Contrast-enhanced ultrasound to predict malignant upgrading of atypical ductal hyperplasia.

BACKGROUND: A malignancy might be found at surgery in cases of atypical ductal hyperplasia (ADH) diagnosed via US-guided core needle biopsy (CNB). The objective of this study was to investigate the diagnostic performance of contrast-enhanced ultrasound (CEUS) in predicting ADH diagnosed by US-guided CNB that was upgraded to malignancy after surgery. METHODS: In this retrospective study, 110 CNB-diagnosed ADH lesions in 109 consecutive women who underwent US, CEUS, and surgery between June 2018 and June 2023 were included. CEUS was incorporated into US BI-RADS and yielded a CEUS-adjusted BI-RADS. The diagnostic performance of US BI-RADS and CEUS-adjusted BI-RADS for ADH were analyzed and compared. RESULTS: The mean age of the 109 women was 49.7 years ± 11.6 (SD). The upgrade rate of ADH at CNB was 48.2% (53 of 110). The sensitivity, specificity, positive predictive value, and negative predictive value of CEUS for identification of malignant upgrading were 96.2%, 66.7%,72.9%, and 95.0%, respectively, based on BI-RADS category 4B threshold. The two false-negative cases were low-grade ductal carcinoma in situ. Compared with the US, CEUS-adjusted BI-RADS had better specificity for lesions smaller than 2 cm (76.7% vs. 96.7%, P = 0.031). After CEUS, 16 (10 malignant and 6 nonmalignant) of the 45 original US BI-RADS category 4A lesions were up-classified to BI-RADS 4B, and 3 (1 malignant and 2 nonmalignant) of the 41 original US BI-RADS category 4B lesions were down-classified to BI-RADS 4A. CONCLUSIONS: CEUS is helpful in predicting malignant upgrading of ADH, especially for lesions smaller than 2 cm and those classified as BI-RADS 4A and 4B on ultrasound.

Value of automated breast ultrasound in screening: Standalone and as a supplemental to digital breast tomosynthesis.

We aimed to determine the value of standalone and supplemental automated breast ultrasound (ABUS) in detecting cancers in an opportunistic screening setting with digital breast tomosynthesis (DBT) and compare this combined screening method to DBT and ABUS alone in women older than 39 years with BI-RADS B-D density categories. In this prospective opportunistic screening study, 3466 women aged 39 or older with BI-RADS B-D density categories and with a mean age of 50 were included. The screening protocol consisted of DBT mediolateral-oblique views, 2D craniocaudal views, and ABUS with three projections for both breasts. ABUS was evaluated blinded to mammography findings. Statistical analysis evaluated diagnostic performance for DBT, ABUS, and combined workflows. Twenty-nine cancers were screen-detected. ABUS and DBT exhibited the same cancer detection rates (CDR) at 7.5/1000 whereas DBT + ABUS showed 8.4/1000, with ABUS contributing an additional CDR of 0.9/1000. Standalone ABUS outperformed DBT in detecting 12.5% more invasive cancers. DBT displayed better accuracy (95%) compared to ABUS (88%) and combined approach (86%). Sensitivities for DBT and ABUS were the same (84%), with DBT + ABUS showing a higher rate (94%). DBT outperformed ABUS in specificity (95% vs. 88%). DBT + ABUS exhibited a higher recall rate (14.89%) compared to ABUS (12.38%) and DBT (6.03%) (p < .001). Standalone ABUS detected more invasive cancers compared to DBT, with a higher recall rate. The combined approach showed a higher CDR by detecting one additional cancer per thousand.

Revolutionizing breast cancer Ki-67 diagnosis: ultrasound radiomics and fully connected neural networks (FCNN) combination method.

PURPOSE: This study aims to assess the diagnostic value of ultrasound habitat sub-region radiomics feature parameters using a fully connected neural networks (FCNN) combination method L2,1-norm in relation to breast cancer Ki-67 status. METHODS: Ultrasound images from 528 cases of female breast cancer at the Affiliated Hospital of Xiangnan University and 232 cases of female breast cancer at the Affiliated Rehabilitation Hospital of Xiangnan University were selected for this study. We utilized deep learning methods to automatically outline the gross tumor volume and perform habitat clustering. Subsequently, habitat sub-regions were extracted to identify radiomics features and underwent feature engineering using the L1,2-norm. A prediction model for the Ki-67 status of breast cancer patients was then developed using a FCNN. The model's performance was evaluated using accuracy, area under the curve (AUC), specificity (Spe), positive predictive value (PPV), negative predictive value (NPV), Recall, and F1. In addition, calibration curves and clinical decision curves were plotted for the test set to visually assess the predictive accuracy and clinical benefit of the models. RESULT: Based on the feature engineering using the L1,2-norm, a total of 9 core features were identified. The predictive model, constructed by the FCNN model based on these 9 features, achieved the following scores: ACC 0.856, AUC 0.915, Spe 0.843, PPV 0.920, NPV 0.747, Recall 0.974, and F1 0.890. Furthermore, calibration curves and clinical decision curves of the validation set demonstrated a high level of confidence in the model's performance and its clinical benefit. CONCLUSION: Habitat clustering of ultrasound images of breast cancer is effectively supported by the combined implementation of the L1,2-norm and FCNN algorithms, allowing for the accurate classification of the Ki-67 status in breast cancer patients.

