Mammographic density measured with quantitative computer-aided method: comparison with radiologists' estimates and BI-RADS categories

PURPOSE: To retrospectively compare computer-aided mammographic density estimation (MDEST) with radiologist estimates of percentage density and Breast Imaging Reporting and Data System (BI-RADS) density classification. MATERIALS AND METHODS: Institutional Review Board approval was obtained for this HIPAA-compliant study; patient informed consent requirements were waived. A fully automated MDEST computer program was used to measure breast density on digitized mammograms in 65 women (mean age, 53 years; range, 24-89 years). Pixel gray levels in detected breast borders were analyzed, and dense areas were segmented. Percentage density was calculated by dividing the number of dense pixels by the total number of pixels within the borders. Seven breast radiologists (five trained with MDEST, two not trained) prospectively assigned qualitative BI-RADS density categories and visually estimated percentage density on 260 mammograms. Qualitative BI-RADS assessments were compared with new quantitative BI-RADS standards. The reference standard density for this study was established by allowing the five trained radiologists to manipulate the MDEST gray-level thresholds, which segmented mammograms into dense and nondense areas. Statistical tests performed include Pearson correlation coefficients, Bland-Altman agreement method, kappa statistics, and unpaired t tests. RESULTS: There was a close correlation between the reference standard and radiologist-estimated density (R = 0.90-0.95) and MDEST density (R = 0.89). Untrained radiologists overestimated percentage density by an average of 37%, versus 6% for trained radiologists (P < .001). MDEST showed better agreement with the reference standard (average overestimate, 1%; range, -15% to +18%). MDEST correlated better with percentage density than with qualitative BI-RADS categories. There were large overlaps and ranges of percentage density in qualitative BI-RADS categories 2-4. Qualitative BI-RADS categories correlated poorly with new quantitative BI-RADS categories, and 16 (6%) of 260 views were erroneously classified by MDEST. CONCLUSION: MDEST compared favorably with radiologist estimates of percentage density and is more reproducible than radiologist estimates when qualitative BI-RADS density categories are used. Qualitative and quantitative BI-RADS density assessments differed markedly.     

Accuracy of assigned BI-RADS breast density category definitions

RATIONALE AND OBJECTIVES: Quantitative criteria for the Breast Imaging Reporting and Data System (BI-RADS) mammographic density categories have recently been defined as <25% dense for almost entirely fatty, 25%-50% dense for scattered fibroglandular densities, 51%-75% for heterogeneously dense, and >75% dense for the extremely dense category. The purpose of this study is to compare the range of percent mammographic densities with radiologist-assigned BI-RADS mammographic density categories and compare with the recently issued definitions. MATERIALS AND METHODS: In this study, 200 consecutive negative analog screening mammograms were assigned BI-RADS mammographic density categories independently by three radiologists blinded to the other readers' density assignment. Quantitative assessment of percent mammographic density was performed using previously validated software. RESULTS: All three readers agreed on BI-RADS mammographic density categories in 98 cases (49%), and two of three readers agreed in all 200 cases. Using two reader's consensus, median mammographic density (range) was 6.0% (0.5%-19.2%) for fatty, 14.8% (1.2%-52.7%) for scattered densities, 51.2% (15.9%-82.2%) for heterogeneously dense, and 78.4% (60.1%-87.9%) for extremely dense breasts. The percent mammographic density ranges for fatty and extremely dense breasts correlated well with BI-RADS definitions, whereas the ranges of densities in the scattered and heterogeneously dense categories were considerably broader. CONCLUSION: Fatty and extremely dense BI-RADS categories compare relatively well to defined criteria, and therefore may be helpful in breast cancer risk models. Scattered fibroglandular densities and heterogeneously dense categories have broad percent mammographic density ranges and may not function well in breast cancer risk models.     

Breast cancer risk prediction model: a nomogram based on common mammographic screening findings

Objectives

To develop a prediction model for breast cancer based on common mammographic findings on screening mammograms aiming to reduce reader variability in assigning BI-RADS.

