Search

A recent study conducted by researchers at Najran University focused on the use of a model called Attention U-Net for identifying breast cancer. The model was applied to the Breast Ultrasound Image Dataset (BUSI), which consists of 780 breast images categorized into benign, malignant, and normal cases. The model demonstrated high accuracy in accurately delineating and segregating tumor borders, achieving an overall accuracy of 0.98, precision of 0.97, recall of 0.90, and a dice score of 0.92. The research emphasizes the importance of early detection in improving diagnostic capabilities and enabling prompt medical interventions. [Extracted from the article]

A recent study conducted at Zhejiang Cancer Hospital in Hangzhou, China, explored the use of deep learning models to distinguish between breast phyllodes tumors (PTs) and fibroadenomas (FAs) based on ultrasound images. The study collected 1180 ultrasound images from 539 patients and trained five deep learning network models using transfer learning and cross-validation methods. The best-performing model, Xception, demonstrated superior diagnostic performance compared to radiologists, improving accuracy for senior, intermediate, and junior radiologists. The researchers concluded that the deep learning model has the potential to enhance diagnostic efficiency and decision-making for identifying and treating these challenging breast tumors. [Extracted from the article]

A study conducted at the University of California Los Angeles (UCLA) examined the impact of routing patients with clinically insignificant breast pain to screening mammograms. The researchers retrospectively reviewed 100 women without a history of breast cancer who underwent diagnostic mammograms and ultrasounds for breast pain. They found that 68 of the patients were eligible for screening mammograms, and with a screen-first approach, there would have been fewer diagnostic tests and a lower total cost. The study concluded that routing eligible patients to screening mammograms improves resource allocation and reduces healthcare spending without missing any cancers. [Extracted from the article]

A recent study conducted at Taiwan Adventist Hospital has found that GATA3-positive circulating tumor cells (CTCs) may be used to detect early-stage breast cancer. The study involved an 86-year-old female patient who had received a mammographic examination with no signs of malignancy. However, CTC testing on the same day revealed four GATA3-positive CTCs in her peripheral blood. Further examination confirmed the presence of invasive carcinoma, leading to surgical intervention. The researchers suggest that GATA3-positive CTCs could serve as a potential biomarker for early-stage breast cancer and should be further explored. [Extracted from the article]

A recent study conducted at Keimyung University in South Korea has developed a computer-aided diagnosis (CAD) system for breast ultrasound. The system utilizes a frequency domain technique to extract mass irregularity features, resulting in improved accuracy in tumor classification. The study used 5252 ultrasound breast tumor images, including both benign and malignant tumors, and employed a Support Vector Machine for classification. The CAD system demonstrated superior performance compared to traditional spatial gray level dependence (SGLD) and other features, providing a precise and quick distinction between benign and malignant breast tumors in clinical settings. [Extracted from the article]

A recent study conducted at the University of Abuja examined the value of ultrasonography in assessing screen-detected calcified ductal carcinoma in situ (DCIS). DCIS is often asymptomatic and detected through screening mammograms. The study found that breast ultrasonography can be a reliable adjunct in detecting microcalcifications and guiding biopsies, particularly for malignant type microcalcifications larger than 15 mm. The research suggests that ultrasonography may be a viable alternative for early breast cancer detection, especially in low- and medium-income countries. [Extracted from the article]

A recent study conducted at the University of Abuja compared the physical and sonographic features of the breast in women with fibroadenoma (FA) and women with breast lumps due to other diseases. The researchers found significant differences in the appearance of FA compared to other breast lesions, including differences in age, parity, lesion size, architecture, echogenicity, borders, capsule, and breast density. The study suggests that these differences could potentially improve the pre-operative diagnosis of FA and contribute to the development of diagnostic criteria for FA in the local context. The full research article can be accessed for free through the Nigerian Postgraduate Medical Journal. [Extracted from the article]

