BI-RADS Research Articles

Abstract 4515: Epidemiological factors associated with breast density in Guam

Abstract Guam's breast cancer incidence is lower than the U.S. (86.6 vs. 126.8 per 100, 000), but mortality rates are higher (27.3 vs. 20.3). This disparity underscores the importance of improving screening and understanding risk factors. Breast density, categorized using the Breast Imaging Reporting and Data System (BIRADS), complicates tumor detection and correlates with malignancy risk. This study investigates the demographic and reproductive factors associated with breast density in Guam. Data from 1, 940 women undergoing mammography at FHP Health Center (Nov 2023-Aug 2024) in the HIPIMR study were analyzed. Associations between demographic factors (ethnicity, BMI) and reproductive factors (family history, age at menarche, menopausal status, parity, age at first live birth) with BIRADS categories were tested using chi-square tests. Reproductive factors were categorized by risk level. Multinomial logistic regression, adjusted for age and BMI, examined associations. Missing data (e.g., BMI, age, parity) were imputed using series means/medians. Women with extremely dense breast tissue (BIRADS D) were younger (mean age 57.13 vs. 58.32, p<0.001) and had lower BMI (74% vs. 37%, p<0.001). They more often experienced menarche after age 12 (58% vs. 51%, p=0.002), had fewer than two births (64% vs. 53%, p<0.001), and gave birth after age 30 (65% vs. 51%, p<0.001). Premenopausal (32%) and early menopause women (51%) had higher BIRADS D proportions (p<0.001, p=0.005). Regression showed BMI inversely associated with BIRADS D (OR = 0.815, 95% CI: 0.787-0.844), BIRADS C (OR = 0.875, 95% CI: 0.842-0.908), and BIRADS B (OR = 0.937, 95% CI: 0.907-0.968) (p<0.001). Family history increased odds of BIRADS D (OR = 1.64, 95% CI: 1.041-2.584, p=0.033) and BIRADS B (OR = 1.674, 95% CI: 1.042-2.69, p=0.033). Nulliparous women were 2.66 times more likely in BIRADS B (OR = 2.66, 95% CI: 1.157-6.114, p=0.021). Micronesians were less likely to belong to BIRADS D (OR = 0.419, 95% CI: 0.19-0.924, p=0.031) or BIRADS C (OR = 0.265, 95% CI: 0.092-0.766, p=0.014). BMI, parity, and childbirth age significantly influence breast density in Guam. Women with lower BMI, fewer births, or later childbirth were more likely to have extremely dense breast tissue, potentially complicating tumor detection. Ethnicity also played a role, with Micronesian women less likely in BIRADS D. While imputation mitigated missing data, bias remains possible. Future research should expand sample sizes and include malignancy risk via BIRADS outcomes for deeper insight. This research was funded by the National Cancer Institute: The University of Guam/University of Hawaii Cancer Center Partnership to Advance Cancer Health Equity (PIPCHE), Grant U54CA143728 (University of Guam, YCP)/Grant U54CA143727 (University of Hawaii Cancer Center). Citation Format: Gian M. Paras, Rodney S. Teria, Su Bin Jin, Grazyna Badowski, John Shepherd, Rose Grino. Epidemiological factors associated with breast density in Guam [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 4515.

A large model for non-invasive and personalized management of breast cancer from multiparametric MRI

Breast Magnetic Resonance Imaging (MRI) demonstrates the highest sensitivity for breast cancer detection among imaging modalities and is standard practice for high-risk women. Interpreting the multi-sequence MRI is time-consuming and prone to subjective variation. We develop a large mixture-of-modality-experts model (MOME) that integrates multiparametric MRI information within a unified structure, leveraging breast MRI scans from 5205 female patients in China for model development and validation. MOME matches four senior radiologists’ performance in identifying breast cancer and outperforms a junior radiologist. The model is able to reduce unnecessary biopsies in Breast Imaging-Reporting and Data System (BI-RADS) 4 patients, classify triple-negative breast cancer, and predict pathological complete response to neoadjuvant chemotherapy. MOME further supports inference with missing modalities and provides decision explanations by highlighting lesions and measuring modality contributions. To summarize, MOME exemplifies an accurate and robust multimodal model for noninvasive, personalized management of breast cancer patients via multiparametric MRI.

Sonographic Glandular Tissue Component: APotential Imaging Marker for Upgrading BI-RADS 4A Breast Masses.

Sonographic Glandular Tissue Component: APotential Imaging Marker for Upgrading BI-RADS 4A Breast Masses.

An explainable and comprehensive BI-RADS assisted diagnosis pipeline for mammograms.

An explainable and comprehensive BI-RADS assisted diagnosis pipeline for mammograms.