Methods

We retrospectively reviewed 352 positive screening mammograms of women participating in the Dutch screening programme (Nijmegen region, 2006–2008). The following mammographic findings were assessed by consensus reading of three expert radiologists: masses and mass density, calcifications, architectural distortion, focal asymmetry and mammographic density, and BI-RADS. Data on age, diagnostic workup and final diagnosis were collected from patient records. Multivariate logistic regression analyses were used to build a breast cancer prediction model, presented as a nomogram.

Results

Breast cancer was diagnosed in 108 cases (31 %). The highest positive predictive value (PPV) was found for spiculated masses (96 %) and the lowest for well-defined masses (10 %). Characteristics included in the nomogram are age, mass, calcifications, architectural distortion and focal asymmetry.

Conclusion

With our nomogram we developed a tool assisting screening radiologists in determining the chance of malignancy based on mammographic findings. We propose cutoff values for assigning BI-RADS in the Dutch programme based on our nomogram, which will need to be validated in future research. These values can easily be adapted for use in other screening programmes.

Key points

? There is substantial reader variability in assigning BI-RADS in mammographic screening. ? There are no strict guidelines linking mammographic findings to BI-RADS categories. ? We developed a model (nomogram) predicting the presence of breast cancer. ? Our nomogram is based on common findings on positive screening mammograms. ? The nomogram aims to assist screening radiologists in assigning BI-RADS categories.     

Influences of Radiology Trainees on Screening Mammography Interpretation

PurposeParticipation of radiology trainees in screening mammographic interpretation is a critical component of radiology residency and fellowship training. The aim of this study was to investigate and quantify the effects of trainee involvement on screening mammographic interpretation and diagnostic outcomes.MethodsScreening mammograms interpreted at an academic medical center by six dedicated breast imagers over a three-year period were identified, with cases interpreted by an attending radiologist alone or in conjunction with a trainee. Trainees included radiology residents, breast imaging fellows, and fellows from other radiology subspecialties during breast imaging rotations. Trainee participation, patient variables, results of diagnostic evaluations, and pathology were recorded.ResultsA total of 47,914 mammograms from 34,867 patients were included, with an overall recall rate for attending radiologists reading alone of 14.7% compared with 18.0% when involving a trainee (P < .0001). Overall cancer detection rate for attending radiologists reading alone was 5.7 per 1,000 compared with 5.2 per 1,000 when reading with a trainee (P = .517). When reading with a trainee, dense breasts represented a greater portion of recalls (P = .0001), and more frequently, greater than one abnormality was described in the breast (P = .013). Detection of ductal carcinoma in situ versus invasive carcinoma or invasive cancer type was not significantly different. The mean size of cancers in patients recalled by attending radiologists alone was smaller, and nodal involvement was less frequent, though not statistically significantly.ConclusionsThese results demonstrate a significant overall increase in recall rate when interpreting screening mammograms with radiology trainees, with no change in cancer detection rate. Radiology faculty members should be aware of this potentiality and mitigate tendencies toward greater false positives.     

Comparison of Breast Density Between Synthesized Versus Standard Digital Mammography

PurposeTo evaluate perceptual difference in breast density classification using synthesized mammography (SM) compared with standard or full-field digital mammography (FFDM) for screening.Materials and MethodsThis institutional review board–approved, retrospective, multireader study evaluated breast density on 200 patients who underwent baseline screening mammogram during which both SM and FFDM were obtained contemporaneously from June 1, 2016, through November 30, 2016. Qualitative breast density was independently assigned by seven readers initially evaluating FFDM alone. Then, in a separate session, these same readers assigned breast density using synthetic views alone on the same 200 patients. The readers were again blinded to each other’s assignment. Qualitative density assessment was based on BI-RADS fifth edition. Interreader agreement was evaluated with κ statistic using 95% confidence intervals. Testing for homogeneity in paired proportions was performed using McNemar’s test with a level of significance of .05.ResultsFor patients across the SM and standard 2-D data set, diagnostic testing with McNemar’s test with P = 0.32 demonstrates that the minimal density transitions across FFDM and SM are not statistically significant density shifts. Taking clinical significance into account, only 8 of 200 (4%) patients had clinically significant transition (dense versus not dense). There was substantial interreader agreement with overall κ in FFDM of 0.71 (minimum 0.53, maximum 0.81) and overall SM κ average of 0.63 (minimum 0.56, maximum 0.87).ConclusionOverall subjective breast density assignment by radiologists on SM is similar to density assignment on standard 2-D mammogram.     

