Breast Imaging Research Articles

Mammography Home Workstations and Remote Diagnostic Breast Imaging: Current Practice Patterns and Planned Future Directions.

Assess current practices and plans regarding home workstations and remote diagnostic breast imaging in the United States. A 43-question survey relating to remote breast imaging was distributed to Society of Breast Imaging members from July 6, 2023, through August 2, 2023. A descriptive summary of responses was performed. Pearson's chi-squared test was used to compare demographic variables of respondents and questions of interest. In total, 424 surveys were completed (response rate 13%, 424/3244). One-third of breast imaging radiologists (31%, 132/424) reported reading examinations from home or a personal remote site for a median of 25% of their clinical time. The most common types of examinations read from home were screening mammography (90%, 119/132), screening US (58%, 77/132), diagnostic mammography and MRI (both 53%, 70/132), and diagnostic US (49%, 65/132). Respondents from private practices were more likely than those from academic practices to read diagnostic imaging from home (67%, 35/52 vs 29%, 15/52; P <.001). Respondents practicing in the West were less likely to read breast imaging examinations from home compared with those in other geographic regions (18%, 12/67 vs 28%-43% for other regions; P = .023). No differences were found among respondents' overall use of home workstations based on age, gender, or having dependents. Most respondents (75%, 318/424) felt that remote breast reading would be a significant practice pattern in the future. Home workstations for mammography and remote diagnostic breast imaging are a considerable U.S. practice pattern. Further research should explore radiologist preferences regarding remote breast imaging and its impact on clinical care and radiologist well-being.

Molecular breast imaging : Positron emission tomography/magnetic resonance imaging and targeted tracers

Molecular imaging has been introduced into breast imaging in recent years, in order to improve breast cancer (BC) depiction as well as our understanding of cancer-associated processes at acellular and molecular level. This review offers an overview of the various molecular imaging modalities implemented in breast imaging as well as of the most significant novel radiotracers and their potential role for the functional evaluation of BC. The applications and the diagnostic potential of different imaging modalities (scintimammography [SM], breast-specific γimaging [BSGI], positron emission tomography [PET] mammography [PEM] and PET/MRI) as well as specific tracers (18-fluormisonidazole [18F‑MISO], 18-fluoro-L-thymidine [18FLT], 18-fluoroestradiol [18FES], 89-zirconium-trastuzumab, 18-Fluoroethylcholine [18FEC] and 68-gallium-fibroblast activation protein inhibitor [68Ga-FAPI]) will be discussed. BSGI increases the sensitivity of SM for small (< 1 cm) lesions, while PEM is more sensitive than whole-body PET scans. Hybrid PET/MRI is the most promising imaging modality for the assessment of BC. While 18F-FDG illustrates the glucose metabolism of cancer cells, novel tracers have other, tumor-specific targets: 18F‑MISO assesses tumor hypoxia, 18FLT the metabolism of DNA, 18FES and 89Zr-trastuzumab the tumor receptor status, 18FEC the metabolism of choline and 68Ga-FAPI cancer-associated fibroblasts. It can be expected that molecular imaging will gain importance for breast imaging in the future, enabling an improved diagnosis, staging, and treatment follow-up.

Transition to GPU-based reconstruction for clinical organ-targeted PET scanner.

