Federated learning (FL) has emerged as a promising paradigm for collaboratively training deep learning models across institutions without exchanging sensitive medical data. However, its effectiveness is often hindered by limited data availability and non-independent, identically distributed data across participating clients, which can degrade model performance and generalization. To address these challenges, we propose a generative AI based data augmentation framework that integrates synthetic image sharing into the federated training process for breast cancer diagnosis via ultrasound images. Specifically, we train two simple class-specific Deep Convolutional Generative Adversarial Networks: one for benign and one for malignant lesions. We then simulate a realistic FL setting using three publicly available breast ultrasound image datasets: BUSI, BUS-BRA, and UDIAT. FedAvg and FedProx are adopted as baseline FL algorithms. Experimental results show that incorporating a suitable number of synthetic images improved the average AUC from 0.9206 to 0.9237 for FedAvg and from 0.9429 to 0.9538 for FedProx. We also note that excessive use of synthetic data reduced performance, underscoring the importance of maintaining a balanced ratio of real and synthetic samples. Our findings highlight the potential of generative AI based data augmentation to enhance FL results in the breast ultrasound image classification task.

To investigate automated breast ultrasound (ABUS) features affecting the use of Multiview convolutional neural network (CNN) for breast lesions according to radiologists' experience. A total of 656 breast lesions (152 malignant and 504 benign lesions) were included and reviewed by six radiologists for background echotexture, glandular tissue component (GTC), and lesion type and size without as well as with Multiview CNN. The sensitivity, specificity, and the area under the receiver operating curve (AUC) for ABUS features were compared between two sessions according to radiologists' experience. Radiology residents showed significant AUC improvement with the Multiview CNN for mass (0.81 to 0.91, P=0.003) and non-mass lesions (0.56 to 0.90, P=0.007), all background echotextures (homogeneous-fat: 0.84 to 0.94, P=0.04; homogeneous-fibroglandular: 0.85 to 0.93, P=0.01; heterogeneous: 0.68 to 0.88, P=0.002), all GTC levels (minimal: 0.86 to 0.93, P=0.001; mild: 0.82 to 0.94, P=0.003; moderate: 0.75 to 0.88, P=0.01; marked: 0.68 to 0.89, P<0.001), and lesions ≤10mm (≤5 mm: 0.69 to 0.86, P<0.001; 6-10 mm: 0.83 to 0.92, P<0.001). Breast specialists showed significant AUC improvement with the Multiview CNN in heterogeneous echotexture (0.90 to 0.95, P=0.03), marked GTC (0.88 to 0.95, P<0.001), and lesions ≤10mm (≤5 mm: 0.89 to 0.93, P=0.02; 6-10 mm: 0.95 to 0.98, P=0.01). With the Multiview CNN, the performance of ABUS in radiology residents was improved regardless of lesion type, background echotexture, or GTC. For breast lesions smaller than 10 mm, both radiology residents and breast specialists showed better performance of ABUS.

<b>Aims/Background</b> Breast nodules are highly prevalent among women, and ultrasound is a widely used screening tool. However, single ultrasound examinations often result in high false-positive rates, leading to unnecessary biopsies. Artificial intelligence (AI) has demonstrated the potential to improve diagnostic accuracy, reducing misdiagnosis and minimising inter-observer variability. This study developed a deep learning-based AI model to evaluate its clinical utility in assisting sonographers with the Breast Imaging Reporting and Data System (BI-RADS) classification of breast nodules. <b>Methods</b> A retrospective analysis was conducted on 558 patients with breast nodules classified as BI-RADS categories 3 to 5, confirmed through pathological examination at The People's Hospital of Pingyang County between December 2019 and December 2023. The image dataset was divided into a training set, validation set, and test set, and a convolutional neural network (CNN) was used to construct a deep learning-based AI model. Patients underwent ultrasound examination and AI-assisted diagnosis. The receiver operating characteristic (ROC) curve was used to analyse the performance of the AI model, physician adjudication results, and the diagnostic efficacy of physicians before and after AI model assistance. Cohen's weighted Kappa coefficient was used to assess the consistency of BI-RADS classification among five ultrasound physicians before and after AI model assistance. Additionally, statistical analyses were performed to evaluate changes in BI-RADS classification results before and after AI model assistance for each physician. <b>Results</b> According to pathological examination, 765 of the 1026 breast nodules were benign, while 261 were malignant. The sensitivity, specificity, and accuracy of routine ultrasonography in diagnosing benign and malignant nodules were 80.85%, 91.59%, and 88.31%, respectively. In comparison, the AI system achieved a sensitivity of 89.36%, specificity of 92.52%, and accuracy of 91.56%. Furthermore, AI model assistance significantly improved the consistency of physicians' BI-RADS classification (<i>p</i> < 0.001). <b>Conclusion</b> A deep learning-based AI model constructed using ultrasound images can enhance the differentiation between benign and malignant breast nodules and improve classification accuracy, thereby reducing the incidence of missed and misdiagnoses.