Roles of artificial intelligence and high frame-rate contrast-enhanced ultrasound in the differential diagnosis of Breast Imaging Reporting and Data System 4 breast nodules.

Breast cancer prevalence and mortality are rising, emphasizing the need for early, accurate diagnosis. Contrast-enhanced ultrasound (CEUS) and artificial intelligence (AI) show promise in distinguishing benign from malignant breast nodules. We compared the diagnostic values of AI, high frame-rate CEUS (HiFR-CEUS), and their combination in Breast Imaging Reporting and Data System (BI-RADS) 4 nodules, using pathology as the gold standard. Patients with BI-RADS 4 breast nodules who were hospitalized at the Department of Thyroid and Breast Surgery, Taizhou People's Hospital from December 2021 to June 2022 were enrolled in the study.80 female patients (80 lesions) underwent preoperative AI and/or HiFR-CEUS. We assessed diagnostic outcomes of AI, HiFR-CEUS, and their combination, calculating sensitivity (SE), specificity (SP), accuracy (ACC), positive/negative predictive values (PPV/NPV). Reliability was compared using Kappa statistics, and AI-HiFR-CEUS correlation was analyzed with Pearson's test. Receiver operating characteristic curves were plotted to compare diagnostic accuracy of AI, HiFR-CEUS, and their combined approach in differentiating BI-RADS 4 lesions. Of the 80 lesions, 18 were pathologically confirmed to be benign, while the remaining 62 were malignant. The SE, SP, ACC, PPV, and NPV were 75.81%, 94.44%, 80.00%, 97.92%, and 53.13% in the AI group, 74.20%, 94.44%, 78.75%, 97.91%, and 51.51% in the HiFR-CEUS group, and 98.39%, 88.89%, 96.25%, 96.83%, and 94.12% in the combination group, respectively. Thus, the SE, ACC, and NPV of the combination group were significantly higher than those of the AI and HiFR-CEUS groups, and the SP of the combination group was lower (all P<0.05); however, no significant difference was found between the groups in terms of the PPV (P>0.05). No statistically significant difference was observed in the diagnostic performance of the AI and HiFR-CEUS groups (all P>0.05). The AI and HiFR-CEUS groups had moderate agreement with the "gold standard" (Kappa =0.551, Kappa =0.530, respectively), while the combination group had high agreement (Kappa =0.890). AI was positively correlated with HiFR-CEUS (r=0.249, P<0.05). The area under the curves (AUCs) of AI, HiFR-CEUS, and both in combination were 0.851±0.039, 0.815±0.047, and 0.936±0.039, respectively. Thus, the AUC of the combination group was significantly higher than those of the AI and HiFR-CEUS groups (Z1=2.207, Z2=2.477, respectively, both P<0.05). The AI group had a higher AUC than the HiFR-CEUS group, but the difference was not statistically significant (Z3=0.554, P>0.05). Compared with AI alone or HiFR-CEUS alone, the combined use of these two methods had higher diagnostic performance in distinguishing between benign and malignant BI-RADS 4 breast nodules. Thus, our combination method could further improve the diagnostic accuracy and guide clinical decision making.

The diagnostic value of real-time ultrasound elastography and contrast-enhanced ultrasound in BI-RADS 4A breast lesions.

The Breast Imaging-Reporting and Data System (BI-RADS) is the primary system for classifying clinical breast lesions. Most early lesions identified via ultrasound are classified as BI-RADS 4A or lower. Although the vast majority of BI-RADS 4A lesions are benign, those lesions still have the possibility of malignancy in clinical practice, which is a controversial and noteworthy issue. This study aimed to assess the diagnostic value of real-time ultrasound elastography (UE) and contrast-enhanced ultrasound (CEUS) in evaluating BI-RADS 4A breast lesions. A retrospective analysis was conducted of the UE and CEUS data of 52 BI-RADS 4A breast lesions from 52 patients between January 2020 and March 2023. All diagnoses were confirmed by surgical pathology. Lesion characteristics, including the margins, echogenicity, size, microcalcifications, blood flow patterns, UE scores, and CEUS features, were analyzed. CEUS scores were based on a five-point system, and the area under the curve (AUC) was calculated using MedCalc version 19.0.4. Based on the postoperative pathology, of the 52 lesions, 27 were benign and 25 were malignant. Compared to those with benign lesions, the patients with malignant lesions were older and had larger lesions (P<0.05). Features such as irregular morphology, indistinct margins, increased blood flow, and calcifications were more common in the malignant lesions than the benign lesions (P<0.05). The malignant lesions also had a higher prevalence of inhomogeneous enhancement, vasa vasorum, irregular enhanced morphology, and crab claw-like signs on CEUS than the benign lesions (P<0.05). The diagnostic accuracy of both the UE and CEUS individually was 76.9%, while the diagnostic accuracy of the UE and CEUS combined reached 80.8%. The AUCs of UE, CEUS, and the UE and CEUS combined were 0.761, 0.773, and 0.813, respectively. UE and CEUS have significant diagnostic value for BI-RADS 4A breast lesions. Combining these techniques improves diagnostic accuracy and can help reduce unnecessary biopsies.