This article explores a new Graphics Processing Unit (GPU)-based techniques for efficient image reconstruction in organ-targeted Positron Emission Tomography (PET) scanners with planar detectors .&#xD;Approach: GPU-based reconstruction is applied to the Radialis low-dose organ-targeted PET technology, developed to overcome the issues of high exposure and limited spatial resolution inherent in traditional whole-body PET/CT (Computed Tomography) scans. The Radialis planar detector technology is based on four-side tileable sensor modules that can be seamlessly combined into a sensing area of the needed size, optimizing the axial field-of-view (AFOV) for specific organs, and maximizing geometric sensitivity. The article explores the transition from Central Processing Unit (CPU)-based Maximum Likelihood Expectation Maximization (MLEM) algorithms to a GPU-based counterpart, demonstrating a tenfold overall speedup in image reconstruction with a hundredfold improvement in iteration speed.&#xD;Main Results: Through standardized PET performance tests and clinical image analysis, this work demonstrates that GPU-based image reconstruction maintains diagnostic image quality while significantly reducing reconstruction times. The application of this technology, particularly in breast imaging using the Radialis Low-Dose Positron Emission Mammography (LD-PEM), significantly reduces exam times thus improving patient comfort and throughput in clinical settings.&#xD;Significance: This study represents an important advancement in the clinical workflow of PET imaging, providing insights into optimizing reconstruction algorithms to effectively leverage the parallel processing capabilities of GPUs.&#xD.

False-Negative Review from the Mammography Audit: Refining Breast Imaging Practice.

Annual review of false-negative (FN) mammograms is a mandatory and critical component of the Mammography Quality Standards Act (MQSA) annual mammography audit. FN review can help hone reading skills and improve the ability to detect cancers at mammography. Subtle architectural distortion, asymmetries (seen only on one view), small lesions, lesions with probably benign appearance (circumscribed regular borders), isolated microcalcifications, and skin thickening are the most common mammographic findings when the malignancy is visible at retrospective review of FN mammograms. Most FN mammograms are not due to radiologist error. There are common and predictable settings in which FN mammograms occur. Patient factors associated with elevated FN mammograms include dense breasts, elevated lifetime risk of breast cancer, and personal history of breast cancer treated with lumpectomy and radiation therapy. About half of FN cancers are detected by supplemental screening examinations and half manifest clinically. The most common manifesting symptoms for interval cancers are a palpable abnormality, nipple discharge, and skin changes. Interval cancers can have more aggressive pathologic features and higher rates of node positivity. The FN review includes Breast Imaging Reporting and Data System (BI-RADS) 3 cases that develop a cancer diagnosis during surveillance. Nonbreast malignancies diagnosed as interval cancers (most commonly lymphoma and metastatic disease) do not need to be counted as FNs for audit purposes. The FN review and annual audit are confidential processes that protect patient and radiologist information while allowing meaningful quality control and improvement. Although FN mammograms are rare, review of these cases is a valuable educational tool. The slide presentation from the RSNA Annual Meeting is available for this article. ©RSNA, 2025.

Enhancement in dry setup of microwave breast imaging based on delay estimation and channel fidelity

Enhancement in dry setup of microwave breast imaging based on delay estimation and channel fidelity

A Novel Slot Spiral Symmetric Array Antenna with a Wide Axial Ratio Beamwidth for Microwave-Induced Thermoacoustic Tomography Applications

Conventional circularly polarized antennas have been employed to deliver microwave illumination in microwave-induced thermoacoustic tomography (TAT). However, these antennas exhibit several limitations in TAT systems, including low efficiency, poor axial ratio (AR) roundness, and narrow axial ratio beamwidth (ARBW). These issues lead to uniform radiation only within a relatively confined area, thereby restricting their effectiveness in clinical applications such as breast imaging. To address these issues, we propose a novel planar slot array antenna that offers a wide ARBW and improved axial ratio (AR) roundness, enabling homogeneous illumination over a larger field. We validated this approach both theoretically and experimentally. Tissue-mimicking phantoms were imaged, demonstrating that the antenna generated a circularly polarized electric field as well as a uniformly illuminated area. These advantages make the antenna proposed in this paper more suitable for clinical imaging compared to traditional microwave radiating antennas.

Improving Wellness Through Reading Room Design and Flexible Scheduling Options.

Breast radiologists have high rates of burnout. Some contributing factors include the sedentary nature of the occupation, reading room design and isolation associated with higher volumes, and increased remote interpretation. Reading rooms can also be filled with numerous distractions and produce conditions that do not support optimal workflow. Identifying and addressing these issues may help prolong physician careers and increase overall productivity. This article presents approaches to improve wellness for breast imaging radiologists and reduce the overall rate of burnout.