<i>"Just Accepted" papers have undergone full peer review and have been accepted for publication in <i>Radiology: Artificial Intelligence</i>. This article will undergo copyediting, layout, and proof review before it is published in its final version. Please note that during production of the final copyedited article, errors may be discovered which could affect the content.</i> Purpose To develop and evaluate an open-source deep learning model for detection and localization of breast cancer on MRI. Materials and Methods In this retrospective study, a deep learning model for breast cancer detection and localization was trained on the largest breast MRI dataset to date. Data included all breast MRIs conducted at a tertiary cancer center in the United States between 2002 and 2019. The model was validated on sagittal MRIs from the primary site (<i>n</i> = 6,615 breasts). Generalizability was assessed by evaluating model performance on axial data from the primary site (<i>n</i> = 7,058 breasts) and a second clinical site (<i>n</i> = 1,840 breasts). Results The primary site dataset included 30,672 sagittal MRI examinations (52,598 breasts) from 9,986 female patients (mean [SD] age, 53 [11] years). The model achieved an area under the receiver operating characteristic curve (AUC) of 0.95 for detecting cancer in the primary site. At 90% specificity (5717/6353), model sensitivity was 83% (217/262), which was comparable to historical performance data for radiologists. The model generalized well to axial examinations, achieving an AUC of 0.92 on data from the same clinical site and 0.92 on data from a secondary site. The model accurately located the tumor in 88.5% (232/262) of sagittal images, 92.8% (272/293) of axial images from the primary site, and 87.7% (807/920) of secondary site axial images. Conclusion The model demonstrated state-of-the-art performance on breast cancer detection. Code and weights are openly available to stimulate further development and validation. ©RSNA, 2025.

To assess whether the diagnostic performance of a commercial artificial intelligence (AI) algorithm for mammography differs between breast-level and lesion-level interpretations and to compare performance to a large population of specialised human readers. We retrospectively analysed 1200 mammograms from the NHS breast cancer screening programme using a commercial AI algorithm and assessments from 1258 trained human readers from the Personal Performance in Mammographic Screening (PERFORMS) external quality assurance programme. For breasts containing pathologically confirmed malignancies, a breast and lesion-level analysis was performed. The latter considered the locations of marked regions of interest for AI and humans. The highest score per lesion was recorded. For non-malignant breasts, a breast-level analysis recorded the highest score per breast. Area under the curve (AUC), sensitivity and specificity were calculated at the developer's recommended threshold for recall. The study was designed to detect a medium-sized effect (odds ratio 3.5 or 0.29) for sensitivity. The test set contained 882 non-malignant (73%) and 318 malignant breasts (27%), with 328 cancer lesions. The AI AUC was 0.942 at breast level and 0.929 at lesion level (difference -0.013, p < 0.01). The mean human AUC was 0.878 at breast level and 0.851 at lesion level (difference -0.027, p < 0.01). AI outperformed human readers at the breast and lesion level (ps < 0.01, respectively) according to the AUC. AI's diagnostic performance significantly decreased at the lesion level, indicating reduced accuracy in localising malignancies. However, its overall performance exceeded that of human readers. Question AI often recalls mammography cases not recalled by humans; to understand why, we as humans must consider the regions of interest it has marked as cancerous. Findings Evaluations of AI typically occur at the breast level, but performance decreases when AI is evaluated on a lesion level. This also occurs for humans. Clinical relevance To improve human-AI collaboration, AI should be assessed at the lesion level; poor accuracy here may lead to automation bias and unnecessary patient procedures.