Development and validation of a semi-automatic radiomics ensemble model for preoperative evaluation of breast masses in mammotome-assisted minimally invasive resection.

Accurate preoperative differentiation of breast masses is critical for guiding individualized treatment strategies in Mammotome-assisted minimally invasive resection. While radiomics shows promise, existing methods rely on manual delineation, which is time-consuming and subjective. This study developed an ultrasound-based semi-automatic segmentation ensemble model to improve preoperative assessment. We retrospectively analyzed preoperative ultrasound images from 773 patients (543 tumors, 230 non-tumors). Semi-automatic segmentation was performed using DeepLabv3_ResNet50 and fully convolutional network (FCN)_ResNet50. Radiomic and deep transfer learning (DTL) features were extracted to construct radiomic, deep learning, and combined models. An ensemble strategy integrated these with clinical models. Performance was evaluated via receiver operating characteristic (ROC) curves and decision curve analysis (DCA). The cohort included 543 tumor patients and 230 non-tumor patients (95 adenosis, 135 other benign lesions). The semi-automatic segmentation model, DeepLabv3_ResNet50, achieved a peak global accuracy of 99.4% and an average Dice coefficient of 92.0% at its best epoch. On the other hand, the FCN_ResNet50 model exhibited a peak global accuracy of 99.5% and an average Dice coefficient of 93.7% at its best epoch. In the task of predicting tumor and non-tumor patients, age, maximum diameter, and BI-RADS (Breast Imaging Reporting and Data System) classification were ultimately identified as key indicators, and the stacking model ultimately demonstrated an area under the curve (AUC) of 0.890 in the training cohort (with a sensitivity of 0.844 and a specificity of 0.815) and an AUC of 0.780 in the testing cohort (with a sensitivity of 0.713 and a specificity of 0.739). In the task of predicting adenosis and other lesion types, focus emerged as a crucial factor, and the stacking model achieved an AUC of 0.813 in the training cohort (with a sensitivity of 0.613 and a specificity of 0.859) and an AUC of 0.771 in the testing cohort (with a sensitivity of 0.759 and a specificity of 0.765). Our study has established an ensemble learning model grounded in semi-automatic segmentation techniques. This model accurately distinguishes between tumor and non-tumor patients preoperatively, as well as discriminating adenosis from other lesion types among the non-tumor cohort, thus providing valuable insights for individualized treatment planning. The proposed stacking model demonstrates significant clinical utility by reducing unnecessary biopsies and saving diagnostic time compared to manual review. These improvements directly address the challenges of overtreatment and diagnostic delays in breast lesion management. By enhancing preoperative accuracy, our model supports tailored surgical planning and alleviates patient anxiety associated with indeterminate diagnoses.

The effectiveness of sono-mammographic findings in predicting the molecular subtypes of breast cancer. Using a sono-mammography score in distinguishing between non-triple-negative and triple-negative breast cancer

BackgroundBreast cancer is the most frequently diagnosed invasive cancer among women. Knowing the molecular subtyping of breast cancer is a standard approach for personalized cancer treatment, helping to evaluate the prognosis of the disease and prevent unnecessary treatment. Assessing the molecular subtypes of breast cancer via imaging can be intricate, as it requires the combination of imaging features from both ultrasound and mammography. The study aimed to assess the effectiveness of mammographic and ultrasound imaging features in predicting breast cancer’s molecular status. It also utilized a scoring system implemented by Shaikh and Rasheed to analyze the imaging characteristics of non-triple-negative versus triple-negative breast cancers.MethodsWe initiated this prospective research study from December 2022 to June 2024, involving 248 patients diagnosed with breast masses. Each patient underwent both digital mammography and ultrasound examinations. We assessed the mammographic and sonographic features of the breast carcinomas using the analytical criteria of the breast imaging reporting and data system (BI-RADS). Subsequently, we conducted a histopathological evaluation to determine the tumor type, grade, and molecular subtypes. We also followed a scoring system previously implemented by Shaikh and Rasheed to comprehend the imaging characteristics of non-triple-negative breast cancer (non-TNBC) in comparison with triple-negative breast cancers (TNBCs).ResultsInvasive ductal carcinoma was the most common pathological type detected in 201 lesions (81%) while grade II was the most encountered among the studied cases found in 135 lesions (54.4%) and Luminal B was the most common molecular subtype presented in 119 cases (48%). Sono-mammographic imaging features showed a statistically significant difference regarding lesions’ shape, margin, number, skin thickening, nipple retraction, and axillary lymphadenopathy between the different molecular subtypes (p < 0.001). Breast lesions with round or oval shapes, well-circumscribed or microlobulated margins, non-calcified or microcalcifications, posterior enhancement, parallel orientation, grade III tumors, and positive KI + 67 are significant predictors of triple-negative breast cancer, with odds ratios of 1.23, 5.76, 2.14, 200.03, 1.106, 2.706, and 1.25, respectively.ConclusionsThe current study showed that sono-mammographic findings can serve as valuable tools for identifying molecular subtypes of breast cancer. Different molecular subtypes exhibit unique imaging characteristics, particularly triple-negative breast cancer, which tends to display the least aggressive features.