Multi-task interaction learning for accurate segmentation and classification of breast tumors in ultrasound images.

In breast diagnostic imaging, the morphological variability of breast tumors and the inherent ambiguity of ultrasound images pose significant challenges. Moreover, multi-task computer-aided diagnosis systems in breast imaging may overlook inherent relationships between pixel-wise segmentation and categorical classification tasks. &#xD;Approach. In this paper, we propose a multi-task learning network with deep inter-task interactions that exploits the inherently relations between two tasks. First, we fuse self-task attention and cross-task attention mechanisms to explore the two types of interaction information, location and semantic, between tasks. In addition, a feature aggregation block is developed based on the channel attention mechanism, which reduces the semantic differences between the decoder and the encoder. To exploit inter-task further, our network uses an circle training strategy to refine heterogeneous feature with the help of segmentation maps obtained from previous training. &#xD;Main results. The experimental results show that our method achieved excellent performance on the BUSI and BUS-B datasets, with DSCs of 81.95% and 86.41% for segmentation tasks, and F1 scores of 82.13% and 69.01% for classification tasks, respectively. &#xD;Significance. The proposed multi-task interaction learning not only enhances the performance of all tasks related to breast tumor segmentation and classification but also promotes research in multi-task learning, providing further insights for clinical applications.}.

Breast ultrasound imaging systems performance evaluation using novel Contrast-Detail (C-D) and Anechoic-Target (A-T) phantoms.

C-D and A-T phantoms were imaged using a range of clinical breast ultrasound systems and imaging modes. A novel sensitive imaging performance metric, the Detectability Score (DS), was proposed which encompasses the Lesion Contrast to Noise Ratio (LCNR) weighted by the lesion depth and diameter. A geometry-based theoretical model comparing LCNR measured using spherical and cylindrical phantom anechoic/lesion targets was developed to investigate the influence of slice thickness on focal lesion detectability. LCNR and DS metrics derived from phantom image measurments were capable of differentiating the imaging performance of a range of ultrasound systems and advanced imaging modes, with the -2 dB contrast lesion targets offered as the most challenging to resolve. The geometry-based theoretical model, validated against phantom measurements, demonstrated the significant influence of slice thickness on focal lesion detectability, highlighting the need for increased availability of low contrast resolution spherical target phantoms for clinically realistic performance evaluation. The performance metrics coupled with the -1 dB contrast targets provide scope for evaluating future technological improvements in ultrasound systems. Given the high dependence of breast cancer care on high quality ultrasound imaging techniques, there is a need for evaluating imaging performance using clinically relevant test objects.

Breast MRI to Screen Women With Extremely Dense Breasts.

Women with extremely dense breasts are at a higher risk of breast cancer, and the sensitivity of mammography in this group is reduced due to the masking effect of overlapping tissue. This review examines supplemental screening methods to improve detection in this population, with a focus on MRI. Morphologic techniques offer limited benefits, digital breast tomosynthesis (DBT) shows inconsistent results, and ultrasound (US), while improving cancer detection rates (CDR), results in a higher rate of false positives. Functional imaging techniques show better performance, molecular breast imaging increases CDR but is limited in availability, and contrast-enhanced mammography is promising, with good results and as an accessible technique, but requires further validation. MRI, with sensitivity ranging from 81% to 100%, is the most supported modality. Despite strong evidence for MRI in this population, high costs, use of contrast, and longer scan times hinder widespread use. Abbreviated MRI protocols aim to overcome these barriers by reducing costs and scan duration. As personalized screening becomes a future focus, MRI remains the most effective option for women with extremely dense breasts. LEVEL OF EVIDENCE: 1 TECHNICAL EFFICACY: Stage 5.