A recent study conducted by researchers at Shanghai University in China explores the use of multi-task learning (MTL) methods in breast imaging for lesion area perception and classification to aid in breast cancer diagnosis and personalized treatment. The study proposes a novel Multi-Feature Fusion Multi-Task (MFFMT) model that addresses information sharing conflicts and extracts richer lesion features. The researchers conducted experiments on two public breast ultrasound imaging datasets to validate the effectiveness of their proposed method. The study provides insights into the working mechanism of the framework and its impact on various factors. [Extracted from the article]

A recent report from the Sanjay Gandhi Postgraduate Institute of Medical Sciences in Lucknow, India, highlights the clinicopathologic profile and reasons for referral of women with benign breast disease to a tertiary care hospital. The study enrolled 245 women aged 18 years and older and found that the majority of patients were premenopausal and from rural areas. The most common presenting symptom was mastalgia, and many patients had visited public hospitals or consulted multiple doctors before seeking care at the tertiary center. The main reasons for visiting the center were fear of malignancy and dissatisfaction with counseling regarding breast cancer risk. The study suggests that counseling for benign breast disease patients should focus on alleviating fear of malignancy, and more insight among treating physicians is needed to address patient concerns. [Extracted from the article]

A study conducted in Evanston, Illinois, has found that the reproducibility of breast lesions found by handheld ultrasound (HHUS) can significantly impact clinical care. The study analyzed the features associated with breast mass position differences during HHUS and evaluated the ability of operators to reproduce the position of small masses with and without a computer-assisted scanning device. The results showed that computer-assisted scanning improved breast mass location reproducibility and exam efficiency. This research provides valuable insights into improving diagnostics in breast ultrasound. [Extracted from the article]

A recent study conducted at Zhejiang Cancer Hospital in China has proposed an automated method for segmenting the pulmonary artery, ascending aorta, and superior vena cava in fetal cardiac ultrasound images. The study introduces a deep learning network called CoFi-Net, which incorporates a coarse-fine collaborative strategy to improve vessel segmentation accuracy. The researchers found that CoFi-Net outperformed other segmentation models and demonstrated potential for enhancing congenital heart disease (CHD) diagnostic efficiency. The study also noted that CoFi-Net showed promise in breast lesion segmentation, despite not being specifically designed for that task. [Extracted from the article]

A recent study conducted at Sun Yat-sen University in Guangzhou, China, has developed a novel ultrasound contrast agent (UCA) called NPs-DPPA(C3F8) that can be used for both breast cancer diagnosis and treatment. The UCA is prepared using a combination of nanoprecipitation and ultrasound vibration methods and exhibits antiangiogenic and antitumor activities. It enables simultaneous real-time imaging, tumor assessment, and antitumor activity. The successful preparation of NPs-DPPA(C3F8) offers a new approach to integrating clinical diagnosis and treatment of triple-negative breast cancer (TNBC). [Extracted from the article]

A clinical trial, NCT06489158, has been launched to explore volatile organic compound (VOC) indicators that can effectively distinguish between benign and malignant breast diseases. The trial aims to compare the accuracy of VOC indicators and ultrasound diagnosis in determining benign and malignant diseases based on pathological gold standards. The trial is observational in nature and will collect exhaled samples from admitted women with breast ultrasound BI-RADS 4 grade. The trial is not yet recruiting and is expected to be completed by July 2025. [Extracted from the article]

A recent report discusses research conducted at RWTH Aachen University on breast cancer and the potential use of ultrasound localization microscopy (ULM) for monitoring the response to neoadjuvant chemotherapy (NAC). The study found that ULM, which provides super-resolution images of vasculature and functional features, may offer improvements in preselection and monitoring of NAC patients. Preliminary results showed that responders to NAC exhibited noticeable reductions in tumor volume and vessel coverage, while non-responders showed minimal tumor size reduction and no decrease in coverage. Although these findings are preliminary, they suggest that ULM has the potential to enhance the monitoring and preselection of NAC patients. [Extracted from the article]