Positive Predictive Rate of Breast Imaging-Reporting and Data System (BI-RADS) Category 4 Lesions For Breast Cancer Taking Histopathology as Gold Standard

Objective: To determine positive predictive rate of Breast imaging-reporting and data system (BI-RADS) category 4 and its sub-categories for breast cancer taking histopathology as gold standard. Study Design: Cross sectional validation study. Place and Duration of Study: Department of Radiology, Pakistan Institute of Medical Sciences, Islamabad Pakistan, from Sep to Nov 2020. Methodology: 60 female patients who were diagnosed as Breast imaging-reporting and data system (BI-RADS) category 4 lesions on mammography were included in the study and afterwards followed for biopsy report. Data was collected and analyzed using SPSSv25. Results: Patient’s mean age was 50.6±9.7yrs. Among sixty patients with BIRADS 4, 13(21.7%) had BIRADS 4A, 22(36.7%) had BIRADS 4B and 25(41.7%) had 4C. Positive predictive value for BIRADS 4 was 68.33%, 4A was 15.3%, 4B was 72.7% and for 4C was 92%. Sensitivity &amp; Specificity of BIRADS 4A being benign and BIRADS 4B, &amp; 4C being malignant is 91.49% and 69.23% respectively. Overall, diagnostic accuracy of BIRADS 4B &amp; 4C classification was found to be 83.33%. Conclusion: Sub-categorization of Breast imaging-reporting and data system (BI-RADS) category 4 is not routinely practiced, however results of current study conclude that it helps to stratify the associated risk of malignancy and update the level of concern to referring physician and pathologist.

Combined Mammographic Breast Density and Breast Arterial Calcification as an Incremental Predictor of Coronary Artery Disease.

Background: Contemporary risk calculators underestimate coronary artery disease (CAD) risk in women. Breast arterial calcification (BAC) associates with CAD. Low breast density (BD) (greater breast adipose tissue) associates with cardiometabolic disease. Both are readily identifiable on screening mammography. We sought to evaluate the association between the combined features of BD, BAC, and CAD. Methods: We retrospectively studied women with clinically indicated mammography and contemporaneous coronary computed tomography angiography. CAD risk was estimated by CAD Consortium Scoring (CCS;>15% high risk). BD was visually assessed by four-level Breast Imaging-Reporting and Data System (BI-RADS) (low:BI-RADS A-B, high:BI-RADS C-D). BAC was visually assessed as present/absent. CAD was categorized as presence/absence of coronary artery plaque. Results are presented with odds ratio (OR) and [95% confidence intervals], and area under the curve (AUC). Results: In 153 patients (age 62 ± 10), low BD (67%) and BAC presence (24%) were both associated with CAD, respectively: OR: 3.21 [1.58-6.60], p = 0.001, and OR: 4.36 [1.58-12.00], p = 0.004. CAD proportion in low BD (68.9%) and BAC (42.9%) was lower than with combined low BD+BAC positive (89.7%). Compared with (high BD+BAC negative), the presence of (low BD+BAC positive) associated with CAD independent of modifiable (OR: 9.12 [2.44-45.83], p = 0.002) and nonmodifiable (OR: 4.87 [1.22-25.02], p = 0.035) risk factors. CCS >15% was seen in 33%. Significant incremental value was seen with the addition of BD/BAC status to CCS (AUC 0.64 versus 0.73, p = 0.004). Conclusions: Mammographic BAC and low BD, both alone and combined, associate with CAD, and improve risk prediction beyond standard coronary risk estimation. Standardized reporting of these features may provide benefit and should be tested in prospective screening studies.

