Multiview deep learning networks based on automated breast volume scanner images for identifying breast cancer in BI-RADS 4.

Objectives: To develop and validate a deep learning (DL) based automatic segmentation and classification system to classify benign and malignant BI-RADS 4 lesions imaged with ABVS. Methods: From May to December 2020, patients with BI-RADS 4 lesions from Centre 1 and Centre 2 were retrospectively enrolled and divided into a training set (Centre 1) and an independent test set (Centre 2). All included patients underwent an ABVS examination within one week before the biopsy. A two-stage DL framework consisting of an automatic segmentation module and an automatic classification module was developed. The preprocessed ABVS images were input into the segmentation module for BI-RADS 4 lesion segmentation. The classification model was constructed to extract features and output the probability of malignancy. The diagnostic performances among different ABVS views (axial, sagittal, coronal, and multi-view) and DL architectures (Inception-v3, ResNet 50, and MobileNet) were compared. Results: A total of 251 BI-RADS 4 lesions from 216 patients were included (178 in the training set and 73 in the independent test set). The average Dice coefficient, precision, and recall of the segmentation module in the test set were 0.817 ± 0.142, 0.903 ± 0.183, and 0.886 ± 0.187, respectively. The DL model based on multiview ABVS images and Inception-v3 achieved the best performance, with an AUC, sensitivity, specificity, PPV, and NPV of 0.949 (95% CI: 0.945-0.953), 82.14%, 95.56%, 92.00%, and 89.58%, respectively, in the test set. Conclusions: The developed multiview DL model enables automatic segmentation and classification of BI-RADS 4 lesions in ABVS images.

Comparison of Visual and Quantra Software Mammographic Density Assessment According to BI-RADS® in 2D and 3D Images.

Mammographic density (MD) assessment is subject to inter- and intra-observer variability. An automated method, such as Quantra software, could be a useful tool for an objective and reproducible MD assessment. Our purpose was to evaluate the performance of Quantra software in assessing MD, according to BI-RADS® Atlas Fifth Edition recommendations, verifying the degree of agreement with the gold standard, given by the consensus of two breast radiologists. A total of 5009 screening examinations were evaluated by two radiologists and analysed by Quantra software to assess MD. The agreement between the three assigned values was expressed as intraclass correlation coefficients (ICCs). The agreement between the software and the two readers (R1 and R2) was moderate with ICC values of 0.725 and 0.713, respectively. A better agreement was demonstrated between the software's assessment and the average score of the values assigned by the two radiologists, with an index of 0.793, which reflects a good correlation. Quantra software appears a promising tool in supporting radiologists in the MD assessment and could be part of a personalised screening protocol soon. However, some fine-tuning is needed to improve its accuracy, reduce its tendency to overestimate, and ensure it excludes high-density structures from its assessment.

The Role of Diffusion-Weighted Imaging in Characterizing Benign and Malignant Breast Lesions: A Retrospective Study.

Introduction  Diffusion-weighted imaging (DWI) is a promising magnetic resonance imaging (MRI) technique for differentiating between benign and malignant breast lesions. This study set out to assess the diagnostic utility of DWI and apparent diffusion coefficient (ADC) values in the characterization of breast lesions. Materials and methods A retrospective analysis comprised 30 patients with breast lesions who had breast MRI with DWI. The histopathological findings, ADC readings, and conventional MRI features were all analyzed. The receiver operating characteristic (ROC) curve analysis method was utilized to assess the diagnostic accuracy of DWI. Results Out of the 30 lesions, 22 (73.3%) were benign and eight (26.7%) were malignant. Malignant lesions exhibited significantly lower ADC values (p < 0.001) compared to benign lesions. An ADC cutoff value of 1.1 × 10-3 mm2/s was optimal for differentiating benign from malignant lesions, yielding 90.81% sensitivity, 91.51% specificity, and 91.5% accuracy. Conclusion Combining DWI with quantitative ADC analysis is a helpful, non-invasive method for the characterization of breast lesions. It shows excellent diagnostic accuracy in identifying benign and malignant lesions, which may cut down on pointless biopsies and help with patient management.

Characterization of Indeterminate Breast Lesions Based on Pressure Estimates by Noninvasive 3D Contrast-Enhanced Ultrasound.