Inter- and intraradiologist variability in the BI-RADS assessment and breast density categories for screening mammograms

Objective

The aim of this study was to evaluate reader variability in screening mammograms according to the American College of Radiology Breast Imaging Reporting and Data System (BI-RADS) assessment and breast density categories.

Methods

A stratified random sample of 100 mammograms was selected from a population-based breast cancer screening programme in Barcelona, Spain: 13 histopathologically confirmed breast cancers and 51 with true-negative and 36 with false-positive results. 21 expert radiologists from radiological units of breast cancer screening programmes in Catalonia, Spain, reviewed the mammography images twice within a 6-month interval. The readers described each mammography using BI-RADS assessment and breast density categories. Inter- and intraradiologist agreement was assessed using percentage of concordance and the kappa (κ) statistic.

Results

Fair interobserver agreement was observed for the BI-RADS assessment [κ=0.37, 95% confidence interval (CI) 0.36–0.38]. When the categories were collapsed in terms of whether additional evaluation was required (Categories III, 0, IV, V) or not (I and II), moderate agreement was found (κ=0.53, 95% CI 0.52–0.54). Intra-observer agreement for BI-RADS assessment was moderate using all categories (κ=0.53, 95% CI 0.50–0.55) and substantial on recall (κ=0.66, 95% CI 0.63–0.70). Regarding breast density, inter- and intraradiologist agreement was substantial (κ=0.73, 95% CI 0.72–0.74 and κ=0.69, 95% CI 0.68–0.70, respectively).

Conclusion

We observed a substantial intra-observer agreement in the BI-RADS assessment but only moderate interobserver agreement. Both inter- and intra-observer agreement in mammographic interpretation of breast density was substantial.

Advances in knowledge

Educational efforts should be made to decrease radiologists'' variability in BI-RADS assessment interpretation in population-based breast screening programmes.Breast cancer is the most commonly diagnosed cancer and the leading cause of cancer death in females worldwide, accounting for 23% of the total new cancer cases and 14% of the total cancer-related deaths in 2008 [1]. In Spain, more than 20 000 cases are diagnosed and approximately 6000 females die each year because of this tumour [2].Breast cancer screening by mammography is the only evidence-based screening procedure currently available to reduce breast cancer mortality. Accuracy of screening mammography depends on various factors, such as the protocols for mammogram reading, the characteristics of the female and of the breast, and the experience of radiologists [3-5]. Great efforts have been made to improve its accuracy. One of these is the implementation of double reading as it increases the cancer detection rate and reduces the further assessment rate [6]. Furthermore, the American College of Radiology developed the Breast Imaging Reporting and Data System (BI-RADS) in order to reduce discordance in the interpretation of mammographic findings, to standardise mammographic reporting and to facilitate follow-up [7,8].A limited number of studies have analysed observer variability in mammography interpretation using BI-RADS assessment as well as breast density categories [9-13]. Thus, the aim of this study was to assess the inter- and intra-observer agreement regarding the assessment and breast density in a breast cancer screening programme in the city of Barcelona, Spain. Furthermore, we wanted to investigate the association between female characteristics and BI-RADS discordance.     

Impact of a Deep Learning Model for Predicting Mammographic Breast Density in Routine Clinical Practice

ObjectiveLegislation in 38 states requires patient notification of dense mammographic breast tissue because increased density is a marker of breast cancer risk and can limit mammographic sensitivity. Because radiologist density assessments vary widely, our objective was to implement and measure the impact of a deep learning (DL) model on mammographic breast density assessments in clinical practice.MethodsThis institutional review board–approved prospective study identified consecutive screening mammograms performed across three clinical sites over two periods: 2017 period (January 1, 2017, through September 30, 2017) and 2019 period (January 1, 2019, through September 30, 2019). The DL model was implemented at sites A (academic practice) and B (community practice) in 2018 for all screening mammograms. Site C (community practice) was never exposed to the DL model. Prospective densities were evaluated, and multivariable logistic regression models evaluated the odds of a dense mammogram classification as a function of time and site.ResultsWe identified 85,124 consecutive screening mammograms across the three sites. Across time intervals, odds of a dense classification decreased at sites exposed to the DL model, site A (adjusted odds ratio [aOR], 0.93; 95% confidence interval [CI], 0.86-0.99; P = .024) and site B (aOR, 0.81 [95% CI, 0.70-0.93]; P = .003), and odds increased at the site unexposed to the model (site C) (aOR, 1.13 [95% CI, 1.01-1.27]; P = .033).DiscussionA DL model reduces the odds of screening mammograms categorized as dense. Accurate density assessments could help health care systems more appropriately use limited supplemental screening resources and help better inform traditional clinical risk models.     