SURGICAL ONCOLOGY-BASED CHECK-UP PROGRAM AND DETERMINING OF ASYMPTOMATIC PATHOLOGIES

Introduction: Our study aims to determine the effectiveness of check-up programs in the early diagnosis of benign and malignant pathologies, and evaluate their importance. Methods: Posteroanterior chest X-ray, thyroid, and abdominal ultrasonography were performed. Breast ultrasonography was performed on all females and mammography on those over 40 years. Fecal occult blood, carcinoembryonic antigen, and prostate-specific antigen tests were performed on the cases along with hemogram and routine biochemical tests. Radiological images and reports were retrospectively scanned, and fecal occult blood, carcinoembryonic antigen, and prostate-specific antigen values were examined. Breast Imaging Reporting and Data Systems (BIRADS) in breast radiology and the Thyroid Imaging Reporting and Data System (TIRADS) in thyroid radiology were used for classification. Results: 1411 cases were included in this study. 49.4% were female and 50.6% male. Median age was 50 years. Benign lesions in the liver were detected in 5.31% of the cases, kidney stones in 4.7%, and cholelithiasis in 5.7%. Benign Prostatic Hyperplasia was observed in 20% of the males. Thyroid gland nodule with a risk of malignancy in 3.8% and biopsy was recommended. Malignancy-suspicious lesions were detected in 0.28%, that performed breast ultrasonography and in 0.8%, that performed mammography, and biopsy was recommended. Normal lung parenchymal findings were detected in 75.4%, minor in 22%, and major in 2.6%, and thoracic computed tomography was recommended. Conclusions: Effective check-up programs are found to be beneficial for public health as they provide diagnosis and follow-up in benign lesions and early diagnosis and treatment opportunities in malignant/premalignant lesions.

Distinguishing Low Expression Levels of Human Epidermal Growth Factor Receptor 2 in Breast Cancer: Insights from Qualitative and Quantitative Magnetic Resonance Imaging Analysis.

The discovery of novel antibody-drug conjugates for low-expression human epidermal growth factor receptor 2 (HER2-low) breast cancer highlights the inadequacy of the conventional binary classification of HER2 status as either negative or positive. Identification of HER2-low breast cancer is crucial for selecting patients who may benefit from targeted therapies. This study aims to determine whether qualitative and quantitative magnetic resonance imaging (MRI) features can effectively reflect low-HER2-expression breast cancer. Pre-treatment breast MRI images from 232 patients with pathologically confirmed breast cancer were retrospectively analyzed. Both clinicopathologic and MRI features were recorded. Qualitative MRI features included Breast Imaging Reporting and Data System (BI-RADS) descriptors from dynamic contrast-enhanced MRI (DCE-MRI), as well as intratumoral T2 hyperintensity and peritumoral edema observed in T2-weighted imaging (T2WI). Quantitative features were derived from diffusion kurtosis imaging (DKI) using multiple b-values and included statistics such as mean, median, 5th and 95th percentiles, skewness, kurtosis, and entropy from apparent diffusion coefficient (ADC), Dapp, and Kapp histograms. Differences in clinicopathologic, qualitative, and quantitative MRI features were compared across groups, with multivariable logistic regression used to identify significant independent predictors of HER2-low breast cancer. The discriminative power of MRI features was assessed using receiver operating characteristic (ROC) curves. HER2 status was categorized as HER2-zero (n = 60), HER2-low (n = 91), and HER2-overexpressed (n = 81). Clinically, estrogen receptor (ER), progesterone receptor (PR), hormone receptor (HR), and Ki-67 levels significantly differed between the HER2-low group and others (all p < 0.001). In MRI analyses, intratumoral T2 hyperintensity was more prevalent in HER2-low cases (p = 0.009, p = 0.008). Mass lesions were more common in the HER2-zero group than in the HER2-low group (p = 0.038), and mass shape (p < 0.001) and margin (p < 0.001) significantly varied between the HER2 groups, with mass shape emerging as an independent predictive factor (HER2-low vs. HER2-zero: p = 0.010, HER2-low vs. HER2-over: p = 0.012). Qualitative MRI features demonstrated an area under the curve (AUC) of 0.763 (95% confidence interval [CI]: 0.667-0.859) for distinguishing HER2-low from HER2-zero status. Quantitative features showed distinct differences between HER2-low and HER2-overexpression groups, particularly in non-mass enhancement (NME) lesions. Combined variables achieved the highest predictive accuracy for HER2-low status, with an AUC of 0.802 (95% CI: 0.701-0.903). Qualitative and quantitative MRI features offer valuable insights into low-HER2-expression breast cancer. While qualitative features are more effective for mass lesions, quantitative features are more suitable for NME lesions. These findings provide a more accessible and cost-effective approach to noninvasively identifying patients who may benefit from targeted therapy.

Detection and BI-RADS Classification of Breast Nodules in Urban Women - China, 2021.