OBJECTIVE: To assess the ability of the pressure gradient between breast lesions and adjacent normal tissue estimated by 3D subharmonic-aided pressure estimation (SHAPE) to characterize indeterminate breast lesions. METHODS: This prospective study enrolled patients scheduled for ultrasound-guided needle biopsies of a breast lesion. Before the biopsy, 3D SHAPE data were collected from the breast lesion during the infusion of an ultrasound contrast agent (Definity) as well as after clearance of the agent. Direct, invasive pressure measurements in the lesion and adjacent normal tissue were then obtained using an intracompartmental pressure monitoring system (C2DX) before tissue sampling as part of the biopsy procedure. The mean SHAPE gradient and invasive measurement gradient between the lesion and adjacent normal tissue were compared to the biopsy results. The SHAPE gradients were also compared to the invasive pressure gradients. RESULTS: There were 8 malignant and 13 benign lesions studied. The SHAPE gradients and invasive pressure gradients were significantly different between the benign and malignant lesions (2.86 ± 3.24 vs. -0.03 ± 1.72 a.u.; p = 0.03 and 9.9 ± 8.5 vs. 20.9 ± 8.0 mmHg; p = 0.008, respectively). The area under the curves, specificities, and sensitivities for detecting malignancy by SHAPE gradients and invasive pressure gradients were 0.79 and 0.88, 77% and 92%, and 88% and 50%, respectively. A weak negative correlation was found between the SHAPE and invasive pressure gradients (r = -0.2). CONCLUSION: The pressure gradient between a breast lesion and adjacent normal tissue estimated by 3D SHAPE shows potential for characterizing indeterminate breast lesions.

Breast Density Status Changes: Frequency, Sequence, and Practice Implications.

OBJECTIVE: Changes in a patient's reported breast density status (dense vs nondense) trigger modifications in their cancer risk profile and supplemental screening recommendations. This study tracked the frequency and longitudinal sequence of breast density status changes among patients who received serial mammograms. METHODS: This IRB-approved, HIPAA-compliant retrospective cohort study tracked breast density changes among patients who received at least 2 mammograms over an 8-year study period. BI-RADS density assessment categories A through D, visually determined at the time of screening, were abstracted from electronic medical records and dichotomized into either nondense (categories A or B) or dense (categories C or D) status. A sequence analysis of longitudinal changes in density status was performed using Microsoft SQL. RESULTS: A total of 58 895 patients underwent 231 997 screening mammograms. Most patients maintained the same BI-RADS density category A through D (87.35% [51 444/58 895]) and density status (93.35% [54 978/58 859]) throughout the study period. Among patients whose density status changed, the majority (97% [3800/3917]) had either scattered or heterogeneously dense tissue, and over half (57% [2235/3917]) alternated between dense and nondense status multiple times. CONCLUSION: Our results suggest that many cases of density status change may be attributable to intra- and interradiologist variability rather than to true underlying changes in density. These results lend support to consideration of automated density assessment because breast density status changes can significantly impact cancer risk assessment and supplemental screening recommendations.

Evaluating text and visual diagnostic capabilities of large language models on questions related to the Breast Imaging Reporting and Data System Atlas 5th edition.