Evaluating Suspicious Breast Lesions: A Head-to-Head Comparison of Contrast-Enhanced Mammography and Breast MRI

Background: Magnetic resonance imaging (MRI) is the most sensitive technique for the detection of breast cancer in contemporary medical practice. Contrast-enhanced mammography (CEM), a novel technology, provides a contrast-enhanced imaging alternative to breast MRI. Objectives: to assess the diagnostic efficacy of CEM in the characterization of suspicious breast lesions, with histopathological results serving as the gold standard. Compare CEM with dynamic contrast enhanced MRI (DCE-MRI). Patients and Methods: 30 individuals with suspicious lesions identified using conventional mammography or ultrasonography underwent CEM and DCE-MRI examinations after receiving ethical approval. Proficient radiologists evaluated all discernible lesions using the Breast Imaging Reporting and Data System (BI-RADS) classifications (categories 1–6). Histopathological data were compared to the morphological descriptions of each lesion that were received from each modality. Results: 30 lesions were identified by the combination of breast MRI and CEM in the 30 patients who were enrolled in CEM/MRI investigations. Histopathology verified that 15 of the 30 lesions were malignant and 15 were benign. Twelve of the fifteen malignant lesions were invasive cancers, while three were in situ cancers. Both breast MRI and CEM demonstrated a sensitivity of 93.33%. The specificity of breast MRI was 93.33%, while CEM had a specificity of 80.00%. The breast MRI achieved an accuracy of 93.33%, while the CEM achieved an accuracy of 86.67%. Conclusions: Contrast-enhanced mammography (CEM) has the potential to replace dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) as a problem-solving instrument for the characterization of indeterminate breast lesions, particularly in situations where MRI is contraindicated or not promptly accessible. CEM provides a viable alternative for evaluating ambiguous lesions, as it utilizes similar principles of contrast enhancement and tumor vascularity. However, it has the additional benefit of being more broadly available, less expensive, and better tolerated by specific patient populations.

Maximizing diagnostic precision: Evaluating the combined Yokohama and BI-RADS scoring system for breast lesions.

The diagnosis of breast cancer necessitates a multifaceted approach integrating cytopathological and radiological assessments. The International Academy of Cytology (IAC) Yokohama system and Breast Imaging Reporting and Data System (BI-RADS) are fundamental frameworks in this context. This study aims to evaluate the diagnostic potential of a combined Yokohama-BI-RADS scoring system for breast lesions. A retrospective analysis was conducted on fine needle aspirates from January to June 2023. The cases were classified using the IAC Yokohama system and sono-mammography BI-RADS score. Sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and diagnostic accuracy were calculated using histopathological diagnoses as the reference standard. Optimal cut-off scores for the combined scoring system were determined. Among 52 patients, cytological diagnoses encompassed non-diagnostic, benign, atypical, suspicious of malignancy, and malignant categories. BI-RADS scores ranged from 1 to 5. The combined Yokohama-BI-RADS score exhibited superior diagnostic accuracy (AUC: 0.986) compared to individual systems. The combined Yokohama-BI-RADS scoring system represents a promising advancement in breast lesion evaluation, providing enhanced diagnostic precision by integrating cytopathological and radiological data. This approach has the potential to optimize clinical decision-making and contribute to improved patient outcomes in breast cancer management.

Comparative Analysis of Automated and Handheld Breast Ultrasound Findings for Small (≤1 cm) Breast Cancers Based on BI-RADS Category.