Screening Outcomes of Mammography with AI in Dense Breasts: A Comparative Study with Supplemental Screening US.

Background Comparative performance between artificial intelligence (AI) and breast US for women with dense breasts undergoing screening mammography remains unclear. Purpose To compare the performance of mammography alone, mammography with AI, and mammography plus supplemental US for screening women with dense breasts, and to investigate the characteristics of the detected cancers. Materials and Methods A retrospective database search identified consecutive asymptomatic women (≥40 years of age) with dense breasts who underwent mammography plus supplemental whole-breast handheld US from January 2017 to December 2018 at a primary health care center. Sequential reading for mammography alone and mammography with the aid of an AI system was conducted by five breast radiologists, and their recall decisions were recorded. Results of the combined mammography and US examinations were collected from the database. A dedicated breast radiologist reviewed marks for mammography alone or with AI to confirm lesion identification. The reference standard was histologic examination and 1-year follow-up data. The cancer detection rate (CDR) per 1000 screening examinations, sensitivity, specificity, and abnormal interpretation rate (AIR) of mammography alone, mammography with AI, and mammography plus US were compared. Results Among 5707 asymptomatic women (mean age, 52.4 years ± 7.9 [SD]), 33 (0.6%) had cancer (median lesion size, 0.7 cm). Mammography with AI had a higher specificity (95.3% [95% CI: 94.7, 95.8], P = .003) and lower AIR (5.0% [95% CI: 4.5, 5.6], P = .004) than mammography alone (94.3% [95% CI: 93.6, 94.8] and 6.0% [95% CI: 5.4, 6.7], respectively). Mammography plus US had a higher CDR (5.6 vs 3.5 per 1000 examinations, P = .002) and sensitivity (97.0% vs 60.6%, P = .002) but lower specificity (77.6% vs 95.3%, P < .001) and higher AIR (22.9% vs 5.0%, P < .001) than mammography with AI. Supplemental US alone helped detect 12 cancers, mostly stage 0 and I (92%, 11 of 12). Conclusion Although AI improved the specificity of mammography interpretation, mammography plus supplemental US helped detect more node-negative early breast cancers that were undetected using mammography with AI. © RSNA, 2024 Supplemental material is available for this article. See also the editorial by Whitman and Destounis in this issue.

Comparison of Mammography and Mammography with Supplemental Whole-Breast US Tomography for Cancer Detection in Patients with Dense Breasts.

BACKGROUND: Women with dense breasts benefit from supplemental cancer screening with US, but US has low specificity. PURPOSE: To evaluate the performance of breast US tomography (UST) combined with full-field digital mammography (FFDM) compared with FFDM alone for breast cancer screening in women with dense breasts. MATERIALS AND METHODS: This retrospective multireader multicase study included women with dense breasts who underwent FFDM and UST at 10 centers between August 2017 and October 2019 as part of a prospective case collection registry. All patients in the registry with cancer were included; patients with benign biopsy or negative follow-up imaging findings were randomly selected for inclusion. Thirty-two Mammography Quality Standards Act-qualified radiologists independently evaluated FFDM followed immediately by FFDM plus UST for suspicious findings and assigned a Breast Imaging Reporting and Data System (BI-RADS) category. The superiority of FFDM plus UST versus FFDM alone for cancer detection (assessed with area under the receiver operating characteristic curve [AUC]), BI-RADS 4 sensitivity, and BI-RADS 3 sensitivity and specificity were evaluated using the two-sided significance level of α = .05. Noninferiority of BI-RADS 4 specificity was evaluated at the one-sided significance level of α = .025 with a -10% margin. RESULTS: Among 140 women (mean age, 56 years ±10 [SD]; 36 with cancer, 104 without), FFDM plus UST achieved superior performance compared with FFDM alone (AUC, 0.60 [95% CI: 0.51, 0.69] vs 0.54 [95% CI: 0.45, 0.64]; P = .03). For FFDM plus UST versus FFDM alone, BI-RADS 4 mean sensitivity was superior (37% [428 of 1152] vs 30% [343 of 1152]; P = .03) and BI-RADS 4 mean specificity was noninferior (82% [2741 of 3328] vs 88% [2916 of 3328]; P = .004). For FFDM plus UST versus FFDM, no difference in BI-RADS 3 mean sensitivity was observed (40% [461 of 1152] vs 33% [385 of 1152]; P = .08), but BI-RADS 3 mean specificity was superior (75% [2491 of 3328] vs 69% [2299 of 3328]; P = .04). CONCLUSION: In women with dense breasts, FFDM plus UST improved cancer detection by radiologists versus FFDM alone. Clinical trial registration nos. NCT03257839 and NCT04260620 Published under a CC BY 4.0 license. Supplemental material is available for this article. See also the editorial by Mann in this issue.

