Improving Multimodal Ultrasound Diagnostics In Non-Palpable Breast Lesions
Abstract
Non-palpable breast lesions represent a significant diagnostic challenge in breast imaging due to their occult nature on clinical examination and variable presentations on conventional modalities. Multimodal ultrasound, integrating B-mode imaging with advanced techniques such as shear wave elastography (SWE), strain elastography, contrast-enhanced ultrasound (CEUS), and microvascular imaging, offers a promising non-invasive approach to enhance detection, characterization, and differentiation of benign and malignant lesions. This article explores advancements in multimodal ultrasound protocols, emphasizing their role in improving diagnostic accuracy, reducing unnecessary biopsies, and optimizing clinical management for non-palpable breast lesions. By combining morphological, vascular, and biomechanical data, multimodal ultrasound achieves superior sensitivity and specificity compared to conventional ultrasound alone, with reported areas under the curve (AUC) often exceeding 0.90 in combined approaches. Key innovations include quantitative stiffness measurements via SWE, perfusion pattern analysis through CEUS, and integration with artificial intelligence for automated feature extraction. These developments not only refine BI-RADS categorization but also support personalized diagnostic pathways, particularly in dense breast tissue where mammography sensitivity is limited. Future directions involve hybrid imaging, real-time fusion techniques, and machine learning models to further elevate precision and reproducibility. This comprehensive review underscores the transformative potential of multimodal ultrasound in early breast cancer detection and lesion characterization.