance on different subpopulations of the screening population, the model's design, training procedure, errors, and properties of its internal representations. Our best models are publicly available at https://github.com/nyukat/breast_cancer_classifier.

Abbreviated Breast MRI: Road to Clinical Implementation.

Breast MRI offers high sensitivity for breast cancer detection, with preferential detection of high-grade invasive cancers when compared to mammography and ultrasound. Despite the clear benefits of breast MRI in cancer screening, its cost, patient tolerance, and low utilization remain key issues. Abbreviated breast MRI, in which only a select number of sequences and postcontrast imaging are acquired, exploits the high sensitivity of breast MRI while reducing table time and reading time to maximize availability, patient tolerance, and accessibility. Worldwide studies of varying patient populations have demonstrated that the comparable diagnostic accuracy of abbreviated breast MRI is comparable to a full diagnostic protocol, highlighting the emerging role of abbreviated MRI screening in patients with an intermediate and high lifetime risk of breast cancer. The purpose of this review is to summarize the background and current literature relating to abbreviated MRI, highlight various protocols utilized in current multicenter clinical trials, describe workflow and clinical implementation issues, and discuss the future of abbreviated protocols, including advanced MRI techniques.

Overstated Harms of Breast Cancer Screening? A Large Outcomes Analysis of Complications Associated With 9-Gauge Stereotactic Vacuum-Assisted Breast Biopsy.

OBJECTIVE: The purpose of this study was to assess the rate, type, and severity of complications related to 9-gauge stereotactic vacuum-assisted breast biopsy (SVAB) and to delineate associated factors that may contribute to a higher rate of complications. MATERIALS AND METHODS: This retrospective study included 4776 patients who underwent SVAB between 2003 and 2016. A total of 319 patients with documented postbiopsy complications were identified. Complications were subcategorized as bleeding, pain, lightheadedness, bruising, and other complications, and their severity was classified as minor, moderate, or severe. Hematoma volumes were correlated with biopsy location and complication severity. A group of control subjects who underwent SVAB but had no complications was compared with the group of study patients with regard to age, biopsy location, lesion type, and pathologic findings. Postbiopsy screening adherence was assessed. Statistical analyses were performed using the Fisher exact, Mann-Whitney, Kruskal-Wallis, and Spearman rank correlation tests. RESULTS: Of the 319 patients with complications who were identified (representing 6.7% of the 4776 patients who underwent SVAB), 307 (96.2%) had mild complications, 12 (3.8%) had moderate complications, and no patients had severe complications. The most common complication was bleeding or hematoma (89.3% of patients [285/319]), followed by pain (6.9% [22/319]), lightheadedness (0.9% [3/319]), bruising (0.9% [3/319]), and other complications (1.9% [6/319]). No significant differences were noted between the study group and the control group in terms of age (p = 0.474), biopsy location (p = 0.065), histologic findings (p = 0.056), or lesion type (p = 0.568). Hematoma volume (median, 7.5 cm3) did not correspond to the severity of complications. Larger hematoma volumes were associated with a posterior biopsy location (p = 0.008). The rate of return to annual screening after biopsy was not adversely affected by the presence of biopsy complications. CONCLUSION: Clinically significant complications associated with SVAB were exceedingly rare (0.3%) in this large study spanning 13 years.

Stereotactic Breast Biopsy With Benign Results Does Not Negatively Affect Future Screening Adherence.