Performance of Supplemental US Screening in Women with Dense Breasts and Varying Breast Cancer Risk: Results from the Breast Cancer Surveillance Consortium.

Background It is unclear whether breast US screening outcomes for women with dense breasts vary with levels of breast cancer risk. Purpose To evaluate US screening outcomes for female patients with dense breasts and different estimated breast cancer risk levels. Materials and Methods This retrospective observational study used data from US screening examinations in female patients with heterogeneously or extremely dense breasts conducted from January 2014 to October 2020 at 24 radiology facilities within three Breast Cancer Surveillance Consortium (BCSC) registries. The primary outcomes were the cancer detection rate, false-positive biopsy recommendation rate, and positive predictive value of biopsies performed (PPV3). Risk classification of participants was performed using established BCSC risk prediction models of estimated 6-year advanced breast cancer risk and 5-year invasive breast cancer risk. Differences in high- versus low- or average-risk categories were assessed using a generalized linear model. Results In total, 34 791 US screening examinations from 26 489 female patients (mean age at screening, 53.9 years ± 9.0 [SD]) were included. The overall cancer detection rate per 1000 examinations was 2.0 (95% CI: 1.6, 2.4) and was higher in patients with high versus low or average risk of 6-year advanced breast cancer (5.5 [95% CI: 3.5, 8.6] vs 1.3 [95% CI: 1.0, 1.8], respectively; P = .003). The overall false-positive biopsy recommendation rate per 1000 examinations was 29.6 (95% CI: 22.6, 38.6) and was higher in patients with high versus low or average 6-year advanced breast cancer risk (37.0 [95% CI: 28.2, 48.4] vs 28.1 [95% CI: 20.9, 37.8], respectively; P = .04). The overall PPV3 was 6.9% (67 of 975; 95% CI: 5.3, 8.9) and was higher in patients with high versus low or average 6-year advanced cancer risk (15.0% [15 of 100; 95% CI: 9.9, 22.2] vs 4.9% [30 of 615; 95% CI: 3.3, 7.2]; P = .01). Similar patterns in outcomes were observed by 5-year invasive breast cancer risk. Conclusion The cancer detection rate and PPV3 of supplemental US screening increased with the estimated risk of advanced and invasive breast cancer. © RSNA, 2024 Supplemental material is available for this article. See also the editorial by Helbich and Kapetas in this issue.

BI-RADS Category Assignments by GPT-3.5, GPT-4, and Google Bard: A Multilanguage Study.

Background The performance of publicly available large language models (LLMs) remains unclear for complex clinical tasks. Purpose To evaluate the agreement between human readers and LLMs for Breast Imaging Reporting and Data System (BI-RADS) categories assigned based on breast imaging reports written in three languages and to assess the impact of discordant category assignments on clinical management. Materials and Methods This retrospective study included reports for women who underwent MRI, mammography, and/or US for breast cancer screening or diagnostic purposes at three referral centers. Reports with findings categorized as BI-RADS 1-5 and written in Italian, English, or Dutch were collected between January 2000 and October 2023. Board-certified breast radiologists and the LLMs GPT-3.5 and GPT-4 (OpenAI) and Bard, now called Gemini (Google), assigned BI-RADS categories using only the findings described by the original radiologists. Agreement between human readers and LLMs for BI-RADS categories was assessed using the Gwet agreement coefficient (AC1 value). Frequencies were calculated for changes in BI-RADS category assignments that would affect clinical management (ie, BI-RADS 0 vs BI-RADS 1 or 2 vs BI-RADS 3 vs BI-RADS 4 or 5) and compared using the McNemar test. Results Across 2400 reports, agreement between the original and reviewing radiologists was almost perfect (AC1 = 0.91), while agreement between the original radiologists and GPT-4, GPT-3.5, and Bard was moderate (AC1 = 0.52, 0.48, and 0.42, respectively). Across human readers and LLMs, differences were observed in the frequency of BI-RADS category upgrades or downgrades that would result in changed clinical management (118 of 2400 [4.9%] for human readers, 611 of 2400 [25.5%] for Bard, 573 of 2400 [23.9%] for GPT-3.5, and 435 of 2400 [18.1%] for GPT-4; P < .001) and that would negatively impact clinical management (37 of 2400 [1.5%] for human readers, 435 of 2400 [18.1%] for Bard, 344 of 2400 [14.3%] for GPT-3.5, and 255 of 2400 [10.6%] for GPT-4; P < .001). Conclusion LLMs achieved moderate agreement with human reader-assigned BI-RADS categories across reports written in three languages but also yielded a high percentage of discordant BI-RADS categories that would negatively impact clinical management. © RSNA, 2024 Supplemental material is available for this article.

Association of clinicopathologic and sonographic features with stromal tumor-infiltrating lymphocytes in triple-negative breast cancer.