Regular mammography screening is essential for early breast cancer detection. Deep learning-based risk prediction methods have sparked interest to adjust screening intervals for high-risk groups. While early methods focused only on current mammograms, recent approaches leverage the temporal aspect of screenings to track breast tissue changes over time, requiring spatial alignment across different time points. Two main strategies for this have emerged: explicit feature alignment through deformable registration and implicit learned alignment using techniques like transformers, with the former providing more control. However, the optimal approach for explicit alignment in mammography remains underexplored. In this study, we provide insights into where explicit alignment should occur (input space vs. representation space) and if alignment and risk prediction should be jointly optimized. We demonstrate that jointly learning explicit alignment in representation space while optimizing risk estimation performance, as done in the current state-of-the-art approach, results in a trade-off between alignment quality and predictive performance and show that image-level alignment is superior to representation-level alignment, leading to better deformation field quality and enhanced risk prediction accuracy. The code is available at https://github.com/sot176/Longitudinal_Mammogram_Alignment.git.

Pathological complete response (pCR) to neoadjuvant systemic therapy (NAST) is an established prognostic marker in breast cancer (BC). Multimodal deep learning (DL), integrating diverse data sources (radiology, pathology, omics, clinical), holds promise for improving pCR prediction accuracy. This systematic review synthesizes evidence on multimodal DL for pCR prediction and compares its performance against unimodal DL. Following PRISMA, we searched PubMed, Embase, and Web of Science (January 2015-April 2025) for studies applying DL to predict pCR in BC patients receiving NAST, using data from radiology, digital pathology (DP), multi-omics, and/or clinical records, and reporting AUC. Data on study design, DL architectures, and performance (AUC) were extracted. A narrative synthesis was conducted due to heterogeneity. Fifty-one studies, mostly retrospective (90.2%, median cohort 281), were included. Magnetic resonance imaging and DP were common primary modalities. Multimodal approaches were used in 52.9% of studies, often combining imaging with clinical data. Convolutional neural networks were the dominant architecture (88.2%). Longitudinal imaging improved prediction over baseline-only (median AUC 0.91 vs. 0.82). Overall, the median AUC across studies was 0.88, with 35.3% achieving AUC ≥ 0.90. Multimodal models showed a modest but consistent improvement over unimodal approaches (median AUC 0.88 vs. 0.83). Omics and clinical text were rarely primary DL inputs. DL models demonstrate promising accuracy for pCR prediction, especially when integrating multiple modalities and longitudinal imaging. However, significant methodological heterogeneity, reliance on retrospective data, and limited external validation hinder clinical translation. Future research should prioritize prospective validation, integration underutilized data (multi-omics, clinical), and explainable AI to advance DL predictors to the clinical setting.