Objectives: This study aimed to compare ultrasound (US) findings between automated and handheld breast ultrasound (ABUS and HHUS, respectively) in small breast cancers, based on the breast imaging reporting and data system (BI-RADS) category. Methods: We included 51 women (mean age: 52 years; range: 39-66 years) with breast cancer (invasive or DCIS), all of whom underwent both ABUS and HHUS. Patients with tumors measuring ≤1 cm on either modality were enrolled. Two breast radiologists retrospectively evaluated multiple imaging features, including shape, orientation, margin, echo pattern, and posterior characteristics and assigned BI-RADS categories. Lesion sizes were compared between US and pathological findings. Statistical analyses were performed using Bowker's test of symmetry, a paired t-test, and a cumulative link mixed model. Results: ABUS assigned lower BI-RADS categories than HHUS while still maintaining malignancy suspicion in categories 4A or higher (54.8% consistent with HHUS; 37.3% downcategorized in ABUS, p = 0.005). While ABUS demonstrated less aggressive margins in some cases (61.3% consistent with HHUS; 25.8% showing fewer suspicious margins in ABUS), this difference was not statistically significant (p = 0.221). Similarly, ABUS exhibited slightly greater height-width ratios compared to HHUS (median, interquartile range: 0.98, 0.7-1.12 vs. 0.86, 0.74-1.10, p = 0.166). No significant differences were observed in other US findings or tumor sizes between the two modalities (all p > 0.05). Conclusions: Small breast cancers exhibited suspicious US features on both ABUS and HHUS, yet they were assigned lower BI-RADS assessment categories on ABUS compared to HHUS. Therefore, when conducting breast cancer screening with ABUS, it is important to remain attentive to even subtle suspicious findings, and active consideration for biopsy may be warranted.

THE ROLE OF BREAST IMAGING MODALITIES IN THE DIAGNOSIS AND MANAGEMENT OF BENIGN BREAST DISEASES

Objective: Accurate diagnosis and management of benign breast diseases are crucial aspects of breast healthcare. Breast imaging modalities such as mammography, ultrasound, and magnetic resonance imaging (MRI) play a pivotal role in differentiating benign from malignant lesions and guiding appropriate management strategies. Methods: This retrospective study was conducted at [Institution Name] from January 2018 to December 2022. It involved a review of medical records, imaging results, and management outcomes of patients diagnosed with benign breast diseases. Imaging modalities included mammography, ultrasound, and MRI. The diagnostic accuracy of each modality was assessed by comparing imaging findings with histopathological results. Statistical analyses were performed using SPSS version 26.0. Results: A total of 200 patients were included, with a mean age of 45.3±12.6 y. Mammography showed a sensitivity of 85% and specificity of 78%, while ultrasound had a sensitivity of 92% and specificity of 84%. MRI exhibited the highest accuracy with a sensitivity of 95% and specificity of 90%. Combined imaging modalities achieved the highest diagnostic performance. Management strategies varied, with watchful waiting, medication, minimally invasive procedures, and surgical interventions employed based on the specific condition. Conclusion: Breast imaging modalities are indispensable in diagnosing and managing benign breast diseases. The integration of mammography, ultrasound, and MRI provides comprehensive diagnostic information, guiding optimal management strategies and improving patient outcomes. Future advancements in imaging technology promise to further enhance diagnostic capabilities.

Hereditary Breast Cancer: Comprehensive Risk Assessment and Prevention Strategies.

Women carrying pathogenic/likely pathogenic (P/LP) variants in moderate- or high-penetrance genes have an increased risk of developing breast cancer. However, most P/LP variants associated with breast cancer risk show incomplete penetrance. Age, gender, family history, polygenic risk, lifestyle, reproductive, hormonal, and environmental factors can affect the expressivity and penetrance of the disease. However, there are gaps in translating how individual genomic variation affects phenotypic presentation. The expansion of criteria for genetic testing and the increasing utilization of comprehensive genetic panels may enhance the identification of individuals carrying P/LP variants linked to hereditary breast cancer. Individualized risk assessment could facilitate the implementation of personalized risk-reduction strategies for these individuals. Preventive interventions encompass lifestyle modifications, chemoprevention, enhanced surveillance through breast imaging, and risk-reducing surgeries. This review addresses the current literature's inconsistencies and limitations, particularly regarding risk factors and the intensity of preventive strategies for women with P/LP variants in moderate- and high-penetrance genes. In addition, it synthesizes the latest evidence on risk assessment and primary and secondary prevention in women at high risk of breast cancer.