BACKGROUND: Increased level of stromal tumor-infiltrating lymphocytes (sTILs) are associated with therapeutic outcomes and prognosis in triple-negative breast cancer (TNBC). This study aimed to investigate the associations of clinicopathologic and sonographic features with sTILs level in TNBC. METHODS: This study included invasive TNBC patients with postoperative evaluation of sTILs after surgical resection. Tumor shape, margin, orientation, echo pattern, posterior features, calcification, and vascularity were retrospectively evaluated. The patients were categorized into high-sTILs (≥ 20%) and low-sTILs (< 20%) level groups. Chi-square or Fisher's exact tests were used to assess the association of clinicopathologic and sonographic features with sTILs level. RESULTS: The 171 patients (mean ± SD age, 54.7 ± 10.3 years [range, 22‒87 years]) included 58.5% (100/171) with low-sTILs level and 41.5% (71/171) with high-sTILs level. The TNBC tumors with high-sTILs level were more likely to be no special type invasive carcinoma (p = 0.008), higher histologic grade (p = 0.029), higher Ki-67 proliferation rate (all p < 0.05), and lower frequency of associated DCIS component (p = 0.026). In addition, the TNBC tumors with high-sTILs level were more likely to be an oval or round shape (p = 0.001), parallel orientation (p = 0.011), circumscribed or micro-lobulated margins (p < 0.001), complex cystic and solid echo patterns (p = 0.001), posterior enhancement (p = 0.002), and less likely to have a heterogeneous pattern (p = 0.001) and no posterior features (p = 0.002). CONCLUSIONS: This preliminary study showed that preoperative sonographic characteristics could be helpful in distinguishing high-sTILs from low-sTILs in TNBC patients.

Diagnostic value of applying preoperative breast ultrasound and clinicopathologic features to predict axillary lymph node burden in early invasive breast cancer: a study of 1247 patients.

BACKGROUND: Since the Z0011 trial, the assessment of axillary lymph node status has been redirected from the previous assessment of the occurrence of lymph node metastasis alone to the assessment of the degree of lymph node loading. Our aim was to apply preoperative breast ultrasound and clinicopathological features to predict the diagnostic value of axillary lymph node load in early invasive breast cancer. METHODS: The 1247 lesions were divided into a high lymph node burden group and a limited lymph node burden group according to axillary lymph node status. Univariate and multifactorial analyses were used to predict the differences in clinicopathological characteristics and breast ultrasound characteristics between the two groups with high and limited lymph node burden. Pathological findings were used as the gold standard. RESULTS: Univariate analysis showed significant differences in ki-67, maximum diameter (MD), lesion distance from the nipple, lesion distance from the skin, MS, and some characteristic ultrasound features (P < 0.05). In multifactorial analysis, the ultrasound features of breast tumors that were associated with a high lymph node burden at the axilla included MD (odds ratio [OR], 1.043; P < 0.001), shape (OR, 2.422; P = 0.0018), hyperechoic halo (OR, 2.546; P < 0.001), shadowing in posterior features (OR, 2.155; P = 0.007), and suspicious lymph nodes on axillary ultrasound (OR, 1.418; P = 0.031). The five risk factors were used to build the predictive model, and it achieved an area under the receiver operating characteristic (ROC) curve (AUC) of 0.702. CONCLUSION: Breast ultrasound features and clinicopathological features are better predictors of high lymph node burden in early invasive breast cancer, and this prediction helps to develop more effective treatment plans.

Spatiotemporal analysis of contrast-enhanced ultrasound for differentiating between malignant and benign breast lesions.

OBJECTIVES: The aim of this study was to apply spatiotemporal analysis of contrast-enhanced ultrasound (CEUS) loops to quantify the enhancement heterogeneity for improving the differentiation between benign and malignant breast lesions. MATERIALS AND METHODS: This retrospective study included 120 women (age range, 18-82 years; mean, 52 years) scheduled for ultrasound-guided biopsy. With the aid of brightness-mode images, the border of each breast lesion was delineated in the CEUS images. Based on visual evaluation and quantitative metrics, the breast lesions were categorized into four grades of different levels of contrast enhancement. Grade-1 (hyper-enhanced) and grade-2 (partly-enhanced) breast lesions were included in the analysis. Four parameters reflecting enhancement heterogeneity were estimated by spatiotemporal analysis of neighboring time-intensity curves (TICs). By setting the threshold on mean parameter, the diagnostic performance of the four parameters for differentiating benign and malignant lesions was evaluated. RESULTS: Sixty-four of the 120 patients were categorized as grade 1 or 2 and used for estimating the four parameters. At the pixel level, mutual information and conditional entropy present significantly different values between the benign and malignant lesions (p < 0.001 in patients of grade 1, p = 0.002 in patients of grade 1 or 2). For the classification of breast lesions, mutual information produces the best diagnostic performance (AUC = 0.893 in patients of grade 1, AUC = 0.848 in patients of grade 1 or 2). CONCLUSIONS: The proposed spatiotemporal analysis for assessing the enhancement heterogeneity shows promising results to aid in the diagnosis of breast cancer by CEUS. CLINICAL RELEVANCE STATEMENT: The proposed spatiotemporal method can be developed as a standardized software to automatically quantify the enhancement heterogeneity of breast cancer on CEUS, possibly leading to the improved diagnostic accuracy of differentiation between benign and malignant lesions. KEY POINTS: • Advanced spatiotemporal analysis of ultrasound contrast-enhanced loops for aiding the differentiation of malignant or benign breast lesions. • Four parameters reflecting the enhancement heterogeneity were estimated in the hyper- and partly-enhanced breast lesions by analyzing the neighboring pixel-level time-intensity curves. • For the classification of hyper-enhanced breast lesions, mutual information produces the best diagnostic performance (AUC = 0.893).

