Conventional diagnostic ultrasound images of the anatomy (as opposed to blood flow) reveal differences in the acoustic properties of soft tissues (mainly echogenicity but also, to some extent, attenuation), whereas ultrasound-based elasticity images are able to reveal the differences in the elastic properties of soft tissues (e.g., elasticity and viscosity). The benefit of elasticity imaging lies in the fact that many soft tissues can share similar ultrasonic echogenicities but may have different mechanical properties that can be used to clearly visualize normal anatomy and delineate pathologic lesions. Typically, all elasticity measurement and imaging methods introduce a mechanical excitation and monitor the resulting tissue response. Some of the most widely available commercial elasticity imaging methods are 'quasi-static' and use external tissue compression to generate images of the resulting tissue strain (or deformation). In addition, many manufacturers now provide shear wave imaging and measurement methods, which deliver stiffness images based upon the shear wave propagation speed. The goal of this review is to describe the fundamental physics and the associated terminology underlying these technologies. We have included a questions and answers section, an extensive appendix, and a glossary of terms in this manuscript. We have also endeavored to ensure that the terminology and descriptions, although not identical, are broadly compatible across the WFUMB and EFSUMB sets of guidelines on elastography (Bamber et al. 2013; Cosgrove et al. 2013).

The World Federation for Ultrasound in Medicine and Biology (WFUMB) has produced these guidelines for the use of elastography techniques in liver disease. For each available technique, the reproducibility, results, and limitations are analyzed, and recommendations are given. Finally, recommendations based on the international literature and the findings of the WFUMB expert group are established as answers to common questions. The document has a clinical perspective and is aimed at assessing the usefulness of elastography in the management of liver diseases.

This paper performs a comprehensive study on the deep-learning-based computer-aided diagnosis (CADx) for the differential diagnosis of benign and malignant nodules/lesions by avoiding the potential errors caused by inaccurate image processing results (e.g., boundary segmentation), as well as the classification bias resulting from a less robust feature set, as involved in most conventional CADx algorithms. Specifically, the stacked denoising auto-encoder (SDAE) is exploited on the two CADx applications for the differentiation of breast ultrasound lesions and lung CT nodules. The SDAE architecture is well equipped with the automatic feature exploration mechanism and noise tolerance advantage, and hence may be suitable to deal with the intrinsically noisy property of medical image data from various imaging modalities. To show the outperformance of SDAE-based CADx over the conventional scheme, two latest conventional CADx algorithms are implemented for comparison. 10 times of 10-fold cross-validations are conducted to illustrate the efficacy of the SDAE-based CADx algorithm. The experimental results show the significant performance boost by the SDAE-based CADx algorithm over the two conventional methods, suggesting that deep learning techniques can potentially change the design paradigm of the CADx systems without the need of explicit design and selection of problem-oriented features.

No abstract