Sparse Image Reconstruction for Contrast Enhanced Cardiac Ultrasound using Diverging Waves
Sparse Image Reconstruction for Contrast Enhanced Cardiac Ultrasound using Diverging Waves
复制标题
DOI：
10.1109/ultsym.2019.8925732
复制
发表时间：
2019-10
期刊：
影响因子：
--
通讯作者：
A. Stanziola;M. Toulemonde;V. Papadopoulou;R. Corbett;N. Duncan;E. Grisan;M. Tang
中科院分区：
文献类型：
--
作者：
A. Stanziola;M. Toulemonde;V. Papadopoulou;R. Corbett;N. Duncan;E. Grisan;M. Tang
经胸超声评估心功能是一项具有挑战性的任务，主要是由于其快速运动和其解剖位置仅允许狭窄的肋间成像窗口。这些因素经常导致使用散波，即使使用造影剂。虽然能够实现非常高的帧率，但从窄光圈发散波的采集会导致严重的图像质量下降。在这方面，通常不可能使用常见的处理方法来缓解这个问题，例如相干复合。在本研究中，我们将重建对比度增强超声图像的问题视为正则化逆问题，类似于压缩感知问题，其中感知矩阵基本上由与飞行时间相关的延迟算子描述。结果表明，与延迟和和（DAS）相比，该框架可将图像的信噪比（SNR）提高5.85dB，因此是重建对比度增强心脏图像的一种很有前途的方法。实验还强调了噪声建模的方式对最终图像质量有重大影响。
Assessing cardiac function with trans-thoracic ultrasound is a challenging task, mainly due to its fast motion and its anatomical position which only allows for a narrow intercostal imaging window. These factors often lead to the use of diverging waves, even when contrast agents are employed. While capable of achieving a very high frames rate, an acquisition with diverging waves from a narrow aperture suffers from serious image quality degradation. In this regard, it is often impossible to mitigate this problem using common processing methods, such as coherent compounding.In this study, we cast the problem of reconstructing the contrast enhanced ultrasound images as regularised inverse problem, analogous to the compressed sensing one, where the sensing matrix is fundamentally described by the delay operator associated with the time of flight. The results show that this framework can improve the Signal to Noise Ratio (SNR) of the image by up to 5.85dB compared to delay and sum (DAS) and is therefore a promising way to reconstruct contrast enhanced cardiac images. The experiments also highlight that the way noise is modelled has a significant impact on the final image quality.