On the Role of Coherent Plane Wave Compounding in Shear Wave Elasticity Imaging: The Convolution Effect and Its Implications

ULTRASOUND IN MEDICINE AND BIOLOGY(2024)

引用 0|浏览8
暂无评分
摘要
Objective: The clinical applicability of shear wave elasticity imaging (SWEI) has been confounded by its appreciable intersystem variability and unsatisfactory sensitivity. While SWEI relies on plane wave imaging (PWI) to achieve real-time rendering, it has been rarely noticed that PWI can affect SWEI's performance. This work is aimed at demonstrating that the use of coherent plane wave compounding (CPWC) can be a factor causing SWEI's underperformance. Methods: We presented a model to formally describe the slow-time behavior of CPWC in motion tracking. This model reveals that CPWC introduces temporal convolution on the observed motion, making the motion sampling process a low-pass filter (LPF). For validation, shear waves were produced in a phantom in the same way but sampled via PWI using different compounding numbers (CN) and pulse repetition frequencies (PRF), with the obtained signals compared with the inferences drawn from our model. Similar experiments were performed to reconstruct two small targets in the phantom in order to appraise the impact of CPWC on SWEI's sensitivity. Discussion: The validation experiment shows that the measurements match well with the model infer-ences, which verifies the LPF nature of CPWC. The phantom study also shows that either increasing CN or decreas-ing PRF can cause the loss of high-frequency motion information, leading to blurred target delineation by SWEI. Conclusion: The convolution effect can help understand the variability of SWEI. Researchers should beware this effect when working on SWEI standardization. Clinicians using SWEI should also be cautious because this effect makes it harder to identify small lesions.
更多
查看译文
关键词
Shear wave elasticity imaging,convolution effect,plane wave imaging,coherent plane wave compounding,motion tracking
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要