PURPOSE: This study aimed to evaluate the performance of large language models (LLMs) and multimodal LLMs in interpreting the Breast Imaging Reporting and Data System (BI-RADS) categories and providing clinical management recommendations for breast radiology in text-based and visual questions. METHODS: This cross-sectional observational study involved two steps. In the first step, we compared ten LLMs (namely ChatGPT 4o, ChatGPT 4, ChatGPT 3.5, Google Gemini 1.5 Pro, Google Gemini 1.0, Microsoft Copilot, Perplexity, Claude 3.5 Sonnet, Claude 3 Opus, and Claude 3 Opus 200K), general radiologists, and a breast radiologist using 100 text-based multiple-choice questions (MCQs) related to the BI-RADS Atlas 5th edition. In the second step, we assessed the performance of five multimodal LLMs (ChatGPT 4o, ChatGPT 4V, Claude 3.5 Sonnet, Claude 3 Opus, and Google Gemini 1.5 Pro) in assigning BI-RADS categories and providing clinical management recommendations on 100 breast ultrasound images. The comparison of correct answers and accuracy by question types was analyzed using McNemar's and chi-squared tests. Management scores were analyzed using the Kruskal- Wallis and Wilcoxon tests. RESULTS: Claude 3.5 Sonnet achieved the highest accuracy in text-based MCQs (90%), followed by ChatGPT 4o (89%), outperforming all other LLMs and general radiologists (78% and 76%) (P < 0.05), except for the Claude 3 Opus models and the breast radiologist (82%) (P > 0.05). Lower-performing LLMs included Google Gemini 1.0 (61%) and ChatGPT 3.5 (60%). Performance across different categories of showed no significant variation among LLMs or radiologists (P > 0.05). For breast ultrasound images, Claude 3.5 Sonnet achieved 59% accuracy, significantly higher than other multimodal LLMs (P < 0.05). Management recommendations were evaluated using a 3-point Likert scale, with Claude 3.5 Sonnet scoring the highest (mean: 2.12 ± 0.97) (P < 0.05). Accuracy varied significantly across BI-RADS categories, except Claude 3 Opus (P < 0.05). Gemini 1.5 Pro failed to answer any BI-RADS 5 questions correctly. Similarly, ChatGPT 4V failed to answer any BI-RADS 1 questions correctly, making them the least accurate in these categories (P < 0.05). CONCLUSION: Although LLMs such as Claude 3.5 Sonnet and ChatGPT 4o show promise in text-based BI-RADS assessments, their limitations in visual diagnostics suggest they should be used cautiously and under radiologists' supervision to avoid misdiagnoses. CLINICAL SIGNIFICANCE: This study demonstrates that while LLMs exhibit strong capabilities in text-based BI-RADS assessments, their visual diagnostic abilities are currently limited, necessitating further development and cautious application in clinical practice.

Radio-pathological characteristics of primary neuroendocrine breast carcinoma: Series of 4 cases.

Neuroendocrine breast cancers (NEBCs) are a rare and distinct subtype of breast tumors, characterized by their neuroendocrine differentiation. Despite accounting for less than 1% of all breast cancers, NEBCs present unique diagnostic and therapeutic challenges due to their heterogeneous nature and variable prognosis. Accurate imaging plays a crucial role in the diagnosis, treatment planning, and follow-up of NEBCs, yet remains a complex area due to the rarity of these tumors and overlapping features with more common breast cancers. We present a series of 4 cases of primary NEBC, emphasizing the imaging features and their histopathological correlations. All patients presented with breast lump. Diagnostic Mammography followed by Ultrasound was performed in each case. All 4 cases were categorized as Breast Imaging- Reporting and Data System (BI-RADS)-4. Trucut biopsy was performed and histopathological analysis revealed the diagnosis of NEBC. Patients underwent Surgery followed by Chemotherapy, Hormonal Therapy or Radiation therapy alone or in combination with each other depending upon the histopathological characteristics.

Downgrading Breast Imaging Reporting and Data System categories in ultrasound using strain elastography and computer-aided diagnosis system: a multicenter, prospective study.

OBJECTIVE: To determine whether adding elastography strain ratio (SR) and a deep learning based computer-aided diagnosis (CAD) system to breast ultrasound (US) can help reclassify Breast Imaging Reporting and Data System (BI-RADS) 3 and 4a-c categories and avoid unnecessary biopsies. METHODS: This prospective, multicentre study included 1049 masses (691 benign, 358 malignant) with assigned BI-RADS 3 and 4a-c between 2020 and 2022. CAD results was dichotomized possibly malignant vs. benign. All patients underwent SR and CAD examinations and histopathological findings were the standard of reference. Reduction of unnecessary biopsies (biopsies in benign lesions) and missed malignancies after reclassified (new BI-RADS 3) with SR and CAD were the outcome measures. RESULTS: Following the routine conventional breast US assessment, 48.6% (336 of 691 masses) underwent unnecessary biopsies. After reclassifying BI-RADS 4a masses (SR cut-off <2.90, CAD dichotomized possibly benign), 25.62% (177 of 691 masses) underwent an unnecessary biopsies corresponding to a 50.14% (177 vs. 355) reduction of unnecessary biopsies. After reclassification, only 1.72% (9 of 523 masses) malignancies were missed in the new BI-RADS 3 group. CONCLUSION: Adding SR and CAD to clinical practice may show an optimal performance in reclassifying BI-RADS 4a to 3 categories, and 50.14% masses would be benefit by keeping the rate of undetected malignancies with an acceptable value of 1.72%. ADVANCES IN KNOWLEDGE: Leveraging the potential of SR in conjunction with CAD holds immense promise in substantially reducing the biopsy frequency associated with BI-RADS 3 and 4A lesions, thereby conferring substantial advantages upon patients encompassed within this cohort.