Female breast nodules represent the most frequently detected lesions during breast ultrasound screening. Notably, nodules classified as Breast Imaging Reporting and Data System (BI-RADS) 4 or 5 indicate an elevated risk of breast cancer. Nevertheless, the detection rate and BI-RADS classification of female breast nodules in China remain largely undocumented. In 2021, breast nodules were detected in 27.9% of urban women in China. Among women with breast nodules marked with BI-RADS classification information, 95.9% were categorized as BI-RADS 2-3, while 4.0% were classified as BI-RADS 4-5. Age, geographic location, per capita gross domestic product (GDP), body mass index (BMI), high triglyceride (TG), high low-density lipoprotein cholesterol (LDL-C), and diabetes were identified as risk factors for BI-RADS 4-5. This study highlights the importance of managing high-risk women with breast nodules through BI-RADS classification. Women with older age, high TG, high LDL-C, or diabetes demonstrate higher detection rates of BI-RADS 4-5, underscoring the need for targeted health interventions for high-risk populations while accounting for regional and socioeconomic disparities.

Ultrasound-based deep learning radiomics for multi-stage assisted diagnosis in reducing unnecessary biopsies of BI-RADS 4A lesions.

Even with the Breast Imaging Reporting and Data System (BI-RADS) guiding risk stratification on ultrasound (US) images, inconsistencies in diagnostic accuracy still exist, leading patients being subjected to unnecessary biopsies in clinical practice. This study investigated the construction of deep learning radiomics (DLR) models to improve the diagnostic consistency and reduce the unnecessary biopsies for BI-RADS 4A lesions. A total of 746 patients with breast lesions were enrolled in this retrospective study. Two DLR models based on US images and clinical variables were developed to conduct breast lesion risk re-stratification as BI-RADS 3 or lower and BI-RADS 4A or higher (DLR_LH), while simultaneously identifying BI-RADS 4A lesions with low malignancy probabilities to avoid unnecessary biopsy (DLR_BM). A three-round reader study with a two-stage artificial intelligence (AI)-assisted diagnosis process was performed to verify the assistive capability and practical benefits of the models in clinical applications. The DLR_LH model achieved areas under the receiver operating characteristic curve (AUCs) of 0.963 and 0.889 with sensitivities of 92.0% and 83.3%, in the internal and external validation cohorts, respectively. The DLR_BM model exhibited AUCs of 0.977 and 0.942, with sensitivities of 94.1% and 86.4%, respectively. Both models were evaluated using integrated features of US images and clinical variables. Ultimately, 27.7% of BI-RADS 4A lesions avoided unnecessary biopsies. In the three-round reader study, all readers achieved significantly higher diagnostic accuracy and specificity, while maintaining outstanding sensitivity comparable to human experts, both before and after model assistance (P<0.05). These findings demonstrate the positive impact of the DLR models in assisting radiologists to enhance their diagnostic capabilities. The models performed well in breast US imaging interpretation and BI-RADS risk re-stratification, and demonstrated potential in reducing unnecessary biopsies of BI-RADS 4A lesions, indicating the promising applicability of the DLR models in clinical diagnosis.

Addition of contrast-enhanced mammography enhancement patterns and morphology for differentiating benign from malignant papillary breast lesions.

To investigate the capability of morphological and enhancement pattern analysis in contrast-enhanced mammography (CEM) to differentiate between benign and malignant papillary breast lesions. This study involved 63 patients diagnosed with papillary breast lesions between January 2018 and December 2022. CEM was performed at two time intervals: between 2-3 min (T1, early phase) and 4-5 min (T2, later phase) after the injection of the contrast agent. For each patient, Breast Imaging Reporting and Data System (BI-RADS) lesion features were recorded. Kinetic enhancement patterns were evaluated for both phases based on these changes, followed by the measurement of diagnostic performance. Among the 63 female patients, a total of 86 lesions were identified, including 18 malignant ones, three intraductal papillomas without atypical proliferation, and 42 benign papillary lesions. The areas under receiver operating characteristic curves for CEM with kinetic enhancement and for CEM alone were 0.856 and 0.800, respectively. Assessing the diagnostic value, specificity, and accuracy of morphological categorization combined with kinetic enhancement showed higher values compared to those of the CEM morphological categorization alone (71.1% vs 60.0% and 79.4% vs 71.4%, respectively). However, the sensitivity and negative predictive value of the CEM with kinetic enhancement were similar to those of CEM alone (100% for all). For BI-RADS 3-5 papillary breast lesions, incorporating CEM kinetic enhancement to morphological patterns improved the confidence level in diagnosis. This article provides valuable references for distinguishing benign and malignant breast papillary lesions.