Ethnic enclaves, neighborhood socioeconomic status, and obesity among Hispanic women in Chicago: a latent profile analysis approach.

The prevalence of obesity, a crucial risk factor for breast cancer, is markedly higher among Hispanic women. The interaction between ethnic enclaves and neighborhood socioeconomic status (SES) as a determinant of this disparity warrants further research. We aimed to identify neighborhood profiles based on ethnic enclaves and socioeconomic status to evaluate the association with obesity among Hispanic women in the metropolitan Chicago region. We used a convenience sample of 24,884 Hispanic women over age 40 who obtained breast imaging from the largest healthcare organization in Chicago between 2010 and 2017. We conducted LPA to characterize neighborhood composition based on tract indicators of ethnic enclaves, disadvantage, and affluence. Multivariate linear and multinomial logistic regression models were used to evaluate the association of neighborhood profiles with BMI. The LPA model identified four latent profiles, labeled based on their most significant characteristic as "middling," "disadvantage" "ethnic enclaves," and "affluent". Close to 50% of women in the disadvantage profile were obese and obese class II. Women in the disadvantage profile had the highest relative risk of being obese II (OR: 2.74 CI 95% 2.23, 3.36), compared to women in the middling profile. Women in the ethnic enclave and affluent profile were positively and negatively associated with obesity, respectively. Using LPA to group individuals according to their combined traits provides empirical evidence to strengthen our understanding of how neighborhoods influence obesity in Hispanic women. The study findings suggest that ethnic enclaves, that are also disadvantage, are associated with obesity in Hispanic women.

Enhance registration precision of transmission breast images utilizing improved Levenberg-Marquardt optimization algorithm with normalized cross-correlation.

Transmission imaging may become a possible advance for breast cancer screening with non-invasive, cost-effective, and radiation-free approaches for early detection. Frame accumulation can successfully eliminate the issue of low SNR, low grayscale and poor quality in transmission image. However, frame accumulation accuracy can be diminished because of inherent human body instability during image acquisition and the light absorption characteristics of breast tissue, resulting in distorted and misplaced image sequences. Therefore, improved Levenberg-Marquardt optimization algorithm with normalized cross-correlation is used as an innovative approach to rectify image sequences before frame accumulation processing. Two separate sets of data, showing breast images with and without markers, were collected using a halogen bulb and a mobile phone camera to validate the suggested method. The approach includes coarse registration utilizing normalized cross-correlation for initial value estimation, followed by fine registration using Levenberg-Marquardt algorithm. The results demonstrate a notable improvement in both accuracy of registration and frame accumulation quality. Specifically, the registration speed showed a remarkable increase, being 8.7 times faster, especially prominent in images that included markers. These images displayed normalized cross-correlation values reaching up to 0.99. The research emphasizes the future potential of the suggested method in overcoming the image quality challenges associated with breast transmission imaging, providing a significant milestone toward more accurate and efficient early breast cancer screening methods. Moreover, transmission imaging systems for the breast have been developed to verify the safety and effectiveness of the implemented technology.

A Radiomic-Clinical Model of Contrast-Enhanced Mammography for Breast Cancer Biopsy Outcome Prediction.

In the USA over 1 million breast biopsies are performed annually. Approximately 9.6% diagnostic exams were given Breast Imaging Reporting and Data System (BI-RADS) ≥4A, most of which are 4A/4B. Contrast-enhanced mammography (CEM) may improve biopsy outcome prediction for this subpopulation, but machine learning-based analysis of CEM is largely unexplored. We aim to develop a machine learning-based analysis of CEM using computer-extracted radiomics and radiologist-assessed descriptors to predict breast biopsy outcomes of BI-RADS 4A/4B/4C or 5 lesions. This HIPPA-compliant, IRB-approved study included women in a single institution who had BI-RADS 4A/4B/4C or 5 lesions and underwent CEM imaging prior to biopsy. Logistic regression models were built to predict biopsy outcomes using radiomics features and four radiologist-assessed qualitative descriptors. A cohort of 201 patients was used for model development/training, and an independent group of 86 patients were used as an internal test set. AUC was used to measure model's performance. Positive predictive value (PPV) was assessed on subgroups of BI-RADS 4A or 4B lesions. Model AUC was 0.90 for radiomics, 0.81 for clinical descriptors and 0.88 for their combination. On patients with an initial BI-RADS 4A or 4B scores, model combining radiomics and clinical descriptors of pre-biopsy CEM increased PPV3 to 18% from the radiologist's 6% for 4A patients, and to 25% from the radiologist's 17% for 4B patients. Machine learning models combining radiomics features and clinical descriptors on CEM can predict breast biopsy outcomes on women with BI-RADS 4A/4B/4C or 5 lesions.