Breast multiparametric ultrasound: a single-center experience.

PURPOSE: To evaluate the role of multiparametric ultrasound (mpUS) in the characterization of focal breast lesions (FBLs). METHODS: This prospective study enrolled patients undergoing multiparametric breast ultrasound for FBLs. An experienced breast radiologist evaluated the following ultrasound features: US BI-RADS category, vascularization pattern (internal, vessels in rim and combined) and presence of penetrating vessels with each Doppler method (Color-Doppler, Power-Doppler, Microvascular imaging), strain ratio (SR) and Tsukuba score (TS) with Strain Elastography (SE), Emax, Emean, Emin and Eratio with 2D-shear wave elastography (2D-SWE). Core biopsy for all BI-RADS 4-5 FBLs and 24-month follow-up for all BI-RADS 2-3 FBLs were considered for standard of reference. The diagnostic performance was assessed with the area under curve (AUCs) and cut-off values were determined according to the Youden's index. RESULTS: A total of 139 FBLs were included with 75/139 (53.9%) benign and 64/139 (46.1%) malignant FBLs. Internal vascularization patterns (p < 0.001), penetrating vessels (p < 0.001), TS 4-5 (p < 0.001) and all 2D-SWE parameters (p < 0.001) were significantly different between benign and malignant FBLs. The BI-RADS score provided an AUC of 0.876 (95% CI 0.810-0.926) for the diagnosis of malignant FBLs. Among the 2D-SWE measurements, an excellent diagnostic performance was observed for Emax with an AUC of 0.915 (95% CI 0.856-0.956) and Emean of 0.908 (95% CI 0.847-0.951). Optimal cutoff for the diagnosis of malignant FBLs were US BI-RADS > 3, Strain Ratio > 2.52, Tsukuba Score > 3, Emax > 82.6 kPa, Emean > 66.0 kPa, Emin > 54.4 kPa and Eratio > 330.8. Multiparametric ultrasound, particularly SWE, can improve specificity in the characterization of FBLs.

Comparative Analysis of the Diagnostic Value of S-Detect Technology in Different Planes Versus the BI-RADS Classification for Breast Lesions.

RATIONALE AND OBJECTIVES: S-Detect, a deep learning-based Computer-Aided Detection system, is recognized as an important tool for diagnosing breast lesions using ultrasound imaging. However, it may exhibit inconsistent findings across multiple imaging planes. This study aims to evaluate the diagnostic performance of S-Detect in different planes and identify factors contributing to these inconsistencies. MATERIALS AND METHODS: A retrospective cohort study was conducted on 711 patients with 756 breast lesions between January 2019 and January 2022. S-Detect was utilized to assess lesions in radial and anti-radial planes. BI-RADS classifications were employed for comparative analysis. The diagnostic performance was compared within each group, and p-values were computed for intergroup comparisons. Univariable and multivariable analyses were conducted to identify factors contributing to diagnostic inconsistency in S-Detect across planes. RESULTS: Among 756 breast lesions, 668 (88.4%) exhibited consistent S-Detect outcomes across planes while 88 (11.6%) were inconsistent. In the consistent group, the diagnostic accuracy and area under the curve (AUC) of S-Detect were significantly higher than those of BI-RADS (accuracy: 91.2% vs. 84.9%, p = 0.045; AUC: 0.916 vs. 0.859, p = 0.036). In the inconsistent group, the diagnostic accuracy and AUC of S-Detect in radial and anti-radial planes were lower than those of BI-RADS (accuracy: 47.7% for radial, 52.2% for anti-radial vs. 69.3% for BI-RADS, p = 0.014, p-anti = 0.039; AUC: 0.503 for radial, 0.497 for anti-radial vs. 0.739 for BI-RADS, p = 0.042, p-anti <0.001). Diagnostic inconsistency in S-Detect across planes was significantly associated with lesion size, indistinct or angular margins, and enhancement posterior acoustic features (p < 0.05). CONCLUSION: S-Detect has outperformed BI-RADS in diagnostic precision under conditions of inter-planar concordance. However, its diagnostic efficacy is compromised in scenarios of inter-planar discordance. Under these circumstances, the results of S-Detect should be carefully referenced.