PURPOSE: To evaluate whether false-positive stereotactic vacuum-assisted breast biopsy (SVAB) affects subsequent mammographic screening adherence. MATERIALS AND METHODS: This Institutional Review Board-approved, HIPAA-compliant retrospective review of women with SVAB was performed between 2012 and 2014. Patient age, clinical history, biopsy pathology, and first postbiopsy screening mammogram were reviewed. Statistical analyses were performed using Fisher's exact, Mann-Whitney, and χ2 tests. RESULTS: There were 913 SVABs performed in 2012 to 2014 for imaging detected lesions; of these, malignant or high-risk lesions or biopsies resulting in a recommendation of surgical excision were excluded, leaving 395 SVABs yielding benign pathology in 395 women. Findings were matched with a control population consisting of 45,126 women who had a BI-RADS 1 or 2 screening mammogram and did not undergo breast biopsy. In all, 191 of 395 (48.4%) women with a biopsy with benign results and 22,668 of 45,126 (50.2%) women without biopsy returned for annual follow-up >9 months and ≤18 months after the index examination (P = .479). In addition, 57 of 395 (14.4%) women with a biopsy with benign results and 3,336 of 45,126 (7.4%) women without biopsy returned for annual follow-up >18 months after the index examination (P < .001). Older women, women with personal history of breast cancer, and women with postbiopsy complication after benign SVAB were more likely to return for screening (P = .026, P = .028, and P = .026, respectively). CONCLUSION: The findings in our study suggest that SVABs with benign results do not negatively impact screening mammography adherence. The previously described "harms" of false-positive mammography and biopsy may be exaggerated.

Canceled MRI-guided Breast Biopsies Due to Nonvisualization: Follow-up and Outcomes.

RATIONALE AND OBJECTIVE: The objective of this study was to evaluate breast lesion outcomes in patients after canceled MRI-guided breast biopsy due to lesion nonvisualization. MATERIALS AND METHODS: Electronic medical records (January 2007-December 2014) were searched for patients with canceled magnetic resonance imaging (MRI)-guided breast biopsies due to lesion nonvisualization. A total of 1403 MRI-detected lesions were scheduled for MRI-guided biopsy and 89 were canceled because of nonvisualization. Imaging studies and medical records were reviewed for patient demographics, lesion characteristics, and subsequent malignancy. Patients without adequate MRI follow-up imaging were excluded. Statistical analysis was employed to determine if patient demographics or lesion characteristics were predictive of lesion resolution or lesion biopsy after subsequent follow-up. RESULTS: Eighty-nine (6.3% [89/1403]; 95% confidence interval, 5.2%-7.7%) biopsies in 89 women were canceled because of nonvisualization. Follow-up MRIs greater than 5.5 months were available for 60.7% (54/89) of women. In 74.1% (40/54) of these patients, the lesions completely resolved on follow-up. In 25.9% (14/54) of the patients, the lesion persisted on follow-up; 42.9% (6/14) of these patients underwent biopsy. One case (1.9% [1/54]) yielded ductal carcinoma in situ with microinvasion at the 6-month follow-up. No patient demographics or lesion features were associated with lesion resolution or lesion biopsy. CONCLUSIONS: The majority of canceled MRI-guided biopsy lesions resolved on later follow-up; however, because of the small possibility of a missed malignancy, follow-up MRI imaging at 6 months is recommended.

Digital Breast Tomosynthesis Practice Patterns Following 2011 FDA Approval: A Survey of Breast Imaging Radiologists.

RATIONALE AND OBJECTIVES: To evaluate uptake, patterns of use, and perception of digital breast tomosynthesis (DBT) among practicing breast radiologists. MATERIALS AND METHODS: Institutional Review Board exemption was obtained for this Health Insurance Portability and Accountability Act-compliant electronic survey, sent to 7023 breast radiologists identified via the Radiological Society of North America database. Respondents were asked of their geographic location and practice type. DBT users reported length of use, selection criteria, interpretive sequences, recall rate, and reading time. Radiologist satisfaction with DBT as a diagnostic tool was assessed (1-5 scale). RESULTS: There were 1156 (16.5%) responders, 65.8% from the United States and 34.2% from abroad. Of these, 749 (68.6%) use DBT; 22.6% in academia, 56.5% private, and 21% other. Participants are equally likely to report use of DBT if they worked in academics versus in private practice (78.2% [169 of 216] vs 71% [423 of 596]) (odds ratio, 1.10; 95% confidence interval: 0.87-1.40; P = 1.000). Of nonusers, 43% (147 of 343) plan to adopt DBT. No US regional differences in uptake were observed (P = 1.000). Although 59.3% (416 of 702) of DBT users include synthetic 2D (s2D) for interpretation, only 24.2% (170 of 702) use s2D alone. Majority (66%; 441 of 672) do not perform DBT-guided procedures. Radiologist (76.6%) (544 of 710) satisfaction with DBT as a diagnostic tool is high (score ≥ 4/5). CONCLUSIONS: DBT is being adopted worldwide across all practice types, yet variations in examination indication, patient selection, utilization of s2D images, and access to DBT-guided procedures persist, highlighting the need for consensus and standardization.

