Variable speed of sound compensation in the linear-array photoacoustic tomography using a multi-stencils fast marching method. (July 2018)

• Record Type: Journal Article
• Authors: Lin, Xiangwei; Sun, Mingjian; Liu, Yang; Shen, Zhiyuan; Shen, Yi; Feng, Naizhang
• Is Part Of: Biomedical signal processing and control. Volume 44(2018)
• Page Start: 67
• Page End: 74
• Publication Date: 2018-07
• Subjects: Linear-array photoacoustic tomography -- Variable speed of sound -- Delay and sum -- Multi-stencils fast marching method; Signal processing -- Periodicals; Biomedical engineering -- Periodicals; Signal Processing, Computer-Assisted -- Periodicals; Image Processing, Computer-Assisted -- Periodicals; Biomedical Engineering -- Periodicals; 610.28
• DOI: 10.1016/j.bspc.2018.04.012
• Languages: English
• ISSNs: 1746-8094

Abstract

Highlights: The combination of MSFM and DAS reconstruction method to compensate the variable speed of sound effect on the conventional DAS reconstruction in the linear-array PAT. The application of DAS-MSFM method to the PA detection of mouse deep brain tumor with an intact skull. Maybe it is the first time to apply the combined DAS and MSFM method for the in vivo PA imaging of brain tumors. To detect the brain tumor with an intact skull, numerical model, phantom, and in vivo PA experiment were sequentially designed. The proposed method could be potentially used in the human study to precisely locate the tumor position and accurately measure the tumor size. Abstract: Despite the promising clinical application of linear-array photoacoustic tomography, it has been shown the variable speed of sound would severely affect the photoacoustic imaging quality, resulting in the target deterioration and inaccurate depth positioning in the conventional constant speed assumed delay and sum (DAS) reconstruction. By contrast, multi-stencils fast marching (MSFM) method is able to produce accurate time delay map for the tissue with inhomogeneous acoustic speed. This study explored the combination of DAS reconstruction algorithm with the MSFM approach to reduce the imaging distortions due to the speed spatial variation, where the target structure and target position in depth could be measured more precisely. To validate the performance of the proposed method, numerical, phantom, and in vivo photoacoustic studies were conducted with the qualitative and quantitative analysis, especially in the detection of mouse deep brain tumor with an intact skull.