生物组织光声成像
摘要
光声成像是一种基于光声效应的新型医学成像方法,它结合了超声成像高空间分辨率和光学成像高对比度的优点。近十年来,光声成像已经成为当今国内外研究的热点课题,并得到飞速发展。光声成像利用脉冲激光照射生物组织,同时利用超声换能器接收光声信号,再从收集的光声信号中反演出光吸收系数分布情况。目前大多数研究都假设换能器可以通过全方位扫描来接收包围成像区域的闭合曲面上的完整数据,并且生物组织具有恒定的声学参数以至于可以忽略光声信号的散射。然而,在实际中,全方位扫描往往是不能实现的,并且生物组织的密度和声速往往也是不均匀的。
本文以有限视角扫描光声重构和声散射媒质中的光声成像作为研究重点,进行了深入系统的研究,并取得了以下研究成果：
本文首先从成像分辨率和成像深度两个角度研究了有限视角扫描对目前广泛应用的反投影重构方法的影响。结果表明有限视角扫描会造成重构图像的畸变,使两个原本独立的物体在重构图像中连接在一起而无法分辨,从而降低成像分辨率。另一方面,即使在物体具有相同强度情况下,有限视角扫描会使深处的物体的重构强度弱于浅处的物体,以至于深处的物体因为重构强度太弱而从重构图像中消失,从而减小成像深度；针对有限视角扫描导致的成像深度问题,本文提出了有效扫描角度的原理,并分析和改进了有效视角情况下光声成像深度。
本文发现用基于均匀媒质假设的重构算法对声散射媒质成像,在重构图像中会出现伪像和畸变；提出了基于声波传播的时间反转不变性的光声重构方法。该方法将接收到的光声信号时间反转后,再发射到一个生物组织的模型中,最终的时间反转声场可以反映由于光声效应而产生的初始声压,进而可以用来重构光声图像。本文详细的分析了模型中散射体的位置和大小误差对时间反转重构方法的影响,结果表明在较大的误差范围之内,重构图像的质量高于基于均匀媒质假设的重构方法,并且重构图像的质量对浅处散射体的误差更加敏感。本文同时指出可以利用超声成像的方法来建立生物组织的模型；由于成像深度有限,超声成像可能无法准确确定甚至完全忽略深处的散射体,数值模拟结果表明即使在这种情况下,时间反转方法依然可以得到满意的重构图像。
目前关于光声成像的研究大多是基于均匀媒质假设,它们认为光声信号的散射会导致重构图像的质量降低。本文提出了一种利用背向散射来提高有限视角光声成像的方法,其核心思想是利用成像区域背后的散射体或者人为放置的背后散射体(如注入声造影剂),使原来向背后传播而无法被换能器接收到的光声信号经散射后最终被换能器接收；同时通过时间反转的方法将这些散射声重新汇聚至吸收体的位置,将原来缺失的背向信息在重构图像中补全,进而提高成像质量。在这个方法中,散射体起到虚拟换能器的作用,即接收散射声并在时间反转过程中发射时间反转后的散射声。本文实验验证了该方法的有效性,结果表明在换能器阵列张角或者换能器扫描角度较小的情况下,该方法可以显著地提高有限视角扫描重构的图像质量。
Photoacoustic tomography (PAT) is a novel biomedical imaging based on photoacoustic effect. It can combine the excellent spatial resolution in ultrasound imaging and the high contrast in optical imaging, and it has been attracting more and more attention. In PAT, pulsed laser is employed to radiate tissues to generate ultrasound, and the ultrasound is recorded by transducer to reconstruct the distribution of optical absorption within the tissues. It is usually assumed that transducer could record the complete data by full-view scanning and the tissues are acoustically homogeneous. However, in practical, transducer cannot scan in full-view way but limited-view way, and the tissues are inhomogeneous. To this end, the dissertation focuses on the limited-view reconstruction and PAT of acoustical scattering tissues. The major work of this dissertation is described as follows:
This dissertation quantitatively studied the influence of the limited-view scanning on the imaging resolution and depth of the back-projection method. It was found that the limited-view scanning could give rise to artifact and distortion which makes two individual absorbers connect with each other and not be separated in the reconstructed image, therefore degrading the resolution. Moreover, the limited-view scanning could also make the reconstructed intensity of deep absorber much weaker than that of the shallow absorber so that the deep absorber cannot be clearly indicated and become indiscernible in the image. The concept of effective scanning angel is proposed to analyze the degradation of imaging depth, and one method based on that concept is proposed to improve imaging depth of limited-view PAT.