Magnetic Resonance Imaging-Directed Ultrasound Imaging of Non-Mass Enhancement in the Breast: Outcomes and Frequency of Malignancy.

OBJECTIVES: This study was performed to determine the frequency, predictors, and outcomes of ultrasound (US) correlates for non-mass enhancement. METHODS: From January 2005 to December 2011, a retrospective review of 5837 consecutive breast magnetic resonance imaging examinations at our institution identified 918 non-mass enhancing lesions for which follow-up or biopsy was recommended. Retrospective review of the images identified 879 of 918 lesions (96%) meeting criteria for non-mass enhancement. Patient demographics, pathologic results, and the presence of an adjacent landmark were recorded. Targeted US examinations were recommended for 331 of 879 cases (38%), and 284 of 331 women (86%) underwent US evaluations. RESULTS: The US correlate rate for non-mass enhancement was 23% (64 of 284). An adjacent landmark was significantly associated with a US correlate (P < .001). Biopsy was recommended for 43 of 64 correlates (67%). Ultrasound-guided biopsy was performed on 39 of 43 (91%); 7 of 39 (18%) were malignant. No correlate was seen for 220 of 284 lesions (77%). At magnetic resonance imaging-guided biopsy, 14 of 117 (12%) were malignancies. For all biopsied non-mass enhancements, the malignancy rate was 18% (55 of 308) and was significantly more prevalent in the setting of a known index cancer (P < .001), older age (P < .001), the presence of a landmark (P = .002), and larger lesion size (P = .019). CONCLUSIONS: Non-mass enhancement with an adjacent landmark is more likely to have a US correlate compared to non-mass enhancement without an adjacent landmark. Non-mass enhancement in the setting of a known index cancer, older age, a landmark, and larger lesion size is more likely to be malignant. However, no statistical difference was detected in the rate of malignancy between non-mass enhancement with (18%) or without (12%) a correlate. Absence of a correlate does not obviate the need to biopsy suspicious non-mass enhancement.

Delineating Extramammary Findings at Breast MR Imaging.

Breast magnetic resonance (MR) imaging is the only breast imaging modality that consistently encompasses extramammary structures in the thorax and upper abdomen. Incidental extramammary findings on breast MR images of patients with a history of breast cancer or other malignancies are significantly more likely to be malignant and may affect staging and treatment. An understanding of the frequency, distribution, and context of extramammary findings on breast MR images and a familiarity with common and uncommon sites of breast cancer metastasis inform the differential diagnosis and prompt the appropriate diagnostic next step, to differentiate benign from malignant findings. High-yield organ systems on breast MR images, as reflected by a high positive predictive value for malignancy, are correlated with known distant sites of breast cancer metastasis in the bone, lung, liver, and lymph nodes. Staging is considered when disease involves the skin and chest wall. Unusual sites of breast cancer metastasis from invasive lobular carcinoma are discussed, including the gastrointestinal tract, peritoneum, and adrenal glands. Nonmalignant clinically important findings involving the cardiovascular and gastrointestinal systems are reviewed, and potential pitfalls in diagnosis and interpretation are highlighted. A consistently systematic diagnostic approach is emphasized for identifying extramammary abnormalities on breast MR images. All things considered, the radiologist should be able to improve diagnostic sensitivity and specificity while interpreting extramammary findings on breast MR images. ©RSNA, 2017.

A Prospective, Single Arm, Multi-site, Clinical Evaluation of a Nonradioactive Surgical Guidance Technology for the Location of Nonpalpable Breast Lesions during Excision.