A conditional inference tree model for predicting cancer risk of non-mass lesions detected on breast ultrasound.

OBJECTIVES: To generate and validate a prediction model based on imaging features for cancer risk of non-mass lesions (NMLs) detected on breast ultrasound (US). METHODS: In this single-center study, consecutive women with 503 NMLs detected on breast US between 2012 and 2019 were retrospectively identified. The lesions were randomly assigned to the training or testing dataset with a 70/30 split. Age, symptoms, lesion size, and US features were collected. Multivariate analyses were employed to identify risk factors associated with malignancy. The predictive model was developed by using conditional inference trees (CTREE). RESULTS: There were 498 patients (50.9 ± 13.29 years; range, 22-88 years) with 503 NMLs with histopathologic results or > 2-year follow-up, including 224 (44.5%) benign and 279 (55.5%) malignant lesions. At multivariate analysis, age (odds ratio (OR) = 1.08, 95% confidence interval (CI), 1.06-1.11, p < 0.001), NMLs with focal mass effect (OR = 3.03, 95% CI, 1.59-5.81, p = 0.001), indistinct glandular-fat interface (GFI) (OR = 4.23, 95% CI, 2.31-7.73, p < 0.001), geographic (OR = 3.47, 95% CI, 1.20-10.8, p = 0.022) and mottled (OR = 3.67, 95% CI, 1.32-10.21, p = 0.013) patterns, and calcifications (OR = 2.15, 95% CI, 1.16-4.01, p = 0.016) were associated with malignancy. The GFI status, architectural patterns, general morphology, and calcifications were consistently identified as the strongest US predictors of malignancy using CTREE analysis. Based on these factors, individuals were stratified into six risk groups. The predictive model showed an area under the curve of 0.797 in the testing dataset. CONCLUSION: The CTREE model efficiently aids in interpreting and managing ultrasound-detected breast NMLs, overcoming BI-RADS limitations by refining cancer risk stratification. CLINICAL RELEVANCE STATEMENT: The CTREE model allows for the reclassification of BI-RADS categories into subgroups with varying malignancy probabilities, thus providing a valuable enhancement to the BI-RADS assessment for the diagnosis of ultrasound-detected NMLs, with the potential to minimize unnecessary biopsies. KEY POINTS: • The indistinct glandular-fat interface (GFI) status, NML with focal mass effect, geographic or mottled patterns, and calcifications are the strongest imaging predictors of malignant non-mass lesions (NMLs) detected on breast US. • A practical system has been created to categorize NMLs found in breast US; each classification is associated with a degree of diagnostic certainty. • The model may contribute to patient stratification by determining the relative likelihood of malignancy and thus support clinical decision-making and evidence-based management.

Diagnostic performance of 3D automated breast ultrasound (3D-ABUS) in a clinical screening setting-a retrospective study.

OBJECTIVES: To assess the diagnostic performance of 3D automated breast ultrasound (3D-ABUS) in breast cancer screening in a clinical setting. MATERIALS AND METHODS: All patients who had 3D-ABUS between January 2014 and January 2022 for screening were included in this retrospective study. The images were reported by 1 of 6 breast radiologists based on the Breast Imaging Reporting and Data Systems (BI-RADS). The 3D-ABUS was reviewed together with the digital breast tomosynthesis (DBT). Recall rate, biopsy rate, positive predictive value (PPV) and cancer detection yield were calculated. RESULTS: In total, 3616 studies were performed in 1555 women (breast density C/D 95.5% (n = 3455/3616), breast density A/B 4.0% (n = 144/3616), density unknown (0.5% (n = 17/3616)). A total of 259 lesions were detected on 3D-ABUS (87.6% (n = 227/259) masses and 12.4% (n = 32/259) architectural distortions). The recall rate was 5.2% (n = 188/3616) (CI 4.5-6.0%) with only 36.7% (n = 69/188) cases recalled to another date. Moreover, recall declined over time. There were 3.4% (n = 123/3616) biopsies performed, with 52.8% (n = 65/123) biopsies due to an abnormality detected in 3D-ABUS alone. Ten of 65 lesions were malignant, resulting in a positive predictive value (PPV) of 15.4% (n = 10/65) (CI 7.6-26.5%)). The cancer detection yield of 3D-ABUS is 2.77 per 1000 screening tests (CI 1.30-5.1). CONCLUSION: The cancer detection yield of 3D-ABUS in a real clinical screening setting is comparable to the results reported in previous prospective studies, with lower recall and biopsy rates. 3D-ABUS also may be an alternative for screening when mammography is not possible or declined. CLINICAL RELEVANCE STATEMENT: 3D automated breast ultrasound screening performance in a clinical setting is comparable to previous prospective studies, with better recall and biopsy rates. KEY POINTS: • 3D automated breast ultrasound is a reliable and reproducible tool that provides a three-dimensional representation of the breast and allows image visualisation in axial, coronal and sagittal. • The diagnostic performance of 3D automated breast ultrasound in a real clinical setting is comparable to its performance in previously published prospective studies, with improved recall and biopsy rates. • 3D automated breast ultrasound is a useful adjunct to mammography in dense breasts and may be an alternative for screening when mammography is not possible or declined.