The Chiraiya project: a retrospective analysis of breast cancer detection gaps addressed via mobile mammography in Jammu Province, India.

BACKGROUND: Breast cancer remains a pervasive threat to women worldwide, with increasing incidence rates necessitating effective screening strategies. Timely detection with mammography has emerged as the primary tool for mass screening. This retrospective study, which is part of the Chiraiya Project, aimed to evaluate breast lesion patients identified during opportunistic mammography screening camps in Jammu Province, India. METHODS: A total of 1505 women aged 40 years and older were screened using a mobile mammographic unit over a five-year period, excluding 2020 and 2021 due to the COVID-19 pandemic. The inclusion criterion was women in the specified age group, while the exclusion criterion was women with open breast wounds, history of breast cancer or a history of breast surgery. The screening process involved comprehensive data collection using a detailed Proforma, followed by mammographic assessments conducted within strategically stationed mobile units. Radiological interpretations utilizing the BI-RADS system were performed, accompanied by meticulous documentation of patient demographics, habits, literacy, medical history, and breastfeeding practices. Participants were recruited through collaborations with NGOs, army camps, village panchayats, and urban cooperatives. Screening camps were scheduled periodically, with each camp accommodating 90 patients or fewer. RESULTS: Among the 1505 patients, most were aged 45-50 years. The number of screenings increased yearly, peaking at 441 in 2022. The BI-RADS II was the most common finding (48.77%), indicating the presence of benign lesions, while the BI-RADS 0 (32.96%) required further evaluation. Higher-risk categories (BI-RADS III, IV, V) were less common, with BI-RADS V being the rarest. Follow-up adherence was highest in the BI-RADS III, IV, and V categories, with BI-RADS V achieving 100% follow-up. However, only 320 of 496 BI-RADS 0 patients were followed up, indicating a gap in continuity of care. The overall follow-up rate was 66.89%. Compared to urban areas, rural areas demonstrated greater screening uptake but lower follow-up rates, highlighting the need for tailored interventions to improve follow-up care access, especially in rural contexts. CONCLUSION: This study underscores the efficacy of a mobile mammographic unit in reaching marginalized populations. Adherence to screening protocols has emerged as a linchpin for early detection, improved prognosis, and holistic public health enhancement. Addressing misconceptions surrounding mammographic screenings, especially in rural settings, is crucial. These findings call for intensified efforts in advocacy and education to promote the benefits of breast cancer screening initiatives. Future interventions should prioritize improving access to follow-up care and addressing screening to enhance breast cancer management in Jammu Province.

US-based radiomics analysis of different machine learning models for differentiating benign and malignant BI-RADS 4A breast lesions.

RATIONALE AND OBJECTIVES: To investigate and authenticate the effectiveness of various radiomics models in distinguishing between benign and malignant BI-RADS 4A lesions. METHODS: A total of 936 patients with pathologically confirmed 4A lesions were included in the study (training cohort: n = 655; test cohort: n = 281). Radiomic features were derived from greyscale US images. Following dimensionality reduction and feature selection, radiomics models were developed using logistic regression (LR), support vector machine (SVM), random forest (RF), eXtreme gradient boosting (XGBoost) and multilayer perceptron (MLP) algorithms. Univariate and multivariable logistic regression analyses were employed to investigate clinical-radiological characteristics and determine variables for creating a clinical model. Five combined models integrating radiomic and clinical parameters were constructed by using each algorithm, and comparison with radiologists' performance was performed. SHapley Additive exPlanations (SHAP) approach was used to elucidate the radiomic model by ranking the significance of features based on their contribution to the evaluation. RESULTS: A total of 1561 radiomic features were extracted. Thirty-six features were deemed significant by dimensionality reduction and selection. The radiomic models showed good performance with AUCs of 0.829-0.945 in training cohort; and 0.805-0.857 in test cohort. The combined model developed by using LR showed the best performance (AUC, training cohort: 0.909; test cohort: 0.905), which is superior to radiologists' performance. Decision curve analysis (DCA) of this combined model indicated better clinical efficacy than clinical and radiomic models. CONCLUSIONS: The combined model integrating radiomic and clinical features demonstrated excellent performance in differentiating between benign and malignant 4A lesions. It may offer a non-invasive and efficient approach to aid in clinical decision-making.