Multi-Institutional Assessment and Crowdsourcing Evaluation of Deep Learning for Automated Classification of Breast Density

ObjectiveWe developed deep learning algorithms to automatically assess BI-RADS breast density.MethodsUsing a large multi-institution patient cohort of 108,230 digital screening mammograms from the Digital Mammographic Imaging Screening Trial, we investigated the effect of data, model, and training parameters on overall model performance and provided crowdsourcing evaluation from the attendees of the ACR 2019 Annual Meeting.ResultsOur best-performing algorithm achieved good agreement with radiologists who were qualified interpreters of mammograms, with a four-class κ of 0.667. When training was performed with randomly sampled images from the data set versus sampling equal number of images from each density category, the model predictions were biased away from the low-prevalence categories such as extremely dense breasts. The net result was an increase in sensitivity and a decrease in specificity for predicting dense breasts for equal class compared with random sampling. We also found that the performance of the model degrades when we evaluate on digital mammography data formats that differ from the one that we trained on, emphasizing the importance of multi-institutional training sets. Lastly, we showed that crowdsourced annotations, including those from attendees who routinely read mammograms, had higher agreement with our algorithm than with the original interpreting radiologists.ConclusionWe demonstrated the possible parameters that can influence the performance of the model and how crowdsourcing can be used for evaluation. This study was performed in tandem with the development of the ACR AI-LAB, a platform for democratizing artificial intelligence.     

Mammographic features of calcifications in DCIS: correlation with oestrogen receptor and human epidermal growth factor receptor 2 status

Objective

This study investigated the correlation of oestrogen receptor (ER) and human epidermal growth factor receptor 2 (HER2) status with the probability of malignancy (POM) of mammographic calcifications in ductal carcinoma in situ (DCIS).

Methods

A total of 101 women (age range, 27–83 years) with pure DCIS that presented as mammographic calcifications were included. Three radiologists independently reviewed mammograms according to the BI-RADS lexicon and provided 100-point POM scores and a BI-RADS category. ER, HER2 and breast cancer subtypes were determined using immunohistochemistry (IHC) and fluorescence in situ hybridisation. Pairwise correlations between POM and IHC biomarker scores were calculated, and mammographic features were compared between breast cancer subtypes.

Results

HER2 level positively correlated with the POM score (P?<?0.0001) and BI-RADS category (P?<?0.0001), and ER level inversely correlated with the POM score (P?<?0.013) and BI-RADS category (P?<?0.010). Fine linear branching (P?=?0.004) and segmental (P?=?0.014) calcifications were significantly associated with HER2-positive cancers, and clustered calcifications were more frequently observed in ER-positive cancers (P?=?0.014).

Conclusion

HER2 status in DCIS correlated positively with the POM of mammographic calcifications, as determined by radiologists on the basis of the BI-RADS lexicon.

Key Points

? Prediction of malignancy on mammographic ductal carcinoma in situ is difficult. ? HER2 level correlated positively with the probability of malignancy assigned by radiologists. ? ER level correlated inversely with the probability of malignancy assigned by radiologists. ? HER2-positive DCIS more frequently exhibited fine linear branching or segmental calcifications. ? ER-positive DCIS more frequently exhibited clustered calcifications.     

Mammographic Density Estimation with Automated Volumetric Breast Density Measurement

Objective

To compare automated volumetric breast density measurement (VBDM) with radiologists'' evaluations based on the Breast Imaging Reporting and Data System (BI-RADS), and to identify the factors associated with technical failure of VBDM.