It also was found that methods based on homogeneity assumption could bring about artifact and distortion when imaging acoustically scattering tissues. This dissertation proposed a method based on the time reversal invariance of the propagation of photoacoustics. In this method, the recorded photoacoustic signals are time-reversed and reemitted into a model medium which has the same deposition of scatterers as the realistic medium, the eventual time reversal field could reveal the initial pressure induced by photoacoustic effect and therefore could be used for imaging. The mismatch between the model medium and the realistic medium was also considered to verify its practicality and robustness; it was shown that the mismatch will degrade the image quality, but within a wide range of mismatch, the method is still superior to the homogeneity-based methods like delay-and-sum method. Ultrasound imaging is a good candidate for establishing such medium model, and it is possible that the deep scatterers cannot be accurately determined and can even be neglected; however, further investigation showed that the method could also provide satisfactory result using a medium model ignoring the deep scatterers.
Acoustical scattering is generally considered as nuisance in PAT because it distorts the propagation of photoacoustic waves. However, this dissertation showed that scattering could also contain useful information for imaging and proposed to use backscattering to improve limited-view PAT. The method employed native scatterers or artificial ones behind the region of interest to make the back-propagating signals accessible to the transducer after backscattering. After time reversal, the backscattered waves could converge to the source and provide backside information lost by direct waves, therefore improving image quality. The practicality and robustness of this method is verified by both simulations and experiments, it was shown that this method could significantly improve the limited-view reconstruction, especially when the aperture of the transducer array is quite limited.
本文以有限视角扫描光声重构和声散射媒质中的光声成像作为研究重点,进行了深入系统的研究,并取得了以下研究成果：
本文首先从成像分辨率和成像深度两个角度研究了有限视角扫描对目前广泛应用的反投影重构方法的影响。结果表明有限视角扫描会造成重构图像的畸变,使两个原本独立的物体在重构图像中连接在一起而无法分辨,从而降低成像分辨率。另一方面,即使在物体具有相同强度情况下,有限视角扫描会使深处的物体的重构强度弱于浅处的物体,以至于深处的物体因为重构强度太弱而从重构图像中消失,从而减小成像深度；针对有限视角扫描导致的成像深度问题,本文提出了有效扫描角度的原理,并分析和改进了有效视角情况下光声成像深度。
本文发现用基于均匀媒质假设的重构算法对声散射媒质成像,在重构图像中会出现伪像和畸变；提出了基于声波传播的时间反转不变性的光声重构方法。该方法将接收到的光声信号时间反转后,再发射到一个生物组织的模型中,最终的时间反转声场可以反映由于光声效应而产生的初始声压,进而可以用来重构光声图像。本文详细的分析了模型中散射体的位置和大小误差对时间反转重构方法的影响,结果表明在较大的误差范围之内,重构图像的质量高于基于均匀媒质假设的重构方法,并且重构图像的质量对浅处散射体的误差更加敏感。本文同时指出可以利用超声成像的方法来建立生物组织的模型；由于成像深度有限,超声成像可能无法准确确定甚至完全忽略深处的散射体,数值模拟结果表明即使在这种情况下,时间反转方法依然可以得到满意的重构图像。
目前关于光声成像的研究大多是基于均匀媒质假设,它们认为光声信号的散射会导致重构图像的质量降低。本文提出了一种利用背向散射来提高有限视角光声成像的方法,其核心思想是利用成像区域背后的散射体或者人为放置的背后散射体(如注入声造影剂),使原来向背后传播而无法被换能器接收到的光声信号经散射后最终被换能器接收；同时通过时间反转的方法将这些散射声重新汇聚至吸收体的位置,将原来缺失的背向信息在重构图像中补全,进而提高成像质量。在这个方法中,散射体起到虚拟换能器的作用,即接收散射声并在时间反转过程中发射时间反转后的散射声。本文实验验证了该方法的有效性,结果表明在换能器阵列张角或者换能器扫描角度较小的情况下,该方法可以显著地提高有限视角扫描重构的图像质量。
Photoacoustic tomography (PAT) is a novel biomedical imaging based on photoacoustic effect. It can combine the excellent spatial resolution in ultrasound imaging and the high contrast in optical imaging, and it has been attracting more and more attention. In PAT, pulsed laser is employed to radiate tissues to generate ultrasound, and the ultrasound is recorded by transducer to reconstruct the distribution of optical absorption within the tissues. It is usually assumed that transducer could record the complete data by full-view scanning and the tissues are acoustically homogeneous. However, in practical, transducer cannot scan in full-view way but limited-view way, and the tissues are inhomogeneous. To this end, the dissertation focuses on the limited-view reconstruction and PAT of acoustical scattering tissues. The major work of this dissertation is described as follows:
This dissertation quantitatively studied the influence of the limited-view scanning on the imaging resolution and depth of the back-projection method. It was found that the limited-view scanning could give rise to artifact and distortion which makes two individual absorbers connect with each other and not be separated in the reconstructed image, therefore degrading the resolution. Moreover, the limited-view scanning could also make the reconstructed intensity of deep absorber much weaker than that of the shallow absorber so that the deep absorber cannot be clearly indicated and become indiscernible in the image. The concept of effective scanning angel is proposed to analyze the degradation of imaging depth, and one method based on that concept is proposed to improve imaging depth of limited-view PAT.