Spectral analysis enhanced net (SAE-Net) to classify breast lesions with BI-RADS category 4 or higher.

Early ultrasound screening for breast cancer reduces mortality significantly. The main evaluation criterion for breast ultrasound screening is the Breast Imaging-Reporting and Data System (BI-RADS), which categorizes breast lesions into categories 0-6 based on ultrasound grayscale images. Due to the limitations of ultrasound grayscale imaging, lesions with categories 4 and 5 necessitate additional biopsy for the confirmation of benign or malignant status. In this paper, the SAE-Net was proposed to combine the tissue microstructure information with the morphological information, thus improving the identification of high-grade breast lesions. The SAE-Net consists of a grayscale image branch and a spectral pattern branch. The grayscale image branch used the classical deep learning backbone model to learn the image morphological features from grayscale images, while the spectral pattern branch is designed to learn the microstructure features from ultrasound radio frequency (RF) signals. Our experimental results show that the best SAE-Net model has an area under the receiver operating characteristic curve (AUROC) of 12% higher and a Youden index of 19% higher than the single backbone model. These results demonstrate the effectiveness of our method, which potentially optimizes biopsy exemption and diagnostic efficiency.

Diagnostic value of contrast-enhanced ultrasound and shear-wave elastography for small breast nodules.

Background: The study aims to evaluate the diagnostic efficacy of contrast-enhanced ultrasound (CEUS) and shear-wave elastography (SWE) in detecting small malignant breast nodules in an effort to inform further refinements of the Breast Imaging Reporting and Data System (BI-RADS) classification system. Methods: This study retrospectively analyzed patients with breast nodules who underwent conventional ultrasound, CEUS, and SWE at Gongli Hospital from November 2015 to December 2019. The inclusion criteria were nodules ≤ 2 cm in diameter with pathological outcomes determined by biopsy, no prior treatments, and solid or predominantly solid nodules. The exclusion criteria included pregnancy or lactation and low-quality images. Imaging features were detailed and classified per BI-RADS. Diagnostic accuracy was assessed using receiver operating characteristic curves. Results: The study included 302 patients with 305 breast nodules, 113 of which were malignant. The diagnostic accuracy was significantly improved by combining the BI-RADS classification with CEUS and SWE. The combined approach yielded a sensitivity of 88.5%, specificity of 87.0%, positive predictive value of 80.0%, negative predictive value of 92.8%, and accuracy of 87.5% with an area under the curve of 0.877. Notably, 55.8% of BI-RADS 4A nodules were downgraded to BI-RADS 3 and confirmed as benign after pathological examination, suggesting the potential to avoid unnecessary biopsies. Conclusion: The integrated use of the BI-RADS classification, CEUS, and SWE enhances the accuracy of differentiating benign and malignant small breast nodule, potentially reducing the need for unnecessary biopsies.

Explainable Precision Medicine in Breast MRI: A Combined Radiomics and Deep Learning Approach for the Classification of Contrast Agent Uptake.

In DCE-MRI, the degree of contrast uptake in normal fibroglandular tissue, i.e., background parenchymal enhancement (BPE), is a crucial biomarker linked to breast cancer risk and treatment outcome. In accordance with the Breast Imaging Reporting & Data System (BI-RADS), it should be visually classified into four classes. The susceptibility of such an assessment to inter-reader variability highlights the urgent need for a standardized classification algorithm. In this retrospective study, the first post-contrast subtraction images for 27 healthy female subjects were included. The BPE was classified slice-wise by two expert radiologists. The extraction of radiomic features from segmented BPE was followed by dataset splitting and dimensionality reduction. The latent representations were then utilized as inputs to a deep neural network classifying BPE into BI-RADS classes. The network's predictions were elucidated at the radiomic feature level with Shapley values. The deep neural network achieved a BPE classification accuracy of 84 ± 2% (p-value < 0.00001). Most of the misclassifications involved adjacent classes. Different radiomic features were decisive for the prediction of each BPE class underlying the complexity of the decision boundaries. A highly precise and explainable pipeline for BPE classification was achieved without user- or algorithm-dependent radiomic feature selection.