A study conducted by researchers at Westlake University in Hangzhou, China, explores the use of deep convolutional neural networks in cancer detection, specifically in the context of automated three-dimensional breast ultrasound. The researchers propose a breast cancer detection framework based on deep learning and a transfer learning method that incorporates cross-organ cancer detection and contrastive learning based on breast imaging reporting and data systems (BI-RADS). The study demonstrates that cross-organ cancer detection parameters and experiences can be mutually referenced, and the contrastive learning method based on BI-RADS improves the detection performance of the model. The research provides valuable insights into the potential of deep learning in medical image analysis for breast cancer detection. [Extracted from the article]

A recent report from Nankai University in Tianjin, China discusses the challenges of accurately segmenting breast tumors from ultrasound images. The researchers propose an enhanced selective kernel convolution method that integrates multiple feature map region representations and adaptively recalibrates the weights of these regions. They conducted extensive experiments using three public breast ultrasound datasets and found that their method achieved superior performance in segmenting breast ultrasound images. The research has been peer-reviewed and the source code is available for further exploration. [Extracted from the article]

A recent report from the School of Humanities in Changsha, People's Republic of China, discusses the challenges of accurately diagnosing and classifying breast tumors using ultrasound imaging. The researchers propose a novel approach called DBL-Net, which is a multitask learning network that combines spatial and frequency domain information for tumor segmentation and classification. They also introduce a feature encoding module and a representational perception enhancer to improve the accuracy of the model. The researchers conducted tests on a breast ultrasound dataset and found that their method outperformed current state-of-the-art methods. This research has been peer-reviewed and published in Biomedical Signal Processing and Control. [Extracted from the article]

A study conducted by researchers at Iran University of Science and Technology in Tehran focuses on the development of a fully-automatic segmentation system for breast cancer detection using 3-D Automated Breast Ultrasound (ABUS). The system utilizes a bidirectional convolutional long short-term memory (Bi-ConvLSTM) network for tumor detection and a 2-D attentive UNet for precise segmentation of breast masses. The proposed modifications to the system achieved a Dice similarity index of 85.82% on a dataset of 60 volumes. This research has been peer-reviewed and published in Expert Systems with Applications. [Extracted from the article]

Researchers at the Second Affiliated Hospital of Xi'an Jiaotong University in China have developed a new approach for categorizing breast cancer using ultrasound imaging. The researchers incorporated a wavelet-based vision transformer network into their model, which allowed for the capture of significant features in the frequency domain. The proposed transformer network demonstrated exceptional performance in breast cancer classification, surpassing conventional deep learning techniques. This study represents a significant advancement in the diagnosis and categorization of breast cancer, highlighting the potential of transformer networks to enhance medical imaging analysis. [Extracted from the article]

A study conducted at the University of Hyderabad in India has developed a deep learning and genetic algorithm-based ensemble model for the classification of breast ultrasound images. The researchers compared the performance of different deep neural networks and found that MobileNet was optimal for feature extraction. They used a genetic algorithm for feature selection and created an ensemble model that demonstrated improved diagnostic accuracy for breast abnormality identification. This research has the potential to improve treatment outcomes for breast cancer patients. [Extracted from the article]

A study conducted at Hangzhou First People's Hospital explored the use of Automated Breast Ultrasound (AB US) in detecting and diagnosing mammographically Breast Imaging Reporting and Data System (BI-RADS) category 4 microcalcifications. The study included 575 patients with confirmed microcalcifications, of which 250 were malignant and 325 were benign. The results showed that AB US had a higher detection rate for malignant lesions compared to mammography, and it had higher sensitivity to malignant microcalcifications. The researchers concluded that AB US could be an effective supplementary examination method for breast microcalcifications, especially in dense breasts. [Extracted from the article]

