Focused ultrasound (FUS) represents an emerging non-invasive technique capable of achieving reversible and targeted opening of the blood-brain barrier (BBB) when combined with microbubbles (MBs), significantly enhancing drug permeability. This technology demonstrates considerable potential in preclinical studies for diagnosing and treating neurological disorders. Ensuring safe and effective BBB opening while preventing tissue injuries such as red blood cell extravasation remains a critical translational challenge. This review examines the mechanisms of FUS combined with MBs induced BBB opening, key influencing factors, and recent advances in strategies for monitoring and controlling in vivo cavitation activity, providing foundational insights for future research.

Cervical cancer is a common gynecologic malignancy worldwide, ranking fourth for both incidence and mortality. Imaging and pathology assessments are incorporated in the revised 2018 International Federation of Gynecology and Obstetrics staging system for cervical cancer. The use of imaging techniques for pre-operative evaluation of cervical cancer has been increasing. Among imaging modalities for evaluating cervical cancer, ultrasound is more easily accessible, faster and more widely available than other options such as computed tomography or magnetic resonance imaging. Advanced technique in ultrasound, such as three-dimension ultrasound and contrast-enhanced ultrasound, have improved the clinical application of ultrasound in cervical cancer. Ultrasound may provide highly accurate information on detecting tumor presence and assessing local extent if performed by well-trained sonographers, as the experience level of readers is also critical for correct pre-operative staging and evaluation of treatment response. In the future, ultrasound imaging with the assistance of artificial intelligence will play an even greater role in management. This review aims to present the most updated applications of ultrasound in the pre-operative evaluation of cervical cancer.

Since 2020, breast cancer has held the highest incidence rate among cancers worldwide. Breast ultrasound (US) imaging technology plays a crucial role in the early diagnosis and intervention treatment of breast cancer patients. Deep learning (DL), as one of the most powerful machine learning techniques in the field of artificial intelligence (AI), has the ability to automatically select features from raw data, achieving remarkable advancements in breast US imaging. This review focuses on the application of convolutional neural networks (CNNs) within DL technology in the field of breast US. It summarizes the use of DL models in breast cancer screening and in preoperative prediction of molecular subtypes, response to neoadjuvant chemotherapy (NAC), and axillary lymph node (ALN) metastasis status. The review also identifies the data limitations of using CNN models in breast US and describes the development history and current applications of DL in breast cancer screening, diagnostic guidance, and prognostic prediction. Furthermore, it discusses the future research directions and potential challenges. Advancing the development of CNN technology in breast US, and improving the generalizability and reproducibility of these models, will significantly promote their translational application in clinical settings.