Breast MRI-the Importance of Type II and III Dynamic Curves Evaluation and Framing in BI-RADS 4C and 5 Score.

BACKGROUND: Breast Magnetic Resonance Imaging (MRI) offers the highest sensitivity in detecting breast cancer among existing clinical and imaging techniques, making it a crucial component of breast imaging protocols. This study aims to investigate MRI importance in correlation with previous imaging discordant procedures performed as echography and/or mammography to evaluate characteristics and framing in high-risk BI-RADS 4C or 5 categories based on morphological features and kinetic curves of masses found in the breasts of patients from our database. METHODS: A retrospective study with related statistical analysis was performed on a group of 33 cases, selected from a total of 488 patients who underwent breast MRI examinations at SPAD Imaging International S.R.L. Craiova, between 01.01.2021 and 31.12.2023, aged between 33 and 75 years. In all patients, MRI images parameters were analysed. RESULTS: In 33 patients, 23 had a single lesion and 10 had multiple lesions, 9 of them in the ipsilateral breast and, as a particularity, one of them, located in the contralateral breast. In 21 of the total patients with multiple or single lesions they had type III curves, which were classified in the BI-RADS 5 category, considering both criteria-morphology and type of curve, where the other previous techniques had not mentioned an increased risk, hence revealing that the situation in a percentage of 63.63 in the case of MRI investigation proved to be clearly superior. CONCLUSION: Combining both kinetic and morphologic criteria can enhance the diagnostic accuracy of MRI in breast lesion evaluation.

Enhancing Accuracy in Breast Density Assessment Using Deep Learning: A Multicentric, Multi-Reader Study.

The evaluation of mammographic breast density, a critical indicator of breast cancer risk, is traditionally performed by radiologists via visual inspection of mammography images, utilizing the Breast Imaging-Reporting and Data System (BI-RADS) breast density categories. However, this method is subject to substantial interobserver variability, leading to inconsistencies and potential inaccuracies in density assessment and subsequent risk estimations. To address this, we present a deep learning-based automatic detection algorithm (DLAD) designed for the automated evaluation of breast density. Our multicentric, multi-reader study leverages a diverse dataset of 122 full-field digital mammography studies (488 images in CC and MLO projections) sourced from three institutions. We invited two experienced radiologists to conduct a retrospective analysis, establishing a ground truth for 72 mammography studies (BI-RADS class A: 18, BI-RADS class B: 43, BI-RADS class C: 7, BI-RADS class D: 4). The efficacy of the DLAD was then compared to the performance of five independent radiologists with varying levels of experience. The DLAD showed robust performance, achieving an accuracy of 0.819 (95% CI: 0.736-0.903), along with an F1 score of 0.798 (0.594-0.905), precision of 0.806 (0.596-0.896), recall of 0.830 (0.650-0.946), and a Cohen's Kappa (κ) of 0.708 (0.562-0.841). The algorithm achieved robust performance that matches and in four cases exceeds that of individual radiologists. The statistical analysis did not reveal a significant difference in accuracy between DLAD and the radiologists, underscoring the model's competitive diagnostic alignment with professional radiologist assessments. These results demonstrate that the deep learning-based automatic detection algorithm can enhance the accuracy and consistency of breast density assessments, offering a reliable tool for improving breast cancer screening outcomes.

Deep Learning for Describing Breast Ultrasound Images with BI-RADS Terms.

