Inter- and intra-observer agreement of BI-RADS-based subjective visual estimation of amount of fibroglandular breast tissue with magnetic resonance imaging: comparison to automated quantitative assessment

Automated Breast Density Assessment in MRI Using Deep Learning and Radiomics: Strategies for Reducing Inter-Observer Variability

BACKGROUND

Accurate breast density evaluation allows for more precise risk estimation but suffers from high inter-observer variability.

PURPOSE

To evaluate the feasibility of reducing inter-observer variability of breast density assessment through artificial intelligence (AI) assisted interpretation.

STUDY TYPE

Retrospective.

POPULATION

Six hundred and twenty-one patients without breast prosthesis or reconstructions were randomly divided into training (N = 377), validation (N = 98), and independent test (N = 146) datasets.

FIELD STRENGTH/SEQUENCE

1.5 T and 3.0 T; T1-weighted spectral attenuated inversion recovery.

ASSESSMENT

Five radiologists independently assessed each scan in the independent test set to establish the inter-observer variability baseline and to reach a reference standard. Deep learning and three radiomics models were developed for three classification tasks: (i) four Breast Imaging-Reporting and Data System (BI-RADS) breast composition categories (A-D), (ii) dense (categories C, D) vs. non-dense (categories A, B), and (iii) extremely dense (category D) vs. moderately dense (categories A-C). The models were tested against the reference standard on the independent test set. AI-assisted interpretation was performed by majority voting between the models and each radiologist's assessment.

STATISTICAL TESTS

Inter-observer variability was assessed using linear-weighted kappa (κ) statistics. Kappa statistics, accuracy, and area under the receiver operating characteristic curve (AUC) were used to assess models against reference standard.

RESULTS

In the independent test set, five readers showed an overall substantial agreement on tasks (i) and (ii), but moderate agreement for task (iii). The best-performing model showed substantial agreement with reference standard for tasks (i) and (ii), but moderate agreement for task (iii). With the assistance of the AI models, almost perfect inter-observer variability was obtained for tasks (i) (mean κ = 0.86), (ii) (mean κ = 0.94), and (iii) (mean κ = 0.94).

DATA CONCLUSION

Deep learning and radiomics models have the potential to help reduce inter-observer variability of breast density assessment.

LEVEL OF EVIDENCE

3 TECHNICAL EFFICACY: Stage 1.

[WITHDRAWN] Predicting breast cancer with AI for individual risk-adjusted MRI screening and early detection

This paper has been withdrawn by Lukas Hirsch. Major revisions and rewriting in progress.

Robustness of radiomics to variations in segmentation methods in multimodal brain MRI

Radiomics in neuroimaging uses fully automatic segmentation to delineate the anatomical areas for which radiomic features are computed. However, differences among these segmentation methods affect radiomic features to an unknown extent. A scan-rescan dataset (n = 46) of T1-weighted and diffusion tensor images was used. Subjects were split into a sleep-deprivation and a control group. Scans were segmented using four segmentation methods from which radiomic features were computed. First, we measured segmentation agreement using the Dice-coefficient. Second, robustness and reproducibility of radiomic features were measured using the intraclass correlation coefficient (ICC). Last, difference in predictive power was assessed using the Friedman-test on performance in a radiomics-based sleep deprivation classification application. Segmentation agreement was generally high (interquartile range = 0.77–0.90) and median feature robustness to segmentation method variation was higher (ICC > 0.7) than scan-rescan reproducibility (ICC 0.3–0.8). However, classification performance differed significantly among segmentation methods (p < 0.001) ranging from 77 to 84%. Accuracy was higher for more recent deep learning-based segmentation methods. Despite high agreement among segmentation methods, subtle differences significantly affected radiomic features and their predictive power. Consequently, the effect of differences in segmentation methods should be taken into account when designing and evaluating radiomics-based research methods.

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.