It also was found that methods based on homogeneity assumption could bring about artifact and distortion when imaging acoustically scattering tissues. This dissertation proposed a method based on the time reversal invariance of the propagation of photoacoustics. In this method, the recorded photoacoustic signals are time-reversed and reemitted into a model medium which has the same deposition of scatterers as the realistic medium, the eventual time reversal field could reveal the initial pressure induced by photoacoustic effect and therefore could be used for imaging. The mismatch between the model medium and the realistic medium was also considered to verify its practicality and robustness; it was shown that the mismatch will degrade the image quality, but within a wide range of mismatch, the method is still superior to the homogeneity-based methods like delay-and-sum method. Ultrasound imaging is a good candidate for establishing such medium model, and it is possible that the deep scatterers cannot be accurately determined and can even be neglected; however, further investigation showed that the method could also provide satisfactory result using a medium model ignoring the deep scatterers.
Acoustical scattering is generally considered as nuisance in PAT because it distorts the propagation of photoacoustic waves. However, this dissertation showed that scattering could also contain useful information for imaging and proposed to use backscattering to improve limited-view PAT. The method employed native scatterers or artificial ones behind the region of interest to make the back-propagating signals accessible to the transducer after backscattering. After time reversal, the backscattered waves could converge to the source and provide backside information lost by direct waves, therefore improving image quality. The practicality and robustness of this method is verified by both simulations and experiments, it was shown that this method could significantly improve the limited-view reconstruction, especially when the aperture of the transducer array is quite limited.
引文
[1]A. G. Bell, Science 1 (12),130-134 (1880).
[2]C. A. S. Lima, M. B. S. Lima, R. Oliva, G. Cardenas and L. C. M. Miranda, Chem. Phys. Lett.332 (5),428-434 (2000).
[3]Y. Yang, X. Liu and S. Zhang, Appl. Phys. Lett.95 (20),3 (2009).
[4]N. Chigarev, J. Zakrzewski, V. Tournat and V. Gusev, J. Appl. Phys.106 (3), (2009).
[5]J. W. Wang, W. Zhang, L. R. Liang and Q. X. Yu, Sensor. Actuat. B-Chem.160 (1),1268-1272(2011).
[6]C. N. Wu and Y. J. Wu, Opt. Commun.185 (4-6),359-362 (2000).
[7]M. Xu and L. V. Wang, Phys. Rev. E 67 (5),15 (2003).
[8]L. Li, R. J. Zemp, G. Lungu, G. Stoica and L. V. Wang, J. Biomed. Opt.12 (2), 3 (2007).
[9]X. M. Yang and L. V. Wang, J. of Biomed. Opt.12 (6),060507 (2007).
[10]C. H. Li and L. V. Wang, Appl. Phys. Lett.93 (3),033902 (2008).
[11]L. V. Wang, Med. Phys.35 (12),5758 (2008).
[12]C. Li and L. V. Wang, Phys. Med. Biol.54,39 (2009).
[13]L. V. Wang and S. Hu, Science 335 (6075),1458-1462 (2012).
[14]C. H. Li, A. Aguirre, J. Gamelin, A. Maurudis, Q. Zhu and L. V. Wang, J. Biomed. Opt.15(1) (2010).
[15]X. Wang, Y. Pang, G. Ku, X. Xie, G. Stoica and L. V. Wang, Nat. Biotechnol. 21 (7),803-806 (2003).
[16]X. Wang, D. L. Chamberland, P. L. Carson, J. B. Fowlkes, R. O. Bude, D. A. Jamadar and B. J. Roessler, Med. Phys.33 (8),2691-2697 (2006).
[17]X. Wang, D. L. Chamberland and D. A. Jamadar, Opt. Lett.32 (20),3002-3004 (2007).
[18]X. Wang, Y. Pang, G. Ku, G. Stoica and L. V. Wang, Opt. Lett.28(19), 1739-1741 (2003).
[19]X. Wang, X. Xie, G. Ku, L. V. Wang and G. Stoica, J. Biomed. Opt.11(2), 024015(2006).