Researchers from Xi'an University of Finance and Economics have developed a domain knowledge-based interpretable deep learning system for improving breast cancer risk prediction using paired multimodal ultrasound images. The system was trained on 4320 images of biopsy-confirmed lesions from 1348 patients and was found to predict breast cancer risk as accurately as experienced radiologists. The system's explainable features can aid junior radiologists in achieving better clinical outcomes and increase confidence levels for senior radiologists. This deep learning system has the potential to be incorporated into future breast cancer screening and support assisted or second-read workflows. [Extracted from the article]

A recent report from the University of Calgary discusses the importance of early diagnosis in breast cancer and presents an automated analysis and classification system for breast cancer using clinical markers. The researchers used a graph convolutional network (GCN) model to classify breast tumors as benign or malignant based on clinical features. The proposed model achieved a test accuracy of 98.73% and outperformed other models in comparison. The study concludes that a GCN model using graph data shows promise as an automated feature-based breast image classification system. [Extracted from the article]

A recent study conducted at Kaohsiung Medical University Hospital in Taiwan explored the efficacy of ultrasound-assisted vacuum-assisted biopsy (VABB) for impalpable breast lesions in Taiwanese women. The study found that VABB offers a non-inferior diagnostic option for patients who are not candidates for stereotactic biopsy. The use of sonography during the biopsy process allows for multiple samples to be obtained from different directions, minimizing the risk to underlying muscles. However, the study suggests that VABB may be more suitable for lower-risk findings, as complete excision rates can vary depending on the operator. [Extracted from the article]

A new report from Menofia University discusses the use of automated breast ultrasound (ABUS) as a supplemental tool for breast cancer screening in women with dense breasts. The study compared the diagnostic yield of full-field digital mammography (FFDM) alone versus FFDM with ABUS. The results showed that the use of ABUS reduced the number of recalls and improved breast cancer detection, with increased positive predictive value. The researchers concluded that ABUS can be a useful tool in improving the sensitivity of cancer detection in women with dense breasts. [Extracted from the article]

A recent study conducted by researchers at the University of Sao Paulo explored the use of machine learning models to predict the malignancy of breast masses identified by ultrasound. The researchers collected clinical and ultrasonographic attributes, as well as images, from 927 lesions and trained and validated five machine learning algorithms. The logistic regression model had the lowest mean errors and was selected to minimize false negative cases. The study concluded that machine learning models have the potential to help physicians reduce unnecessary ultrasound-guided breast biopsies by accurately predicting malignancy. [Extracted from the article]

A recent report discusses the use of artificial intelligence (AI) in breast cancer ultrasound screening. The researchers from the Department of Ultrasound in Chongqing, China, highlight that while ultrasound is the primary method for screening breast cancer, it can still be influenced by subjective factors. They explain that AI technology, specifically deep learning models, can improve the consistency and accuracy of diagnosis in breast cancer screening, diagnosis, and treatment. The researchers emphasize the potential of AI in enhancing the field of oncology and women's health. [Extracted from the article]

A recent study conducted at the University of Rochester has focused on breast cancer and the development of a new diagnostic tool called SoftVue breast tomographic ultrasound (BTUS). This technology has received approval from the Food and Drug Administration as an adjunct to breast cancer screening in women with dense breasts. SoftVue provides detailed images of breast tissue and can also be used as a cost-effective platform for artificial intelligence-assisted breast cancer identification. Additionally, SoftVue's quantitative acoustic maps allow for noninvasive temperature monitoring and targeted treatment options. The study suggests that women prefer the comfort of SoftVue over traditional mammograms and are seeking less invasive options for breast care. [Extracted from the article]

A study conducted by researchers from Sungkyunkwan University in Seoul, South Korea, evaluated the diagnostic performance of radiologists using artificial intelligence (AI)-based computer-aided detection software (CAD) for automated breast ultrasounds (ABUS). The study included 262 breast lesions with histopathological verification and found that the implementation of AI-CAD significantly improved the diagnostic performance of radiologists. The researchers concluded that AI-CAD could serve as a valuable diagnostic tool. The study was published in Academic Radiology and has been peer-reviewed. [Extracted from the article]