Materials and Methods

In this study, 1129 women aged 19-82 years who underwent mammography from December 2011 to January 2012 were included. Breast density evaluations by radiologists based on BI-RADS and by VBDM (Volpara Version 1.5.1) were compared. The agreement in interpreting breast density between radiologists and VBDM was determined based on four density grades (D1, D2, D3, and D4) and a binary classification of fatty (D1-2) vs. dense (D3-4) breast using kappa statistics. The association between technical failure of VBDM and patient age, total breast volume, fibroglandular tissue volume, history of partial mastectomy, the frequency of mass > 3 cm, and breast density was analyzed.

Results

The agreement between breast density evaluations by radiologists and VBDM was fair (k value = 0.26) when the four density grades (D1/D2/D3/D4) were used and moderate (k value = 0.47) for the binary classification (D1-2/D3-4). Twenty-seven women (2.4%) showed failure of VBDM. Small total breast volume, history of partial mastectomy, and high breast density were significantly associated with technical failure of VBDM (p = 0.001 to 0.015).

Conclusion

There is fair or moderate agreement in breast density evaluation between radiologists and VBDM. Technical failure of VBDM may be related to small total breast volume, a history of partial mastectomy, and high breast density.     

Breast Density Knowledge in a Screening Mammography Population Exposed to Density Notification

ObjectiveWomen are increasingly informed about their breast density due to state density reporting laws. However, accuracy of personal breast density knowledge remains unclear. We compared self-reported with clinically assessed breast density and assessed knowledge of density implications and feelings about future screening.MethodsFrom December 2017 to January 2020, we surveyed women aged 40 to 74 years without prior breast cancer, with a normal screening mammogram in the prior year, and ≥1 recorded breast density measures in four Breast Cancer Surveillance Consortium registries with density reporting laws. We measured agreement between self-reported and BI-RADS breast density categorized as “ever-dense” if heterogeneously or extremely dense within the past 5 years or “never-dense” otherwise, knowledge of dense breast implications, and feelings about future screening.ResultsSurvey participation was 28% (1,528 of 5,408), and 59% (896 of 1,528) of participants had ever-dense breasts. Concordance between self-report versus clinical density was 76% (677 of 896) among women with ever-dense breasts and 14% (89 of 632) among women with never-dense breasts, and 34% (217 of 632) with never-dense breasts reported being told they had dense breasts. Desire for supplemental screening was more frequent among those who reported having dense breasts 29% (256 of 893) or asked to imagine having dense breasts 30% (152 of 513) versus those reporting nondense breasts 15% (15 of 102) (P = .003, P = .002, respectively). Women with never-dense breasts had 6.3-fold higher odds (95% confidence interval:3.39-11.80) of accurate knowledge in states reporting density to all compared to states reporting only to women with dense breasts.DiscussionStandardized communications of breast density results to all women may increase density knowledge and are needed to support informed screening decisions.     

Correlation between breast density in mammography and background enhancement in MR mammography

Purpose

We aimed to analyse the influence of mammographic breast density on background enhancement (BE) at magnetic resonance (MR) mammography in preand postmenopausal women. In addition, we questioned predictability of contrast-enhancement dynamics of normal fibroglandular tissue (NFT) at MR mammography according to mammographic breast density.

Materials and methods

Twenty-six patients (mean age 51.54±11.5 years; range 37–79 years) who underwent both MR mammography and conventional mammography were included in this retrospective study. Fourteen patients were premenopausal and 12 were postmenopausal. The ethics committee of our institution approved the study. The mammograms were retrospectively reviewed for overall breast density according to the four-point scale (I–IV) of the Breast Imaging Reporting and Data System (BI-RADS) classification. Two radiologists, who were unaware of the clinical data, separately assessed the MR mammography images. Images were assessed for enhancement kinetic features (enhancement kinetic curve and the early-phase enhancement rate) and BE. MR mammography and conventional mammography findings were compared according to BI-RADS breast density category and menopausal status.

Results

Percentage of increased signal intensity values during the first minute did not change according to mammographic breast density, and the mean early-phase enhancement rate scores were similar among breast density groups (p=0.942). There was no significant difference between pre- and postmenopausal groups. Enhancement kinetic features of the different groups based on BI-RADS breast density category and menopausal status were similar. There was no correlation between breast density and BE in either premenopausal (p=0.211) or in postmenopausal (p=0.735) groups.