The potential of predictive and prognostic breast MRI (P2-bMRI)

Magnetic resonance imaging (MRI) is an important part of breast cancer diagnosis and multimodal workup. It provides unsurpassed soft tissue contrast to analyse the underlying pathophysiology, and it is adopted for a variety of clinical indications. Predictive and prognostic breast MRI (P2-bMRI) is an emerging application next to these indications. The general objective of P2-bMRI is to provide predictive and/or prognostic biomarkers in order to support personalisation of breast cancer treatment. We believe P2-bMRI has a great clinical potential, thanks to the in vivo examination of the whole tumour and of the surrounding tissue, establishing a link between pathophysiology and response to therapy (prediction) as well as patient outcome (prognostication). The tools used for P2-bMRI cover a wide spectrum: standard and advanced multiparametric pulse sequences; structured reporting criteria (for instance BI-RADS descriptors); artificial intelligence methods, including machine learning (with emphasis on radiomics data analysis); and deep learning that have shown compelling potential for this purpose. P2-bMRI reuses the imaging data of examinations performed in the current practice. Accordingly, P2-bMRI could optimise clinical workflow, enabling cost savings and ultimately improving personalisation of treatment. This review introduces the concept of P2-bMRI, focusing on the clinical application of P2-bMRI by using semantic criteria.

Magnetic resonance classification proposal for fibrous capsules in breast silicone implants

OBJECTIVE

To propose a new classification system for evaluating fibrous capsules around silicone implants using magnetic resonance.

METHODS

We retrospectively evaluated 90 consecutive patients who underwent breast MRI scans at a single center from February to March 2022. All patients with silicone implants and contrast dynamic sequences were included. Two radiologists classified the fibrous capsules according to the proposed classification in four grades. Interobserver variability was calculated for the final score. For comparison purposes, the inter-rater agreement of background parenchymal enhancement and the amount of fibroglandular tissue were also calculated.

RESULTS

Reader 1 classified 2 (2.2) fibrous capsules as grade 1, 7 (7.8) as grade 2, 18 (20.0) as grade 3, and 63 (70.0) as grade 4, whereas reader 2 classified 1 (1.1), 9 (10.0), 24 (26.7), and 56 (62.2) fibrous capsules, respectively, for each grade. The interobserver agreement for fibrous capsule classification was moderate (ĸ = 0.65). The inter-rater agreement of background parenchymal enhancement and amount of fibroglandular tissue were fair: ĸ = 0.50 and ĸ = 0.44, respectively.

CONCLUSION

Our study proposes classifying FC by MRI in patients with SI to standardize the description and classification of the findings with good interobserver agreement.

Quantitative Mammographic Density Measurements and Molecular Subtypes in Chinese Women With Breast Cancer

Background

Studies investigating associations between mammographic density (MD) and breast cancer subtypes have generated mixed results. We previously showed that having extremely dense breasts was associated with the human epidermal growth factor receptor-2 (HER2)-enriched subtype in Chinese breast cancer patients.

Methods

In this study, we reevaluated the MD-subtype association in 1549 Chinese breast cancer patients, using VolparaDensity software to obtain quantitative MD measures. All statistical tests were 2-sided.

Results

Compared with women with luminal A tumors, women with luminal B/HER2- (odds ratio [OR] = 1.20, 95% confidence interval [CI] = 1.04 to 1.38; P = .01), luminal B/HER2+ (OR = 1.22, 95% CI = 1.03 to 1.46; P = .03), and HER2-enriched tumors (OR = 1.30, 95% CI = 1.06 to 1.59; P = .01) had higher fibroglandular dense volume. These associations were stronger in patients with smaller tumors (<2 cm). In contrast, the triple-negative subtype was associated with lower nondense volume (OR = 0.82, 95% CI = 0.68 to 0.99; P = .04), and the association was only seen among older women (age 50 years or older).

Conclusion

Although biological mechanisms remain to be investigated, the associations for the HER2-enriched and luminal B subtypes with increasing MD may partially explain the higher prevalence of luminal B and HER2+ breast cancers previously reported in Asian women.

MRI background parenchymal enhancement, fibroglandular tissue, and mammographic breast density in patients with invasive lobular breast cancer on adjuvant endocrine hormonal treatment: associations with survival

BACKGROUND

To investigate if baseline and/or changes in contralateral background parenchymal enhancement (BPE) and fibroglandular tissue (FGT) measured on magnetic resonance imaging (MRI) and mammographic breast density (MD) can be used as imaging biomarkers for overall and recurrence-free survival in patients with invasive lobular carcinomas (ILCs) undergoing adjuvant endocrine treatment.