A recent meta-analysis conducted by Guangxi Medical University in China compared the effectiveness of breast-specific gamma imaging (BSGI), mammography, and ultrasound for diagnosing breast cancer. The study found that BSGI had the highest sensitivity and specificity in distinguishing between malignant and benign breast lesions, making it a valuable tool for diagnosing breast cancer in women with dense breasts. The research concluded that BSGI could play a crucial role in improving the diagnosis of breast cancer. This study has been peer-reviewed and provides important insights into breast cancer screening methods. [Extracted from the article]

The document provides information about a clinical trial, NCT06303232, being conducted by Riphah International University in Pakistan. The trial aims to investigate the combined effects of manual lymphatic drainage and therapeutic ultrasound on breast engorgement, pain, and self-efficacy in postpartum women. The primary outcome measures include pain scores, breast engorgement, and breastfeeding self-efficacy. The trial is currently ongoing and is recruiting participants, with an estimated completion date of September 1, 2024. The document also includes a list of publications related to breast engorgement and breastfeeding. [Extracted from the article]

A new report from the Department of Anatomy and Embriology in Timisoara, Romania, discusses the shifting demographic patterns of breast cancer, with an increasing incidence in younger age groups. The report evaluates the feasibility of using breast ultrasound as the primary screening method for breast cancer in low- and middle-income countries. Breast ultrasound offers advantages such as radiation-free imaging, suitability for repeated screenings, and preference for younger populations. However, limitations include reduced specificity and challenges in detecting microcalcifications. The report emphasizes the need for a comprehensive approach to breast cancer screening, international collaboration, and addressing limitations in resource-constrained settings. [Extracted from the article]

A study conducted at the University of Illinois in Champaign has proposed a new model for joint localization and classification of breast masses on ultrasound images. The model utilizes attention mechanisms and a sequential semi-supervised learning strategy to improve performance and reduce the need for data annotation. The study found that the proposed method was effective in two different breast ultrasound image datasets. The research has been peer-reviewed and the code for the model is available online. [Extracted from the article]

A recent study conducted by researchers at Universidad Tecnica Particular de Loja in Ecuador explores the use of generative adversarial networks (GANs) to reduce speckle noise in breast ultrasound images. The researchers trained two GAN models, Conditional GAN and Wasserstein GAN, using different architectures and datasets. The study found that the Conditional GAN model achieved better performance in terms of image quality, as measured by Peak Signal to Noise Ratio (PSNR) and Structural Similarity Index (SSIM), compared to the Wasserstein GAN model. The findings suggest that the use of GANs can improve the accuracy of breast ultrasound image interpretation and may have implications for computer-aided detection/diagnosis (CAD) systems. [Extracted from the article]

A systematic review conducted by researchers at Tehran University of Medical Sciences in Iran examined the diagnostic performance of magnetic resonance imaging (MRI) in the evaluation of non-mass lesions of the breast. The review included two studies and found that if any enhancement is present on MRI, most ultrasound non-mass lesions (NMLs) will exhibit non-mass enhancement. The distribution of non-mass enhancement lesions was primarily segmental and regional, with the highest number of malignancies associated with segmental distributions. The researchers concluded that non-mass lesions of the breast that do not enhance in MRI have a good prognosis, while cases with regional and segmental enhancement may warrant further pathological examination. [Extracted from the article]

A recent study conducted at Seoul National University Bundang Hospital explored the potential of using Quantitative Ultrasound (QUS) enhanced by artificial intelligence (AI) to improve the accuracy of breast cancer diagnosis. The researchers collected B-mode images and radiofrequency signals from breast lesions and analyzed parameters such as Attenuation Coefficient (AC), Speed of Sound (SoS), Effective Scatterer Diameter (ESD), and Effective Scatterer Concentration (ESC). The study found that AC, SoS, and ESD showed significant differences between benign and malignant lesions, while ESC did not. A logistic regression model developed using these parameters achieved a high level of accuracy in distinguishing between benign and malignant lesions. However, further studies are needed to validate these findings and optimize the system for clinical use. [Extracted from the article]

