Share Email Print

Proceedings Paper

Adaptive photoacoustic imaging quality optimization with EMD and reconstruction
Author(s): Chengwen Guo; Yao Ding; Jie Yuan; Guan Xu; Xueding Wang; Paul L. Carson
Format Member Price Non-Member Price
PDF $17.00 $21.00

Paper Abstract

Biomedical photoacoustic (PA) signal is characterized with extremely low signal to noise ratio which will yield significant artifacts in photoacoustic tomography (PAT) images. Since PA signals acquired by ultrasound transducers are non-linear and non-stationary, traditional data analysis methods such as Fourier and wavelet method cannot give useful information for further research. In this paper, we introduce an adaptive method to improve the quality of PA imaging based on empirical mode decomposition (EMD) and reconstruction. Data acquired by ultrasound transducers are adaptively decomposed into several intrinsic mode functions (IMFs) after a sifting pre-process. Since noise is randomly distributed in different IMFs, depressing IMFs with more noise while enhancing IMFs with less noise can effectively enhance the quality of reconstructed PAT images. However, searching optimal parameters by means of brute force searching algorithms will cost too much time, which prevent this method from practical use. To find parameters within reasonable time, heuristic algorithms, which are designed for finding good solutions more efficiently when traditional methods are too slow, are adopted in our method. Two of the heuristic algorithms, Simulated Annealing Algorithm, a probabilistic method to approximate the global optimal solution, and Artificial Bee Colony Algorithm, an optimization method inspired by the foraging behavior of bee swarm, are selected to search optimal parameters of IMFs in this paper. The effectiveness of our proposed method is proved both on simulated data and PA signals from real biomedical tissue, which might bear the potential for future clinical PA imaging de-noising.

Paper Details

Date Published: 31 October 2016
PDF: 9 pages
Proc. SPIE 10024, Optics in Health Care and Biomedical Optics VII, 100241S (31 October 2016); doi: 10.1117/12.2244916
Show Author Affiliations
Chengwen Guo, Nanjing Univ. (China)
Yao Ding, Nanjing Univ. (China)
Jie Yuan, Nanjing Univ. (China)
Guan Xu, Univ. of Michigan (United States)
Xueding Wang, Univ. of Michigan (United States)
Paul L. Carson, Univ. of Michigan (United States)

Published in SPIE Proceedings Vol. 10024:
Optics in Health Care and Biomedical Optics VII
Qingming Luo; Xingde Li; Ying Gu; Yuguo Tang, Editor(s)

© SPIE. Terms of Use
Back to Top