METHODS

Women who fulfilled the following inclusion criteria were included in this retrospective HIPAA-compliant IRB-approved study: unilateral ILC, pre-treatment breast MRI and/or mammography from 2000 to 2010, adjuvant endocrine treatment, follow-up MRI, and/or mammography 1-2 years after treatment onset. BPE, FGT, and mammographic MD of the contralateral breast were independently graded by four dedicated breast radiologists according to BI-RADS. Associations between the baseline levels and change in levels of BPE, FGT, and MD with overall survival and recurrence-free survival were assessed using Kaplan-Meier survival curves and Cox regression analysis.

RESULTS

Two hundred ninety-eight patients (average age = 54.1 years, range = 31-79) fulfilled the inclusion criteria. The average follow-up duration was 11.8 years (range = 2-19). Baseline and change in levels of BPE, FGT, and MD were not significantly associated with recurrence-free or overall survival. Recurrence-free and overall survival were affected by histological subtype (p < 0.0001), number of metastatic axillary lymph nodes (p < 0.0001), age (p = 0.01), and adjuvant endocrine treatment duration (p < 0.001).

CONCLUSIONS

Qualitative evaluation of BPE, FGT, and mammographic MD changes cannot predict which patients are more likely to benefit from adjuvant endocrine treatment.

Quantitative Volumetric K-Means Cluster Segmentation of Fibroglandular Tissue and Skin in Breast MRI

Mammographic breast density (MBD) is the most commonly used method to assess the volume of fibroglandular tissue (FGT). However, MRI could provide a clinically feasible and more accurate alternative. There were three aims in this study: (1) to evaluate a clinically feasible method to quantify FGT with MRI, (2) to assess the inter-rater agreement of MRI-based volumetric measurements and (3) to compare them to measurements acquired using digital mammography and 3D tomosynthesis. This retrospective study examined 72 women (mean age 52.4 ± 12.3 years) with 105 disease-free breasts undergoing diagnostic 3.0-T breast MRI and either digital mammography or tomosynthesis. Two observers analyzed MRI images for breast and FGT volumes and FGT-% from T1-weighted images (0.7-, 2.0-, and 4.0-mm-thick slices) using K-means clustering, data from histogram, and active contour algorithms. Reference values were obtained with Quantra software. Inter-rater agreement for MRI measurements made with 2-mm-thick slices was excellent: for FGT-%, r = 0.994 (95% CI 0.990-0.997); for breast volume, r = 0.985 (95% CI 0.934-0.994); and for FGT volume, r = 0.979 (95% CI 0.958-0.989). MRI-based FGT-% correlated strongly with MBD in mammography (r = 0.819-0.904, P < 0.001) and moderately to high with MBD in tomosynthesis (r = 0.630-0.738, P < 0.001). K-means clustering-based assessments of the proportion of the fibroglandular tissue in the breast at MRI are highly reproducible. In the future, quantitative assessment of FGT-% to complement visual estimation of FGT should be performed on a more regular basis as it provides a component which can be incorporated into the individual's breast cancer risk stratification.

Multimodality Imaging of Breast Parenchymal Density and Correlation with Risk Assessment

Purpose of Review

Breast density, or the amount of fibroglandular tissue in the breast, has become a recognized and independent marker for breast cancer risk. Public awareness of breast density as a possible risk factor for breast cancer has resulted in legislation for risk stratification purposes in many US states. This review will provide a comprehensive overview of the currently available imaging modalities for qualitative and quantitative breast density assessment and the current evidence on breast density and breast cancer risk assessment.

Recent Findings

To date, breast density assessment is mainly performed with mammography and to some extent with magnetic resonance imaging. Data indicate that computerized, quantitative techniques in comparison with subjective visual estimations are characterized by higher reproducibility and robustness.

Summary

Breast density reduces the sensitivity of mammography due to a masking effect and is also a recognized independent risk factor for breast cancer. Standardized breast density assessment using automated volumetric quantitative methods has the potential to be used for risk prediction and stratification and in determining the best screening plan for each woman.

Density and tailored breast cancer screening: practice and prediction - an overview

Mammography, as the primary screening modality, has facilitated a substantial decrease in breast cancer-related mortality in the general population. However, the sensitivity of mammography for breast cancer detection is decreased in women with higher breast densities, which is an independent risk factor for breast cancer. With increasing public awareness of the implications of a high breast density, there is an increasing demand for supplemental screening in these patients. Yet, improvements in breast cancer detection with supplemental screening methods come at the expense of increased false-positives, recall rates, patient anxiety, and costs. Therefore, breast cancer screening practice must change from a general one-size-fits-all approach to a more personalized, risk-based one that is tailored to the individual woman’s risk, personal beliefs, and preferences, while accounting for cost, potential harm, and benefits.