Imaging the dense breast.

The sensitivity of mammography reduces as breast density increases, which impacts breast screening and locoregional staging in breast cancer. Supplementary imaging with other modalities can offer improved cancer detection, but this often comes at the cost of more false positives. Magnetic resonance imaging and contrast-enhanced mammography, which assess tumour enhancement following contrast administration, are more sensitive than digital breast tomosynthesis and ultrasound, which predominantly rely on the assessment of tumour morphology.

A neural network with a human learning paradigm for breast fibroadenoma segmentation in sonography.

BACKGROUND: Breast fibroadenoma poses a significant health concern, particularly for young women. Computer-aided diagnosis has emerged as an effective and efficient method for the early and accurate detection of various solid tumors. Automatic segmentation of the breast fibroadenoma is important and potentially reduces unnecessary biopsies, but challenging due to the low image quality and presence of various artifacts in sonography. METHODS: Human learning involves modularizing complete information and then integrating it through dense contextual connections in an intuitive and efficient way. Here, a human learning paradigm was introduced to guide the neural network by using two consecutive phases: the feature fragmentation stage and the information aggregation stage. To optimize this paradigm, three fragmentation attention mechanisms and information aggregation mechanisms were adapted according to the characteristics of sonography. The evaluation was conducted using a local dataset comprising 600 breast ultrasound images from 30 patients at Suining Central Hospital in China. Additionally, a public dataset consisting of 246 breast ultrasound images from Dataset_BUSI and DatasetB was used to further validate the robustness of the proposed network. Segmentation performance and inference speed were assessed by Dice similarity coefficient (DSC), Hausdorff distance (HD), and training time and then compared with those of the baseline model (TransUNet) and other state-of-the-art methods. RESULTS: Most models guided by the human learning paradigm demonstrated improved segmentation on the local dataset with the best one (incorporating C3ECA and LogSparse Attention modules) outperforming the baseline model by 0.76% in DSC and 3.14 mm in HD and reducing the training time by 31.25%. Its robustness and efficiency on the public dataset are also confirmed, surpassing TransUNet by 0.42% in DSC and 5.13 mm in HD. CONCLUSIONS: Our proposed human learning paradigm has demonstrated the superiority and efficiency of ultrasound breast fibroadenoma segmentation across both public and local datasets. This intuitive and efficient learning paradigm as the core of neural networks holds immense potential in medical image processing.

Breast cancer screening and early diagnosis in China: a systematic review and meta-analysis on 10.72 million women.

BACKGROUND: The incidence of breast cancer among Chinese women has gradually increased in recent years. This study aims to analyze the situation of breast cancer screening programs in China and compare the cancer detection rates (CDRs), early-stage cancer detection rates (ECDRs), and the proportions of early-stage cancer among different programs. METHODS: We conducted a systematic review and meta-analysis of studies in multiple literature databases. Studies that were published between January 1, 2010 and June 30, 2023 were retrieved. A random effects model was employed to pool the single group rate, and subgroup analyses were carried out based on screening model, time, process, age, population, and follow-up method. RESULTS: A total of 35 studies, including 47 databases, satisfied the inclusion criteria. Compared with opportunistic screening, the CDR (1.32‰, 95% CI: 1.10‰-1.56‰) and the ECDR (0.82‰, 95% CI: 0.66‰-0.99‰) were lower for population screening, but the proportion of early-stage breast cancer (80.17%, 95% CI: 71.40%-87.83%) was higher. In subgroup analysis, the CDR of population screening was higher in the urban group (2.28‰, 95% CI: 1.70‰-2.94‰), in the breast ultrasonography (BUS) in parallel with mammography (MAM) group (3.29‰, 95% CI: 2.48‰-4.21‰), and in the second screening follow-up group (2.47‰, 95% CI: 1.64‰-3.47‰), and the proportion of early-stage breast cancer was 85.70% (95% CI: 68.73%-97.29%), 88.18% (95% CI: 84.53%-91.46%), and 90.05% (95% CI: 84.07%-94.95%), respectively. CONCLUSION: There were significant differences between opportunistic and population screening programs. The results of these population screening studies were influenced by the screening process, age, population, and follow-up method. In the future, China should carry out more high-quality and systematic population-based screening programs to improve screening coverage and service.

