Advances in Ultrasound Imaging and Deep Learning for the Diagnosis and Treatment of Breast and Thyroid Cancers

Ultrasound is one of the most widely used medical imaging modalities in hospitals worldwide and is applied across nearly all fields of medicine. In recent years, several novel techniques have emerged, including three-dimensional ultrasound (3D US), contrast-enhanced ultrasound (CEUS), ultrafast imaging, elastography, and ultrasound tomography. These advances have substantially broadened the scope of ultrasound-assisted diagnosis and therapeutic intervention.Cancer remains one of the most prevalent diseases worldwide and places a heavy burden on patients, their families, and healthcare insurance systems. Breast and thyroid cancers are among the most common malignancies. Early diagnosis is critical for effective management, as it reduces mortality and improves five-year survival rates. With recent advances in deep learning, several studies have reported diagnostic accuracy and robustness comparable to those of experienced clinicians. Such artificial intelligence (AI) tools are especially valuable for primary and county-level hospitals, where specialist resources are limited.This Research Topic aims to gather state-of-the-art research that bridges ultrasound imaging, deep learning methods, and the clinical management of breast and thyroid cancers. It seeks to highlight technical innovations in lesion detection, segmentation, risk stratification, prognosis prediction, and image-guided therapy. It also aims to address practical bottlenecks, such as model generalizability, interpretability, and real-world clinical deployment. By bringing together radiologists, oncologists, biomedical engineers, and computer scientists, this Research Topic aims to consolidate current evidence, identify remaining challenges, and promote the clinical translation of ultrasound-based deep learning tools. The ultimate goals are to improve early diagnosis, reduce unnecessary interventions, and optimize patient outcomes.We welcome Original Research, Reviews, Brief Research Reports, and Case Reports on ultrasound imaging combined with deep learning for the diagnosis and treatment of breast and thyroid cancers. Topics of interest include, but are not limited to:• Ultrasound-based lesion detection and segmentation• Risk stratification and malignancy classification• Multimodal image fusion• Prognostic evaluation• Image-guided interventional therapy• Model interpretability and domain adaptation• Dataset construction and real-world clinical validationStudies that address model interpretability, domain adaptation, dataset construction, and real-world clinical validation are especially encouraged. Topic editors: MINGYUE Ding, Chuan Zhou, Xin YANG Source: https://www.frontiersin.org/research-topics/87378/advances-in-ultrasound-imaging-and-deep-learning-for-the-diagnosis-and-treatment-of-breast-and-thyroid-cancers