Conclusions

We determined no correlation between mammographic breast density and so-called BE in MR mammography in either premenopausal or postmenopausal women. NFT at MR mammography cannot be predicted on the basis of mammographic breast density.     

A dedicated BI-RADS training programme: effect on the inter-observer variation among screening radiologists

Introduction

The Breast Imaging Reporting and Data System (BI-RADS) was introduced in the Dutch breast cancer screening programme to improve communication between medical specialists. Following introduction, a substantial variation in the use of the BI-RADS lexicon for final assessment categories was noted among screening radiologists. We set up a dedicated training programme to reduce this variation. This study evaluates whether this programme was effective.

Materials and methods

Two comparable test sets were read before and after completion of the training programme. Each set contained 30 screening mammograms of referred women selected from screening practice. The sets were read by 25 experienced and 30 new screening radiologists. Cohen's kappa (κ) was used to calculate the inter-observer agreement. The BI-RADS 2003 version was implemented in the screening programme as the BI-RADS 2008 version requires the availability of diagnostic work-up, and this is unavailable.

Results

The inter-observer agreement of all participating radiologists (n = 55) with the expert panel increased from a pre-training κ-value of 0.44 to a post-training κ-value of 0.48 (p = 0.14). The inter-observer agreement of the new screening radiologists (n = 30) with the expert panel increased from κ = 0.41 to κ = 0.50 (p = 0.01), whereas there was no difference in agreement among the 25 experienced radiologists (from κ = 0.48 to κ = 0.46, p = 0.60).

Conclusion

Our training programme in the BI-RADS lexicon resulted in a significant improvement of agreement among new screening radiologists. Overall, the agreement among radiologists was moderate (guidelines Landis and Koch). This is in line with results found in the literature.     

Teleradiology with uncompressed digital mammograms: Clinical assessment

Purpose

The purpose of our study was to demonstrate the feasibility of sending uncompressed digital mammograms in a teleradiologic setting without loss of information by comparing image quality, lesion detection, and BI-RADS assessment.

Materials and methods

CDMAM phantoms were sent bidirectionally to two hospitals via the network. For the clinical aspect of the study, 200 patients were selected based on the BI-RAD system: 50% BI-RADS I and II; and 50% BI-RADS IV and V. Two hundred digital mammograms (800 views) were sent to two different institutions via a teleradiology network. Three readers evaluated those 200 mammography studies at institution 1 where the images originated, and in the two other institutions (institutions 2 and 3) where the images were sent. The readers assessed image quality, lesion detection, and BI-RADS classification.

Results

Automatic readout showed that CDMAM image quality was identical before and after transmission. The image quality of the 200 studies (total 600 mammograms) was rated as very good or good in 90–97% before and after transmission. Depending on the institution and the reader, only 2.5–9.5% of all studies were rated as poor. The congruence of the readers with respect to the final BI-RADS assessment ranged from 90% and 91% at institution 1 vs. institution 2, and from 86% to 92% at institution 1 vs. institution 3. The agreement was even higher for conformity of content (BI-RADS I or II and BI-RADS IV or V). Reader agreement in the three different institutions with regard to the detection of masses and calcifications, as well as BI-RADS classification, was very good (κ: 0.775–0.884). Results for interreader agreement were similar.

Conclusion

Uncompressed digital mammograms can be transmitted to different institutions with different workstations, without loss of information. The transmission process does not significantly influence image quality, lesion detection, or BI-RADS rating.     