Differential diagnosis of breast mucinous carcinoma with an oval shape from fibroadenoma based on ultrasonographic features.

BACKGROUND: Approximately 50% of breast mucinous carcinomas (MCs) are oval and have the possibility of being misdiagnosed as fibroadenomas (FAs). We aimed to identify the key features that can help differentiate breast MC with an oval shape from FA on ultrasonography (US). METHODS: Seventy-six MCs from 71 consecutive patients and 50 FAs with an oval shape from 50 consecutive patients were included in our study. All lesions pathologically diagnosed. According to the Breast Imaging Reporting and Data System (BI-RADS), first, the ultrasonographic features of the MCs and FAs were recorded and a final category was assessed. Then, the differences in ultrasonographic characteristics between category 4 A (low-risk group) and category 4B-5 (medium-high- risk group) MCs were identified. Finally, other ultrasonographic features of MC and FA both with an oval shape were compared to determine the key factors for differential diagnosis. The Mann-Whitney test, χ2 test or Fisher's exact test was used to compare data between groups. RESULTS: MCs with an oval shape (81.2%) and a circumscribed margin (25%) on US were more commonly assessed in the low-risk group (BI-RADS 4 A) than in the medium-high-risk group (BI-RADS 4B-5) (20%, p < 0.001 and 0%, p = 0.001, respectively). Compared with those with FA, patients with MC were older, and tended to have masses with non-hypoechoic patterns, not circumscribed margins, and a posterior echo enhancement on US (p < 0.001, p < 0.001, and p = 0.003, respectively). CONCLUSION: The oval shape was the main reason for the underestimation of MCs. On US, an oval mass found in the breast of women of older age with non-hypoechoic patterns, not circumscribed margins, and a posterior echo enhancement was associated with an increased risk of being an MC, and should be subjected to active biopsy.

Use of ultrasound imaging Omics in predicting molecular typing and assessing the risk of postoperative recurrence in breast cancer.

BACKGROUND: The aim of this study is to assess the efficacy of a multiparametric ultrasound imaging omics model in predicting the risk of postoperative recurrence and molecular typing of breast cancer. METHODS: A retrospective analysis was conducted on 534 female patients diagnosed with breast cancer through preoperative ultrasonography and pathology, from January 2018 to June 2023 at the Affiliated Cancer Hospital of Xinjiang Medical University. Univariate analysis and multifactorial logistic regression modeling were used to identify independent risk factors associated with clinical characteristics. The PyRadiomics package was used to delineate the region of interest in selected ultrasound images and extract radiomic features. Subsequently, radiomic scores were established through Least Absolute Shrinkage and Selection Operator (LASSO) regression and Support Vector Machine (SVM) methods. The predictive performance of the model was assessed using the receiver operating characteristic (ROC) curve, and the area under the curve (AUC) was calculated. Evaluation of diagnostic efficacy and clinical practicability was conducted through calibration curves and decision curves. RESULTS: In the training set, the AUC values for the postoperative recurrence risk prediction model were 0.9489, and for the validation set, they were 0.8491. Regarding the molecular typing prediction model, the AUC values in the training set and validation set were 0.93 and 0.92 for the HER-2 overexpression phenotype, 0.94 and 0.74 for the TNBC phenotype, 1.00 and 0.97 for the luminal A phenotype, and 1.00 and 0.89 for the luminal B phenotype, respectively. Based on a comprehensive analysis of calibration and decision curves, it was established that the model exhibits strong predictive performance and clinical practicability. CONCLUSION: The use of multiparametric ultrasound imaging omics proves to be of significant value in predicting both the risk of postoperative recurrence and molecular typing in breast cancer. This non-invasive approach offers crucial guidance for the diagnosis and treatment of the condition.

Ultrasound-based deep learning radiomics nomogram for differentiating mass mastitis from invasive breast cancer.