Intra- and peri-tumoral radiomics based on dynamic contrast-enhanced MRI for prediction of benign disease in BI-RADS 4 breast lesions: a multicentre study

Background and purposeThe study aimed to create a radiomics model based on breast intra- and peri-tumoral regions in dynamic contrast-enhanced (DCE) MRI to distinguish benign from malignant breast lesions of Breast Imaging Reporting and Data System (BI-RADS) 4.Materials and methodsA total of 516 patients from Hospital 1 were assigned to the training cohort. Then, 146 and 52 patients were enrolled from Hospital 2 and 3, respectively, as the internal and external test cohort. Seven classification models were built, using features extracted from the intra- and peri-tumoral regions. Diagnostic performance was evaluated by receiver operating characteristics (ROC) analysis and compared by the DeLong test. Subgroup analysis was performed after stratifying all lesions by enhancement pattern and the subdivision of BI-RADS 4.ResultsThe Comb2 model, built with features from peri-tumoral 2 mm and intra-tumoral region, demonstrated the best performance with AUCs of 0.828 and 0.844 in the internal and external test cohort, respectively. The Comb2 model was robust in both mass and non-mass enhancement (NME) lesions. At the three exploratory cutoff values on the ROC curve, the model identified 9.1% (sensitivity of C1 ≥ 98%), 27.3% (sensitivity of C2 ≥ 95%) and 36.4% (sensitivity of C3 ≥ 90%) of the benign lesions in the external test cohort. Applying the identified cutoff values in the external test cohort showed the potential to lower the number of unnecessary biopsies of benign lesions.ConclusionAn MRI-based radiomics model built with features extracted from the intra-tumoral region combined with the peri-tumoral 2 mm showed the best potential to reduce false-positive diagnoses and may avoid unnecessary biopsies with a low underestimate risk.

Combining ultrasound gray scale ratio and BI-RADS grading for differentiating benign and malignant breast masses

Background The evaluation of breast mass echo in the breast imaging reporting and data system (BI-RADS) is subjective, whereas ultrasound grayscale ratio (UGSR) provides a quantitative measure of ultrasound echo intensity. The purpose of this study is to investigate the value of UGSR, in conjunction with BI-RADS, for distinguishing between benign and malignant breast masses. Methods A retrospective analysis was conducted on ultrasound imaging data from 179 malignant breast mass cases and 219 benign breast mass cases. The masses were categorized according to the BI-RADS grading standard, and the grayscale values of the masses, adjacent fat lobules, and mammary glands were measured using the RADinfo system. The f-UGSR (ratio of ultrasound grayscale of mass to ultrasound grayscale of fat lobule) and g-UGSR (ratio of ultrasound grayscale of mass to ultrasound grayscale of gland) were calculated. The receiver operating characteristic (ROC) curve was constructed, optimal thresholds for f-UGSR and g-UGSR were determined, and the diagnostic efficacy of the combination of f-UGSR, g-UGSR and BI-RADS grading was assessed. Results The f-UGSR for malignant and benign breast masses were 0.57 ± 0.20 and 0.85 ± 0.23, respectively, while the g-UGSR were 0.23 (0.18, 0.30) and 0.38 (0.30, 0.47) respectively. The area under the ROC curve (AUC) for f-UGSR and g-UGSR in the diagnosis of breast malignant masses were 0.819 and 0.823, respectively, with the optimal thresholds of 0.69 and 0.29, respectively. The AUC for BI-RADS, g-UGSR, and their combination in the diagnosis of breast malignant masses were 0.849, 0.823, and 0.821, respectively. Sensitivity rates were approximately 74.30%, 73.74%, and 89.39%, and specificity rates were 95.43%, 77.17%, and 74.89%, respectively. Conclusion UGSR provides an objective and accurate method to evaluate the benign or malignant nature of breast masses. Combining UGSR with BI-RADS enhances sensitivity, reduces the missed diagnosis rate, and improves confidence in distinguishing between benign and malignant breast masses.

The application of contrast-enhanced ultrasound and MicroFlow Imaging in the diagnosis of breast cancer.

The purpose of this study was to compare the value of contrast-enhanced ultrasound (CEUS) plus MicroFlow Imaging (CEUS-MFI), MicroFlow Imaging (MFI) alone, and color Doppler flow imaging (CDFI) in the differential diagnosis of benign and malignant breast lesions. A total of 116 patients with 116 breast lesions categorized as Breast Imaging Reporting and Data System (BI-RADS) category 4 were enrolled in this prospective study. CEUS-MFI, MFI, and CDFI were used to evaluate the microvascular morphology and distribution types of breast lesions. Pathological results were considered the gold standard. Compared with conventional CDFI and MFI, the CEUS-MFI technique can reveal microvasculature and distribution types at a higher resolution in breast masses. The sensitivity, specificity, and accuracy of CEUS-MFI were 96.3%, 80.6%, and 91.4%, respectively. The sensitivity, specificity, and accuracy of MFI were 87.5%, 75.0%, and 83.6%, respectively. The sensitivity, specificity, and accuracy of CDFI were 58.8%, 72.2%, and 62.9%, respectively. The accuracy was significantly different between CEUS-MFI and MFI (P = 0.024), and the accuracy was significantly different between CEUS-MFI and CDFI (P = 0.000). US BI-RADS 4A masses were downgraded based on CEUS-MFI features without any malignancy missed, with the biopsy rate decreasing from 100% (29/29) to 31.0% (9/29). CEUS-MFI provides improved diagnostic efficacy for breast lesions. The CEUS-MFI technique can be used as an effective supplement to conventional ultrasound in the diagnosis of breast tumors.