Reflection Ultrasound Computed Tomography (UCT) is gaining prominence as an essential instrument for breast cancer screening. However, reflection UCT quality is often compromised by the variability in sound speed across breast tissue. Traditionally, reflection UCT utilizes the Delay and Sum (DAS) algorithm, where the Time of Flight significantly affects the coherence of the reflected radio frequency (RF) data, based on an oversimplified assumption of uniform sound speed. This study introduces three meticulously engineered modules that leverage the spatial correlation of receiving arrays to improve the coherence of RF data and enable more effective summation. These modules include the self-supervised blind RF data segment block (BSegB) and the state-space model-based strong reflection prediction block (SSM-SRP), followed by a polarity-based adaptive replacing refinement (PARR) strategy to suppress sidelobe noise caused by aperture narrowing. To assess the effectiveness of our method, we utilized standard image quality metrics, including Peak Signal-to-Noise Ratio (PSNR), Structural Similarity Index Measure (SSIM), and Root Mean Squared Error (RMSE). Additionally, coherence factor (CF) and variance (Var) were employed to verify the method's ability to enhance signal coherence at the RF data level. The findings reveal that our approach greatly improves performance, achieving an average PSNR of 19.64 dB, an average SSIM of 0.71, and an average RMSE of 0.10, notably under conditions of sparse transmission. The conducted experimental analyses affirm the superior performance of our framework compared to alternative enhancement strategies, including adaptive beamforming methods and deep learning-based beamforming approaches.

This study aimed to evaluate the ability of ChatGPT and Breast Ultrasound Helper, a special ChatGPT-based subprogram trained on ultrasound image analysis, to analyze and differentiate benign and malignant breast lesions on ultrasound images. Ultrasound images of histopathologically confirmed breast cancer and fibroadenoma patients were read GPT-4o (the latest ChatGPT version) and Breast Ultrasound Helper (BUH), a tool from the "Explore" section of ChatGPT. Both were prompted in English using ACR BI-RADS Breast Ultrasound Lexicon criteria: lesion shape, orientation, margin, internal echo pattern, echogenicity, posterior acoustic features, microcalcifications or hyperechoic foci, perilesional hyperechoic rim, edema or architectural distortion, lesion size, and BI-RADS category. Two experienced radiologists evaluated the images and the responses of the programs in consensus. The outputs, BI-RADS category agreement, and benign/malignant discrimination were statistically compared. A total of 232 ultrasound images were analyzed, of which 133 (57.3%) were malignant and 99 (42.7%) benign. In comparative analysis, BUH showed superior performance overall, with higher kappa values and statistically significant results across multiple features (P .001). However, the overall level of agreement with the radiologists' consensus for all features was similar for BUH (κ: 0.387-0.755) and GPT-4o (κ: 0.317-0.803). On the other hand, BI-RADS category agreement was slightly higher in GPT-4o than in BUH (69.4% versus 65.9%), but BUH was slightly more successful in distinguishing benign lesions from malignant lesions (65.9% versus 67.7%). Although both AI tools show moderate-good performance in ultrasound image analysis, their limited compatibility with radiologists' evaluations and BI-RADS categorization suggests that their clinical application in breast ultrasound interpretation is still early and unreliable.

Breast cancer is the most prevalent malignancy in women and a leading cause of mortality. Accurate assessment of axillary lymph node metastasis (LNM) is critical for breast cancer management. Exploring non-invasive methods such as radiomics for the detection of LNM is highly important. We systematically searched Pubmed, Embase, Scopus, Web of Science and google scholar until 11 March 2024. To assess the risk of bias and quality of studies, we utilized the quality assessment of diagnostic accuracy studies (QUADAS) tool as well as the radiomics quality score (RQS). Area under the curve (AUC), sensitivity, specificity and accuracy were determined for each study to evaluate the diagnostic accuracy of radiomics in magnetic resonance imaging (MRI) for detecting LNM in patients with breast cancer. This meta-analysis of 20 studies (5072 patients) demonstrated an overall AUC of 0.83 (95% confidence interval (CI): 0.80-0.86). Subgroup analysis revealed a trend towards higher specificity when radiomics was combined with clinical factors (0.83) compared to radiomics alone (0.79). Sensitivity analysis confirmed the robustness of the findings and publication bias was not evident. The radiomics models increased the likelihood of a positive LNM outcome from 37% to 73.2% when initial probability was positive and decreased the likelihood to 8% when initial probability was negative, highlighting their potential clinical utility. Radiomics as a non-invasive method demonstrates strong potential for detecting LNM in breast cancer, offering clinical promise. However, further standardization and validation are needed in future studies.