BACKGROUND: The purpose of this study is to develop and validate the potential value of the deep learning radiomics nomogram (DLRN) based on ultrasound to differentiate mass mastitis (MM) and invasive breast cancer (IBC). METHODS: 50 cases of MM and 180 cases of IBC with ultrasound Breast Imaging Reporting and Data System 4 category were recruited (training cohort, n = 161, validation cohort, n = 69). Based on PyRadiomics and ResNet50 extractors, radiomics and deep learning features were extracted, respectively. Based on supervised machine learning methods such as logistic regression, random forest, and support vector machine, as well as unsupervised machine learning methods using K-means clustering analysis, the differences in features between MM and IBC were analyzed to develop DLRN. The performance of DLRN had been evaluated by receiver operating characteristic curve, calibration, and clinical practicality. RESULTS: Supervised machine learning results showed that compared with radiomics models, especially random forest models, deep learning models were better at recognizing MM and IBC. The area under the curve (AUC) of the validation cohort was 0.84, the accuracy was 0.83, the sensitivity was 0.73, and the specificity was 0.83. Compared to radiomics or deep learning models, DLRN even further improved discrimination ability (AUC of 0.90 and 0.90, accuracy of 0.83 and 0.88 for training and validation cohorts), which had better clinical benefits and good calibratability. In addition, the information heterogeneity of deep learning features in MM and IBC was validated again through unsupervised machine learning clustering analysis, indicating that MM had a unique features phenotype. CONCLUSION: The DLRN developed based on radiomics and deep learning features of ultrasound images has potential clinical value in effectively distinguishing between MM and IBC. DLRN breaks through visual limitations and quantifies more image information related to MM based on computers, further utilizing machine learning to effectively utilize this information for clinical decision-making. As DLRN becomes an autonomous screening system, it will improve the recognition rate of MM in grassroots hospitals and reduce the possibility of incorrect treatment and overtreatment.

Spatial and geometric learning for classification of breast tumors from multi-center ultrasound images: a hybrid learning approach.

BACKGROUND: Breast cancer is the most common cancer among women, and ultrasound is a usual tool for early screening. Nowadays, deep learning technique is applied as an auxiliary tool to provide the predictive results for doctors to decide whether to make further examinations or treatments. This study aimed to develop a hybrid learning approach for breast ultrasound classification by extracting more potential features from local and multi-center ultrasound data. METHODS: We proposed a hybrid learning approach to classify the breast tumors into benign and malignant. Three multi-center datasets (BUSI, BUS, OASBUD) were used to pretrain a model by federated learning, then every dataset was fine-tuned at local. The proposed model consisted of a convolutional neural network (CNN) and a graph neural network (GNN), aiming to extract features from images at a spatial level and from graphs at a geometric level. The input images are small-sized and free from pixel-level labels, and the input graphs are generated automatically in an unsupervised manner, which saves the costs of labor and memory space. RESULTS: The classification AUCROC of our proposed method is 0.911, 0.871 and 0.767 for BUSI, BUS and OASBUD. The balanced accuracy is 87.6%, 85.2% and 61.4% respectively. The results show that our method outperforms conventional methods. CONCLUSIONS: Our hybrid approach can learn the inter-feature among multi-center data and the intra-feature of local data. It shows potential in aiding doctors for breast tumor classification in ultrasound at an early stage.

Detection and diagnosis of automated breast ultrasound in patients with BI-RADS category 4 microcalcifications: a retrospective study.

BACKGROUND: Automated Breast Ultrasound (AB US) has shown good application value and prospects in breast disease screening and diagnosis. The aim of the study was to explore the ability of AB US to detect and diagnose mammographically Breast Imaging Reporting and Data System (BI-RADS) category 4 microcalcifications. METHODS: 575 pathologically confirmed mammographically BI-RADS category 4 microcalcifications from January 2017 to June 2021 were included. All patients also completed AB US examinations. Based on the final pathological results, analyzed and summarized the AB US image features, and compared the evaluation results with mammography, to explore the detection and diagnostic ability of AB US for these suspicious microcalcifications. RESULTS: 250 were finally confirmed as malignant and 325 were benign. Mammographic findings including microcalcifications morphology (61/80 with amorphous, coarse heterogeneous and fine pleomorphic, 13/14 with fine-linear or branching), calcification distribution (189/346 with grouped, 40/67 with linear and segmental), associated features (70/96 with asymmetric shadow), higher BI-RADS category with 4B (88/120) and 4 C (73/38) showed higher incidence in malignant lesions, and were the independent factors associated with malignant microcalcifications. 477 (477/575, 83.0%) microcalcifications were detected by AB US, including 223 malignant and 254 benign, with a significantly higher detection rate for malignant lesions (x2 = 12.20, P < 0.001). Logistic regression analysis showed microcalcifications with architectural distortion (odds ratio [OR] = 0.30, P = 0.014), with amorphous, coarse heterogeneous and fine pleomorphic morphology (OR = 3.15, P = 0.037), grouped (OR = 1.90, P = 0.017), liner and segmental distribution (OR = 8.93, P = 0.004) were the independent factors which could affect the detectability of AB US for microcalcifications. In AB US, malignant calcification was more frequent in a mass (104/154) or intraductal (20/32), and with ductal changes (30/41) or architectural distortion (58/68), especially with the both (12/12). BI-RADS category results also showed that AB US had higher sensitivity to malignant calcification than mammography (64.8% vs. 46.8%). CONCLUSIONS: AB US has good detectability for mammographically BI-RADS category 4 microcalcifications, especially for malignant lesions. Malignant calcification is more common in a mass and intraductal in AB US, and tend to associated with architectural distortion or duct changes. Also, AB US has higher sensitivity than mammography to malignant microcalcification, which is expected to become an effective supplementary examination method for breast microcalcifications, especially in dense breasts.