Suppr超能文献

组织中具有非均一声速的内窥光声层析成像的图像重建方法。

An image reconstruction method for endoscopic photoacoustic tomography in tissues with heterogeneous sound speed.

机构信息

Department of Electronic and Communication Engineering, North China Electric Power University, Baoding, 071003, China.

Department of Electronic and Communication Engineering, North China Electric Power University, Baoding, 071003, China.

出版信息

Comput Biol Med. 2019 Jul;110:15-28. doi: 10.1016/j.compbiomed.2019.05.008. Epub 2019 May 10.

Abstract

The idealized assumption of a constant speed of sound (SOS) in acoustically inhomogeneous biological tissues usually results in blurred details, acoustic distortion and artifacts in in vivo endoscopic photoacoustic tomographic (EPAT) images. In this paper, we propose an image reconstruction method to improve EPAT imaging for luminal structures with the variable SOS. In our method, an optimal SOS providing the maximal local focusing of a measuring location within the imaging region is firstly determined. The deviation in the ultrasonic propagation time caused by the variable SOS is then compensated. The grayscale images of the optical absorption distribution on the cross-sections of the luminal structures are finally reconstructed with a filtered back-projection (FBP) algorithm based on the corrected propagation time. Any prior knowledge of the SOS distribution in the imaged tissues is not required. The results of numerical simulation experiments demonstrated that the proposed method can effectively improve the image quality by reducing the misalignment of tissues, acoustic distortion and artifacts caused by the variable SOS.

摘要

在声不均匀的生物组织中,理想情况下假设声速(SOS)恒定,通常会导致体内内窥镜光声层析成像(EPAT)图像中的细节模糊、声失真和伪影。本文提出了一种图像重建方法,用于改善具有可变 SOS 的腔道结构的 EPAT 成像。在我们的方法中,首先确定在成像区域内提供测量位置最大局部聚焦的最佳 SOS。然后补偿由可变 SOS 引起的超声传播时间偏差。最后,基于校正后的传播时间,使用基于滤波反投影(FBP)算法重建腔道结构横截面的光学吸收分布的灰度图像。不需要事先了解成像组织中的 SOS 分布。数值模拟实验结果表明,该方法可以通过减少由可变 SOS 引起的组织错位、声失真和伪影,有效提高图像质量。

文献AI研究员

20分钟写一篇综述,助力文献阅读效率提升50倍。

立即体验

用中文搜PubMed

大模型驱动的PubMed中文搜索引擎

马上搜索

文档翻译

学术文献翻译模型,支持多种主流文档格式。

立即体验