A study conducted at Seoul National University Hospital in South Korea examined the accuracy of preoperative imaging in predicting tumor size in breast cancer patients who underwent neoadjuvant chemotherapy. The researchers found that the estimated tumor extent after chemotherapy often did not match the actual pathological extent, leading to unnecessary mastectomies for patients who could have been eligible for breast-conserving surgery. The study also found that magnetic resonance imaging (MRI) contributed the most to false size predictions. The findings highlight the importance of improving imaging techniques to avoid unnecessary surgeries. [Extracted from the article]

A recent study conducted by researchers at the National Institute of Technology Rourkela in Odisha, India, has proposed an efficient deep learning-based ensemble classifier for the detection of breast cancer using mammogram and ultrasound breast images. The study highlights the importance of early and accurate breast cancer detection for proper treatment planning. The proposed model integrates three transfer learning models and utilizes various techniques to improve performance. The experimental results demonstrate high accuracy in detecting abnormality and malignancy in both mammogram and ultrasound datasets. The researchers suggest that the proposed model can be used for breast cancer detection in multimodal datasets. [Extracted from the article]

A report from the University of Blida 1 in Algeria discusses the use of transfer learning and convolutional neural networks for the classification of ultrasound breast cancer data. The researchers propose a new adaptive pre-trained convolutional neural network model that analyzes two public datasets to distinguish between malignant and benign breast tumors. The results show that the Inception V3 model is the most efficient, with high accuracy, specificity, sensitivity, and F1-score. This research has been peer-reviewed and provides valuable insights into the use of emerging technologies for computer-aided diagnosis of breast cancer. [Extracted from the article]

A recent study conducted at Bundang Jesaeng General Hospital in South Korea explored the use of ultrasonography for diagnosing and classifying silicone breast implant ruptures. The study found that high-resolution ultrasonography was a reliable method for detecting ruptures and determining the extent of silicone migration. The researchers recommended updating the current binary classification system to include three types of ruptures based on pathology. This research is important for improving the accuracy of diagnosing and treating silicone breast implant-related issues. [Extracted from the article]

A research study conducted by Central South University in Changsha, China, has developed a new framework called Swin-Net for the segmentation of breast tumor ultrasound images. The framework combines Transformer and Convolutional Neural Networks (CNNs) to improve the accuracy of breast ultrasound segmentation. The Swin-Net framework utilizes a swin-transformer encoder, feature refinement and enhancement module, and hierarchical multi-scale feature fusion module to extract precise features, refine feature maps, and achieve effective cross-layer feature fusion. Experimental results show that Swin-Net outperforms other methods on benchmark datasets, demonstrating its advancements in breast ultrasound image segmentation. The study also provides a new dataset of breast ultrasound images to further validate the effectiveness of the Swin-Net framework. [Extracted from the article]

A recent study conducted by researchers at St. Joseph's Institute of Technology in Tamil Nadu, India, focuses on the detection of benign and malignant breast cancer in ultrasound images. The study proposes a Convolutional Neural Network (CNN) framework that utilizes deep feature mining and handcrafted features to classify breast ultrasound images. Experimental results show that the proposed hybrid features achieve a classification accuracy of 99.50% with the K-Nearest Neighbor (KNN) classifier. The study emphasizes the significance of the proposed scheme for the early diagnosis and treatment of breast cancer. [Extracted from the article]

A systematic review conducted by researchers at Southern Medical University examined the accuracy of deep learning (DL) in ultrasound diagnosis of breast cancer compared to human readers. The review included 16 studies involving 9,238 female participants. The results varied across the studies, with DL showing lower sensitivities in some cases and outperforming human readers in others. However, the researchers concluded that current evidence is insufficient to determine whether DL outperforms human readers or enhances the accuracy of diagnostic breast ultrasound in a clinical setting. They emphasized the need for standardization of study methodologies to improve reproducibility and generalizability of DL research. [Extracted from the article]