Breast cancer is the most common cancer in women. Ultrasound is one of the most used techniques for diagnosis, but an expert in the field is necessary to interpret the test. Computer-aided diagnosis (CAD) systems aim to help physicians during this process. Experts use the Breast Imaging-Reporting and Data System (BI-RADS) to describe tumors according to several features (shape, margin, orientation...) and estimate their malignancy, with a common language. To aid in tumor diagnosis with BI-RADS explanations, this paper presents a deep neural network for tumor detection, description, and classification. An expert radiologist described with BI-RADS terms 749 nodules taken from public datasets. The YOLO detection algorithm is used to obtain Regions of Interest (ROIs), and then a model, based on a multi-class classification architecture, receives as input each ROI and outputs the BI-RADS descriptors, the BI-RADS classification (with 6 categories), and a Boolean classification of malignancy. Six hundred of the nodules were used for 10-fold cross-validation (CV) and 149 for testing. The accuracy of this model was compared with state-of-the-art CNNs for the same task. This model outperforms plain classifiers in the agreement with the expert (Cohen's kappa), with a mean over the descriptors of 0.58 in CV and 0.64 in testing, while the second best model yielded kappas of 0.55 and 0.59, respectively. Adding YOLO to the model significantly enhances the performance (0.16 in CV and 0.09 in testing). More importantly, training the model with BI-RADS descriptors enables the explainability of the Boolean malignancy classification without reducing accuracy.

Limited impact of adding digital breast tomosynthesis to full field digital mammography in an elevated breast cancer risk population.

PURPOSE: To investigate the impact of adding digital breast tomosynthesis (DBT) to full field digital mammography (FFDM) in screening asymptomatic women with an elevated breast cancer life time risk (BCLTR) but without known genetic mutation. METHODS: This IRB-approved single-institution multi-reader study on prospectively acquired FFDM + DBT images included 429 asymptomatic women (39-69y) with an elevated BC risk on their request form. The BCLTR was calculated for each patient using the IBISrisk calculator v8.0b. The screening protocol and reader study consisted of 4-view FFDM + DBT, which were read by four independent radiologists using the BI-RADS lexicon. Standard of care (SOC) included ultrasound (US) and magnetic resonance imaging (MRI) for women with > 30 % BCLTR. Breast cancer detection rate (BCDR), sensitivity and positive predictive value were assessed for FFDM and FFDM + DBT and detection outcomes were compared with McNemar-test. RESULTS: In total 7/429 women in this clinically elevated breast cancer risk group were diagnosed with BC using SOC (BCDR 16.3/1000) of which 4 were detected with FFDM. Supplemental DBT did not detect additional cancers and BCDR was the same for FFDM vs FFDM + DBT (9.3/1000, McNemar p = 1). Moderate inter-reader agreement for diagnostic BI-RADS score was found for both study arms (ICC for FFDM and FFDM + DBT was 0.43, resp. 0.46). CONCLUSION: In this single institution study, supplemental screening with DBT in addition to standard FFDM did not increase BCDR in this higher-than-average BC risk group, objectively documented using the IBISrisk calculator.

Spatial N-glycomics of the normal breast microenvironment reveals fucosylated and high-mannose N-glycan signatures related to BI-RADS density and ancestry.

Higher breast cancer mortality rates continue to disproportionally affect black women (BW) compared to white women (WW). This disparity is largely due to differences in tumor aggressiveness that can be related to distinct ancestry-associated breast tumor microenvironments (TMEs). Yet, characterization of the normal microenvironment (NME) in breast tissue and how they associate with breast cancer risk factors remains unknown. N-glycans, a glucose metabolism-linked post-translational modification, has not been characterized in normal breast tissue. We hypothesized that normal female breast tissue with distinct Breast Imaging and Reporting Data Systems (BI-RADS) categories have unique microenvironments based on N-glycan signatures that varies with genetic ancestries. Profiles of N-glycans were characterized in normal breast tissue from BW (n = 20) and WW (n = 20) at risk for breast cancer using matrix assisted laser desorption/ionization (MALDI) mass spectrometry imaging (MSI). A total of 176 N-glycans (32 core-fucosylated and 144 noncore-fucosylated) were identified in the NME. We found that certain core-fucosylated, outer-arm fucosylated and high-mannose N-glycan structures had specific intensity patterns and histological distributions in the breast NME dependent on BI-RADS densities and ancestry. Normal breast tissue from BW, and not WW, with heterogeneously dense breast densities followed high-mannose patterns as seen in invasive ductal and lobular carcinomas. Lastly, lifestyles factors (e.g. age, menopausal status, Gail score, BMI, BI-RADS) differentially associated with fucosylated and high-mannose N-glycans based on ancestry. This study aims to decipher the molecular signatures in the breast NME from distinct ancestries towards improving the overall disparities in breast cancer burden.