OBJECTIVES: This study was a multicenter evaluation of the SAVI SCOUT(®) breast localization and surgical guidance system using micro-impulse radar technology for the removal of nonpalpable breast lesions. The study was designed to validate the results of a recent 50-patient pilot study in a larger multi-institution trial. The primary endpoints were the rates of successful reflector placement, localization, and removal. METHODS: This multicenter, prospective trial enrolled patients scheduled to have excisional biopsy or breast-conserving surgery of a nonpalpable breast lesion. From March to November 2015, 154 patients were consented and evaluated by 20 radiologists and 16 surgeons at 11 participating centers. Patients had SCOUT(®) reflectors placed up to 7 days before surgery, and placement was confirmed by mammography or ultrasonography. Implanted reflectors were detected by the SCOUT(®) handpiece and console. Presence of the reflector in the excised surgical specimen was confirmed radiographically, and specimens were sent for routine pathology. RESULTS: SCOUT(®) reflectors were successfully placed in 153 of 154 patients. In one case, the reflector was placed at a distance from the target that required a wire to be placed. All 154 lesions and reflectors were successfully removed during surgery. For 101 patients with a preoperative diagnosis of cancer, 86 (85.1 %) had clear margins, and 17 (16.8 %) patients required margin reexcision. CONCLUSIONS: SCOUT(®) provides a reliable and effective alternative method for the localization and surgical excision of nonpalpable breast lesions using no wires or radioactive materials, with excellent patient, radiologist, and surgeon acceptance.

Outcomes of Preoperative MRI-Guided Needle Localization of Nonpalpable Mammographically Occult Breast Lesions.

OBJECTIVE: MRI-guided needle localization allows access to MRI-detected mammographically occult breast lesions that are not amenable to MRI-guided biopsy. The purpose of this study was to examine the safety and outcomes of MRI-guided needle localization. MATERIALS AND METHODS: Ninety-nine consecutive breast lesions that underwent preoperative MRI-guided needle localization were identified. Clinical indications for breast MRI, reasons for performing MRI-guided needle localization, and surgical pathology results were recorded. Lesion characteristics, procedure time, and complications were assessed. RESULTS: Of 99 lesions, 60 (60.6%) were in a location inaccessible for MRI biopsy, necessitating MRI-guided needle localization. Histologic evaluation revealed 38 (38.4%) carcinomas, 31 (31.3%) high-risk lesions, and 30 (30.3%) benign lesions. Carcinoma was more likely to be found in women with known cancer (31/61 [50.8%]; p = 0.003) than in women undergoing imaging for high-risk screening (2/18 [11.1%]) or problem solving (6/20 [30%]). Masses (p = 0.013) and foci (p < 0.001) were more likely to be malignant than were lesions with nonmass enhancement. Foci were significantly more often malignant compared with all other lesion types (9/10 [90%]; p < 0.001). The mean (± SD) procedure time was 32.9 ± 9.39 minutes. All lesions were occult on specimen radiographs. There were no procedure-related complications. CONCLUSION: The positive predictive value of MRI-guided needle localization (38.4%) is comparable to that of mammography- and tomosynthesis-guided localizations and is highest in women with a known diagnosis of cancer. It is highly accurate in targeting small enhancing lesions, thereby improving surgical management. MRI-guided needle localization is a safe, accurate, and time-efficient procedure.

Frequency of Discordant Lesions and False-negative Cancers at Stereotactic Vacuum-assisted Biopsy.

RATIONALE AND OBJECTIVES: This study aimed to determine frequency of discordant lesions and discordant false-negative cancers at stereotactic vacuum-assisted biopsy (SVAB). MATERIALS AND METHODS: Institutional database was searched for discordant SVAB results between January 1, 2005 and December 31, 2012, in this retrospective institutional review board-approved Health Insurance Portability and Accountability Act-compliant study. Patient age, indication for initial mammogram, breast density, lesion size, Breast Imaging Reporting and Data System categorization, operator experience, biopsy needle gauge, biopsy histology, and final surgical histology of discordant lesions were collected and entered into a Microsoft Excel spreadsheet. Discordant rate and false-negative rates were calculated. Fisher exact test was used to assess prevalence of discordance using 11-Gauge needles versus 9-Gauge needles. Patient age, lesion Breast Imaging Reporting and Data System, operator days of experience, mammographic density, and lesion size were evaluated for association with false-negative discordant lesions using an exact Mann-Whitney U test. RESULTS: A total of 1861 SVABs were performed, 224 (12%) with an 11-Gauge VAB device and 1637 (88%) with a 9G Suros or Eviva device. Majority (1409 of 1861; 76%) of SVABs targeted calcifications. Twenty-three of 1861 (1.2%) discordant lesions were identified in 23 women. Seven of 23 (30%) discordant lesions were found to be cancers after final surgical pathology. Needle gauge was not associated with discordance. Operator experience was not associated with false-negative discordance. CONCLUSIONS: A relatively low discordance rate (1.2%) was observed. However, a high percentage (30%; range in literature 11.7%-53.8%) of our discordant lesions were false negatives. This study emphasizes the need for careful radiological-pathologic review after SVAB and for repeat biopsy or surgical excision in the setting of discordance.