Trends of Supplemental Screening in Women With Dense Breasts

PurposeThe objective of this study was to survey current trends in supplemental screening across various practice types and to understand factors that affect these practice patterns.MethodsIn this institutional review board–exempt study, a 12-question survey was sent out to ACR lead interpreting physicians. The survey inquired about practice features and the utilization of supplemental screening.ResultsA total of 902 of 4,688 lead interpreting breast imaging physicians (19.2%) responded to our survey. Of those respondents, 617 respondents (68.4%) worked in practices that offered supplemental breast cancer screening. Screening ultrasound was the most commonly utilized supplemental screening modality (53%). There was variability in methods of referral for supplemental screening, with referral through the ordering provider (56%) being the most common. Academic practices, private practices with breast specialization, and practices in the Northeast were more likely to provide supplemental screening (P < .05). There were significant relationships between the presence of state density notification legislation, the number of breast imaging trained radiologists, and the volume of mammographic studies performed per day and the availability of supplemental screening (P < .05). The use of automated breast density assessment software and patient education brochures about density and supplemental screening also had significant relationships with the availability of supplemental screening (P < .05).ConclusionsThe majority of practices surveyed are utilizing supplemental screening, but there is significant variability in the modalities used and the methods of referral. There are practice features that correlate with the availability of supplemental screening, and understanding these features provides further insight into current trends in supplemental screening utilization.     

Breast screening: What can the interval cancer review teach us? Are we perhaps being a bit too hard on ourselves?

PurposeThe aim of this study was to determine the features that make interval cancers apparent on the preceding screening mammogram and determine whether changes in the ways of performing the interval cancer review will affect the true interval cancer rate.Materials and methodsThis study was approved by the clinical governance committee. Mammograms of women diagnosed with an interval cancer were included in the study if they had been allocated to either the “suspicious signs” group or “subtle signs” group, during the historic interval cancer review. Three radiologists, individually and blinded to the site of interval cancer, reviewed the mammograms and documented the presence, site, characteristics and classification of any abnormality. Findings were compared with the appearances of the abnormality at the site of subsequent cancer development by a different breast radiologist. The chi-squared test was used in the analysis of the results, seeking associations between recall concordance and cancer mammographic or histological characteristics.Results111/590 interval cancers fulfilled the study inclusion criteria.In 17% of the cases none of the readers identified the relevant abnormality on the screening mammogram. 1/3 readers identified the relevant lesion in 22% of the cases, 2/3 readers in 28% of cases and all 3 readers in 33% of cases. The commonest unanimously recalled abnormality was microcalcification and the most challenging mammographic abnormality to detect was asymmetric density.We did not find any statistically significant association between recall concordance and time to interval cancer, position of lesion in the breast, breast density or cancer grade.ConclusionEven the simple step of performing an independent blinded review of interval cancers reduces the rate of interval cancers classified as missed by up to 39%.     

Characterization of common breast MRI abnormalities: comparison between abbreviated and full MRI protocols

Rationale and objectivesTo evaluate the diagnostic performance of abbreviated MRI (AB-MRI) in comparison to a full protocol MRI (FP-MRI) when evaluating common MRI abnormalities of a mass, non-mass enhancement and focus.Materials and methodsThis retrospective reader study was Institutional Review Board approved and Health Insurance Portability and Accountability Act (HIPAA) compliant. AB-MRIs were reviewed from May 2018–December 2019 to identify women with an abnormal AB-MRI, FP-MRI within six months of the AB-MRI and an elevated risk for breast cancer. Six breast radiologists initially interpreted and recorded findings from the AB-MRI. Immediately after reviewing the AB-MRI, the same radiologists interpreted and recorded findings from the FP-MRI. Findings were recorded in an electronic data collection form. Cohen's Kappa test was used to calculate agreement. P < 0.05 was considered statistically significant.ResultsOf 119 patients who had an AB-MRI, our final study comprised of 32 patients who had 64 breast MRIs (32 AB-MRI and 32 FP-MRI). The amount of fibroglandular tissue for AB-MRI and FP-MRI showed excellent intra-reader agreement [Kappa: 0.89–1.00 (P < 0.0001)]. Substantial to excellent intra-reader agreement [Kappa: 0.74–0.93 (P < 0.0001)] was demonstrated for all 6 readers when identifying abnormalities seen on AB-MRI and FP-MRI. Moderate to excellent intra-reader agreement [Kappa: 0.41–0.87(P < 0.0001)] was demonstrated between the AB-MRI and FP-MRI for the final BI-RADS assessment.ConclusionAB-MRI has acceptable intra-reader agreement with FP-MRI when characterizing common MRI abnormalities such as a mass, non-mass enhancement and focus suggesting that subsequent FP-MRI may not be needed.