Researchers at Kunming Medical University in Yunnan, China have developed a new method for segmenting breast tumors from ultrasound images. The method, called BCUIS-Net, addresses challenges such as indistinct lesion boundaries and noise artifacts. It incorporates a boundary-aware module to accurately locate tumor regions and boundaries, and a shape feature fusion module to extract complementary features and correct errors in segmentation. The network was evaluated on two public breast ultrasound datasets and demonstrated superior segmentation results compared to existing methods. This research has been peer-reviewed and published in the International Journal of Imaging Systems and Technology. [Extracted from the article]

A study conducted by researchers at Chongqing Medical University has developed a neural network with a human learning paradigm for the segmentation of breast fibroadenoma in sonography. Fibroadenoma is a significant health concern, especially for young women, and computer-aided diagnosis has proven to be an effective method for early and accurate detection of solid tumors. The study utilized a local dataset of 600 breast ultrasound images and a public dataset of 246 images to evaluate the performance of the proposed network. The results showed that the human learning paradigm improved segmentation accuracy and reduced training time, demonstrating its potential in medical image processing. [Extracted from the article]

A study conducted at Technical University Cluj Napoca in Romania explores the use of deep learning models and transfer learning to classify breast tissue from ultrasound images into three categories: benign, malignant, and normal. The researchers compared the performance of six state-of-the-art models and introduced a custom model trained on a dataset of 780 images. The results showed that the models trained on an augmented dataset with 7800 images achieved the highest accuracy. The proposed model outperformed some existing models in terms of accuracy and training time. This research provides valuable insights into the automation of early breast cancer detection and classification. [Extracted from the article]

Researchers at Hokuto Hospital in Hokkaido, Japan have conducted a study to evaluate the potential usefulness of a coronal view obtained with an automated breast ultrasound system (ABUS) in detecting breast lesions. The study included 114 ABUS cases and involved ten physicians interpreting the coronal and transverse views independently. The researchers found that the coronal view was associated with significantly shorter reading times compared to the transverse view, while maintaining breast lesion detection accuracy. This study provides new insights into breast cancer screening and the potential benefits of using ABUS in combination with screening mammography. [Extracted from the article]

A researcher at Heidelberg University Hospital in Germany has conducted a study on the use of shear wave elastography (SWE) as a supplemental tool in assessing unsuspicious axillary lymph nodes in patients undergoing breast ultrasound examination. The study involved evaluating 177 axillary lymph nodes in 175 women using SWE. The results showed that SWE is a reliable method for quantifying tissue stiffness in axillary lymph nodes, and these findings could help in identifying lymph node metastasis in breast cancer patients when combined with other lymph node features. The study provides reference values for SWE in unsuspicious axillary lymph nodes. [Extracted from the article]

A researcher from Tokyo Medical and Dental University Hospital conducted a study comparing the clinical properties of original breast ultrasound images and those synthesized by a generative adversarial network (GAN). The study found that the GAN-synthesized images closely resembled the original ultrasound images in clinical characteristics, suggesting their potential utility in clinical education and training for enhancing diagnostic skills in breast ultrasound imaging. The research was supported by the Japan Society For The Promotion of Science. [Extracted from the article]

A patent has been issued to Koninklijke Philips N.V. for an ultrasound system and method that improves the correlation between ultrasound breast images and images acquired by other imaging modalities. The system includes a user interface with a display and a user input device, a memory, and a processor. It uses a deformation model to predict the location of a lesion when imaging the tissue using ultrasound, based on previously acquired image data from another modality. The system provides visual indications of the predicted lesion location and suggested position of the ultrasound probe, and allows for modification of the deformation model based on the actual lesion location. [Extracted from the article]