A nomogram for diagnosis of BI-RADS 4 breast nodules based on three-dimensional volume ultrasound

ObjectivesThe classification of malignant breast nodules into four categories according to the Breast Imaging Reporting and Data System (BI-RADS) presents significant variability, posing challenges in clinical diagnosis. This study investigates whether a nomogram prediction model incorporating automated breast ultrasound system (ABUS) can improve the accuracy of differentiating benign and malignant BI-RADS 4 breast nodules.MethodsData were collected for a total of 257 nodules with breast nodules corresponding to BI-RADS 4 who underwent ABUS examination and for whom pathology results were obtained from January 2019 to August 2022. The participants were divided into a benign group (188 cases) and a malignant group (69 cases) using a retrospective study method. Ultrasound imaging features were recorded. Logistic regression analysis was used to screen the clinical and ultrasound characteristics. Using the results of these analyses, a nomogram prediction model was established accordingly.ResultsAge, distance between nodule and nipple, calcification and C-plane convergence sign were independent risk factors that enabled differentiation between benign and malignant breast nodules (all P < 0.05). A nomogram model was established based on these variables. The area under curve (AUC) values for the nomogram model, age, distance between nodule and nipple, calcification, and C-plane convergence sign were 0.86, 0.735, 0.645, 0.697, and 0.685, respectively. Thus, the AUC value for the model was significantly higher than a single variable.ConclusionsA nomogram based on the clinical and ultrasound imaging features of ABUS can be used to improve the accuracy of the diagnosis of benign and malignant BI-RADS 4 nodules. It can function as a relatively accurate predictive tool for sonographers and clinicians and is therefore clinically useful.Advances in knowledge statementwe retrospectively analyzed the clinical and ultrasound characteristics of ABUS BI-RADS 4 nodules and established a nomogram model to improve the efficiency of the majority of ABUS readers in the diagnosis of BI-RADS 4 nodules.

An Analysis of the Results of Breast Imaging Reporting and Data System 4 Lesions in Tertiary Care Center in South India.

The Breast Imaging Reporting and Data System (BIRADS) is a reporting system used to describe the results of mammogram, breast ultrasound, or breast MRI in a standard way. BIRADS ranks the test findings according to one of the seven categories, ranging from normal or benign to highly suspicious of cancer or malignant. This has specific accuracy for breast cancer. Out of the seven categories, BIRADS 4 is linked with a greater possibility for carcinoma breast. The research objective was to establish the rate of malignancy and clinical net result of BIRADS 4 lesions. The retrospective study, conducted in in Lakeshore Hospital and Research Centre (LHRC), Kochi, for a duration of 3 years, from December 2019 to December 2022, includes patients diagnosed as BIRADS 4 by mammographic or ultrasonographic or MRI findings. Previous medical records and electronic database were used to collect data. The study includes patients with BIRADS 4 lesions who went through surgery. Predictors such as patient demographics, comorbidities, and imaging features were considered. The rate of malignancy and positive predictive factor of BIRADS 4 lesions associated with carcinoma breast was calculated. A total of 394 lesions were classified as BIRADS 4 according to mammogram, ultrasound, or MRI for a time period of 3 years, from December 2019 to December 2022 at LHRC. Nevertheless, only 383 BIRADS 4 lesions for whom surgery was done were included in the survey. Out of the 383 lesions, the final histopathological report of the biopsied samples revealed 339 lesions to be malignant. Multivariate logistic regression analysis showed that menopausal status, mass lesions and spiculations in mammogram, and ductal abnormality in ultrasonography were greatly linked with carcinoma breast. Invasive breast carcinoma was the most common malignant lesion while fibro-cystic disease remained the predominant benign pathology. This study showed that majority of the BIRADS 4 lesions in LHRC were malignant. Both mammogram and ultrasonography were able to pick up early-stage breast cancer. Tissue diagnosis had higher sensitivity and is useful to avoid surgeries for non-significant lesions which can be followed up. Routine health check-up should be done according to the recent guidelines to detect early-stage breast cancer.