Preoperative comprehensive risk estimation for axillary lymph node metastasis in breast cancer: development and verification of a network-based prediction model

To prevent the overaggressive treatment of axillary lymph nodes (ALNs) in breast cancer, it is necessary to develop a convenient analysis method that accurately and comprehensively reflects whether ALNs are metastatic or nonmetastatic. We retrospectively analyzed data from patients who underwent surgery for breast cancer at the Weifang Hospital of Traditional Chinese Medicine between January 2019 and June 2023. Binary logistic regression analysis was used to predict the metastasis status of ALNs. The developmental data set included 531 patients (January 2019–June 2023). The validation set included separate data points (n = 178, January 2019–June 2023). Multivariate analysis revealed that positive findings on breast physical examination, ultrasound grades of ALNs, lymphovascular invasion, and Her-2 status had significant predictive value for metastatic ALNs. Based on these findings, a 5-grade risk scoring system and 3-level management recommendations were developed. The risk of metastasis ranged from 11.25 to 93.46%, which was positively correlated with an increase in risk grade. The areas under the curve of the development and validation sets were 0.895 and 0.865, respectively. Ultimately, a convenient, accurate and comprehensive web-based predictive model was constructed using various breast cancer clinical, imaging and pathological criteria to stratify ALNs according to the metastasis probability.

Accelerated High-resolution T1- and T2-weighted Breast MRI with Deep Learning Super-resolution Reconstruction.

To assess the performance of an industry-developed deep learning (DL) algorithm to reconstruct low-resolution Cartesian T1-weighted dynamic contrast-enhanced (T1w) and T2-weighted turbo-spin-echo (T2w) sequences and compare them to standard sequences. Female patients with indications for breast MRI were included in this prospective study. The study protocol at 1.5 Tesla MRI included T1w and T2w. Both sequences were acquired in standard resolution (T1S and T2S) and in low-resolution with following DL reconstructions (T1DL and T2DL). For DL reconstruction, two convolutional networks were used: (1) Adaptive-CS-Net for denoising with compressed sensing, and (2) Precise-Image-Net for resolution upscaling of previously downscaled images. Overall image quality was assessed using 5-point-Likert scale (from 1=non-diagnostic to 5=excellent). Apparent signal-to-noise (aSNR) and contrast-to-noise (aCNR) ratios were calculated. Breast Imaging Reporting and Data System (BI-RADS) agreement between different sequence types was assessed. A total of 47 patients were included (mean age, 58±11 years). Acquisition time for T1DL and T2DL were reduced by 51% (44 vs. 90s per dynamic phase) and 46% (102 vs. 192s), respectively. T1DL and T2DL showed higher overall image quality (e.g., 4 [IQR, 4-4] for T1S vs. 5 [IQR, 5-5] for T1DL, P<0.001). Both, T1DL and T2DL revealed higher aSNR and aCNR than T1S and T2S (e.g., aSNR: 32.35±10.23 for T2S vs. 27.88±6.86 for T2DL, P=0.014). Cohen k agreement by BI-RADS assessment was excellent (0.962, P<0.001). DL for denoising and resolution upscaling reduces acquisition time and improves image quality for T1w and T2w breast MRI.