This overview will provide an overview of the available breast density assessment modalities, the current breast density screening recommendations for women at average risk of breast cancer, and supplemental methods for breast cancer screening. In addition, we will provide a look at the possibilities for a risk-adapted breast cancer screening.

Breast cancer MRI radiomics: An overview of algorithmic features and impact of inter-reader variability in annotating tumors

PURPOSE

To review features used in MRI radiomics of breast cancer and study the inter-reader stability of the features.

METHODS

We implemented 529 algorithmic features that can be extracted from tumor and fibroglandular tissue (FGT) in breast MRIs. The features were identified based on a review of the existing literature with consideration of their usage, prognostic ability, and uniqueness. The set was then extended so that it comprehensively describes breast cancer imaging characteristics. The features were classified into 10 groups based on the type of data used to extract them and the type of calculation being performed. For the assessment of inter-reader variability, four fellowship-trained readers annotated tumors on preoperative dynamic contrast-enhanced MRIs for 50 breast cancer patients. Based on the annotations, an algorithm automatically segmented the image and extracted all features resulting in one set of features for each reader. For a given feature, the inter-reader stability was defined as the intraclass correlation coefficient (ICC) computed using the feature values obtained through all readers for all cases.

RESULTS

The average inter-reader stability for all features was 0.8474 (95% CI: 0.8068-0.8858). The mean inter-reader stability was lower for tumor-based features (0.6348, 95% CI: 0.5391-0.7257) than FGT-based features (0.9984, 95% CI: 0.9970-0.9992). The feature group with the highest inter-reader stability quantifies breast and FGT volume. The feature group with the lowest inter-reader stability quantifies variations in tumor enhancement.

CONCLUSIONS

Breast MRI radiomics features widely vary in terms of their stability in the presence of inter-reader variability. Appropriate measures need to be taken for reducing this variability in tumor-based radiomics.

MRI-based quantification of residual fibroglandular tissue of the breast after conservative mastectomies

PURPOSE

Skin-sparing and nipple-sparing mastectomies (SSM; NSM) remove the breast's fibroglandular tissue (FGT), thereby reducing breast cancer risk. The postoperative presence of residual FGT (RFGT) is associated with remaining cancer risk. This study evaluated the role of MRI in the quantitative assessment of RFGT and its impact on the estimation of the remaining breast cancer risk.

METHODS

The postoperative MRI scans (following EUSOMA recommendations) of 58 patients who had undergone SSM or NSM between 2003 and 2013, as well as preoperative MRI scans that were available in 25 of these patients, were retrospectively evaluated for the presence and location of RFGT by three radiologists. Two different observers quantitatively assessed the volume and percentage of retromamillary and other RFGT (RFGT_rm and RFGT_other) were assessed. The Fisher's exact test, the Student's t-test, and intraclass coherence were used to compare patient groups and to assess reproducibility.

RESULTS

RFGT was found in 20% of all breasts and significantly more frequently after NSM than SSM (50% vs. 13%, p = .003). RFGT_rm and RFGT_other were more prevalent after NSM (p < 0.001; p = .127). RFGT ranged from 0.5 to 26% of the preoperative FGT, with higher percentages after NSM than SSM (p = .181).

CONCLUSIONS

The prevalence and percentage of RFGT found on MRI indicate a considerable remaining postoperative breast cancer risk in some women.