The role of dynamic contrast-enhanced screening breast MRI in populations at increased risk for breast cancer.

Breast MRI is more sensitive than mammography in detecting breast cancer. However, MRI as a screening tool is limited to high-risk patients due to cost, low specificity and insufficient evidence for its use in intermediate-risk populations. Nonetheless, in the past decade, there has been a dramatic increase in the use of breast-screening MRI in the community setting. In this review, we set to describe the current literature on the use of screening MRI in high- and intermediate-risk populations. We will also describe novel applications of breast MRI including abbreviated breast MRI protocols, background parenchymal enhancement and diffusion-weighted imaging.

A pilot study of letrozole for one year in women at enhanced risk of developing breast cancer: effects on mammographic density.

BACKGROUND: Tamoxifen or raloxifen for 5 years reduces the risk of developing invasive breast cancer by 40%. To address safety concerns and seek enhanced efficacy, studies of new chemopreventive agents using mammographic density as a surrogate end point are attractive. PATIENTS AND METHODS: Postmenopausal women with risk factors for developing breast cancer were given letrozole 2.5 mg daily for one year, and mammographic density was the biomarker of breast cancer risk modification. It was assessed (blinded to the reader) at baseline, 6, and 12 months in 16 evaluable women among 20 enrolled. RESULTS: Eight patients exhibited decreased mammographic density at six months, and eleven at 12 months. Toxicities included joint aches not precluding continued treatment. CONCLUSION: This pilot study supports the use of letrozole for reducing breast cancer risk. In addition, it encourages prospective studies of serial changes in mammographic density as a biomarker of risk modification within a selected high-risk population.

The relationship of mammographic density and age: implications for breast cancer screening.

OBJECTIVE: Breast density is increasingly recognized as an independent risk factor for the development of breast cancer, because it has been shown to be associated with a four- to sixfold increase in a woman's risk of malignant breast disease. Increased breast density as identified on mammography is also known to decrease the diagnostic sensitivity of the examination, which is of great concern to women at increased risk for breast cancer. Dense tissue has generally been associated with younger age and premenopausal status, with the assumption that breast density gradually decreases after menopause. However, the actual proportion of older women with dense breasts is unknown. The purpose of this study was to examine the relationship between age and breast density, particularly focusing on postmenopausal women. MATERIALS AND METHODS: All screening mammograms completed at the New York University Langone Medical Center in 2008 were retrospectively reviewed. Analysis of variance and descriptive analyses were used to evaluate the relationship between patient age and breast density. RESULTS: A total of 7007 screening mammograms were performed. The median age of our cohort was 57 years. Within each subgroup categorized by decade of age, there was a normal distribution among the categories of breast density. There was a significant inverse relationship between age and breast density (p < 0.001). Seventy-four percent of patients between 40 and 49 years old had dense breasts. This percentage decreased to 57% of women in their 50s. However, 44% of women in their 60s and 36% of women in their 70s had dense breasts as characterized on their screening mammograms. CONCLUSION: In general, we found an inverse relationship between patient age and mammographic breast density. However, there were outliers at the extremes of age. A meaningful proportion of young women had predominantly fatty breasts and a subset of older women had extremely dense breasts. Increased density renders mammography a less sensitive tool for early detection. Breast density should be considered when evaluating the potential benefit of extended imaging for breast cancer screening, especially for women at increased risk for the disease.