Quantitative assessment of background parenchymal enhancement in breast magnetic resonance images predicts the risk of breast cancer

The objective of this study was to evaluate the association betweenthe quantitative assessment of background parenchymal enhancement rate (BPER) and breast cancer. From 14,033 consecutive patients who underwent breast MRI in our center, we randomly selected 101 normal controls. Then, we selected 101 women with benign breast lesions and 101 women with breast cancer who were matched for age and menstruation status. We evaluated BPER at early (2 minutes), medium (4 minutes) and late (6 minutes) enhanced time phases of breast MRI for quantitative assessment. Odds ratios (ORs) for risk of breast cancer were calculated using the receiver operating curve. The BPER increased in a time-dependent manner after enhancement in both premenopausal and postmenopausal women. Premenopausal women had higher BPER than postmenopausal women at early, medium and late enhanced phases. In the normal population, the OR for probability of breast cancer for premenopausal women with high BPER was 4.1 (95% CI: 1.7–9.7) and 4.6 (95% CI: 1.7–12.0) for postmenopausal women. The OR of breast cancer morbidity in premenopausal women with high BPER was 2.6 (95% CI: 1.1–6.4) and 2.8 (95% CI: 1.2–6.1) for postmenopausal women. The BPER was found to be a predictive factor of breast cancer morbidity. Different time phases should be used to assess BPER in premenopausal and postmenopausal women.

Accuracy of fully automated, quantitative, volumetric measurement of the amount of fibroglandular breast tissue using MRI: correlation with anthropomorphic breast phantoms

To demonstrate the accuracy of fully automated, quantitative, volumetric measurement of the amount of fibroglandular breast tissue (FGT), using MRI, and to investigate the impact of different MRI sequences using anthropomorphic breast phantoms as the ground truth. In this study, 10 anthropomorphic breast phantoms that consisted of different known fractions of adipose and protein tissue, which closely resembled normal breast parenchyma, were developed. Anthropomorphic breast phantoms were imaged with a 1.5 T unit (Siemens, Avantofit) using an 18-channel breast coil. The sequence protocol consisted of an isotropic Dixon sequence (Di), an anisotropic Dixon sequence (Da), and T₁ 3D FLASH sequences with and without fat saturation (T1). Fully automated, quantitative, volumetric measurement of FGT for all anthropomorphic phantoms and sequences was performed and correlated with the amounts of fatty and protein components in the phantoms as the ground truth. Fully automated, quantitative, volumetric measurements of FGT with MRI for all sequences ranged from 5.86 to 61.05% (mean 33.36%). The isotropic Dixon sequence yielded the highest accuracy (median 0.51%-0.78%) and precision (median range 0.19%) compared with anisotropic Dixon (median 1.92%-2.09%; median range 0.55%) and T₁ -weighted sequences (median 2.54%-2.46%; median range 0.82%). All sequences yielded good correlation with the FGT content of the anthropomorphic phantoms. The best correlation of FGT measurements was identified for Dixon sequences (Di, R²  = 0.999; Da, R²  = 0.998) compared with conventional T₁ -weighted sequences (R²  = 0.971). MRI yields accurate, fully automated, quantitative, volumetric measurements of FGT, an increasingly important and sensitive imaging biomarker for breast cancer risk. Compared with conventional T₁ sequences, Dixon-type sequences show the highest correlation and reproducibility for automated, quantitative, volumetric FGT measurements using anthropomorphic breast phantoms as the ground truth.

Breast density estimation from high spectral and spatial resolution MRI

A three-dimensional breast density estimation method is presented for high spectral and spatial resolution (HiSS) MR imaging. Twenty-two patients were recruited (under an Institutional Review Board--approved Health Insurance Portability and Accountability Act-compliant protocol) for high-risk breast cancer screening. Each patient received standard-of-care clinical digital x-ray mammograms and MR scans, as well as HiSS scans. The algorithm for breast density estimation includes breast mask generating, breast skin removal, and breast percentage density calculation. The inter- and intra-user variabilities of the HiSS-based density estimation were determined using correlation analysis and limits of agreement. Correlation analysis was also performed between the HiSS-based density estimation and radiologists' breast imaging-reporting and data system (BI-RADS) density ratings. A correlation coefficient of 0.91 ([Formula: see text]) was obtained between left and right breast density estimations. An interclass correlation coefficient of 0.99 ([Formula: see text]) indicated high reliability for the inter-user variability of the HiSS-based breast density estimations. A moderate correlation coefficient of 0.55 ([Formula: see text]) was observed between HiSS-based breast density estimations and radiologists' BI-RADS. In summary, an objective density estimation method using HiSS spectral data from breast MRI was developed. The high reproducibility with low inter- and low intra-user variabilities shown in this preliminary study suggest that such a HiSS-based density metric may be potentially beneficial in programs requiring breast density such as in breast cancer risk assessment and monitoring effects of therapy.