[20]R. I. Siphanto, K. K. Thumma, R. G. M. Kolkman, T. G. van Leeuwen, F. F. M. de Mul, J. W. van Neck, L. N. A. van Adrichem and W. Steenbergen, Opt. Express 13 (1),89-95 (2005).
[21]F. Ye, S. H. Yang and D. Xing, Appl. Phys. Lett.97 (21),213702 (2010).
[22]J. J. Niederhauser, M. Jaeger, R. Lemor, P. Weber and M. Frenz, IEEE T. Med. Imaging 24 (4),436-440 (2005).
[23]R. A. Kruger, P. Y. Liu, Y. R. Fang and C. R. Appledorn, Med.Phys.22 (10), 1605-1609 (1995).
[24]M. Xu and L. V. Wang, IEEE Trans. Med. Imaging 21 (7),814-822 (2002).
[25]M. Xu, Y. Xu and L. V. Wang, IEEE T. Biomed. Eng.50 (9),1086-1099 (2003).
[26]M. Xu and L. V. Wang, Phys. Rev. E 71 (1),016706 (2005).
[27]Y. Xu, L. V. Wang, G. Ambartsoumian and P. Kuchment, Med. Phys.31 (4), 724-733 (2004).
[28]Y. Xu and L. V. Wang, IEEE T. Ultrason. Ferr.50 (9),1134-1146 (2003).
[29]X. Jin and L. V. Wang, Phys. Med. Biol.51 (24),6437-6448 (2006).
[30]R. Courant, K. O. Friedrichs and H. Lewy, IBM J.11,215-234 (1956).
[31]X. C. Pan, Y. Zou and M. A. Anastasio, IEEE T. Image. Process.12 (7), 784-795 (2003).
[32]X. L. Dean-Ben, V. Ntziachristos and D. Razansky, Appl. Phys. Lett.98 (17), 171110(2011).
[33]M. Fink, F. Wu, J.-L. Thomas and M. Fink, IEEE T. Ultrason. Ferroelec. Fre. Contr.39 (5),24 (1992).
[34]J. L. Thomas, F. Wu and M. Fink, Ultrasonic Imaging 18 (2),16 (1996).
[35]M. Fink, D. Cassereau, A. Derode, C. Prada, P. Roux, M. Tanter, J. L. Thomas and F. Wu, Rep. Prog. Phys.63 (12),1933-1995 (2000).
[36]P. Blomgren, G. Papanicolaou and H. Zhao, J. Acoust. Soc. Am. 111 (1),19 (2001).
[37]G. Lerosey, J. de Rosny, A. Tourin, A. Derode, G. Montaldo and M. Fink, Phys. Rev. Lett.92 (19),193904 (2004).
[38]A. Sutin, A. Sarvazyan, P. A. Johnson and J. A. TenCate, J. Acoust. Soc. Am. 115 (5),2384-2384 (2004).
[39]Y. Xu and L. V. Wang, Phys. Rev. Lett.92 (3),033902 (2004).
[40]S. Cathelinea, M. Fink, N. Quieffin and R. K. Ing, Appl. Phys. Lett.90 (6),3 (2007).
[41]M. Fink, G. Lerosey, J. De Rosny and A. Tourin, Science 315 (5815), 1120-1122(2007).
[42]G. Bartal, G. Lerosey and X. Zhang, Phys. Rev. B 79 (20),201103 (2009).
[43]O. Couture, J.-F. Aubry, M. Tanter and M. Fink, Appl. Phys. Lett.94 (17), 173901 (2009).
[44]F. Lemoult, G. Lerosey, J. de Rosny and M. Fink, Phys. Rev. Lett 103 (17), 173902(2009).
[45]C. S. Larmat, R. A. Guyer and P. A. Johnson, Phys. Today 63 (8),31-35 (2010).
[46]C. Draeger and M. Fink, Phys. Rev. Lett.79 (3),407-410 (1997).
[47]B. T. Cox, S. R. Arridge and P. C. Beard, Inverse Probl.23 (6), S95-S112 (2007).
[48]D. Cassereau and M. Fink, IEEE T. Ultrason. Ferroelectr. Freq. Control.39 (5), 579-592(1992).
[49]M. Fink and C. Prada, Inverse Probl.17,38 (2001).
[50]C. Tao and X. Liu, Opt. Express 18 (3),2760 (2010).
[51]I. Vasconcelos, R. Snieder and H. Douma, Phys. Rev. E 80 (3),036605 (2009).
[52]A. Aubry and A. Derode, Phys. Rev. Lett 102 (8),084301 (2009).
[2]C. A. S. Lima, M. B. S. Lima, R. Oliva, G. Cardenas and L. C. M. Miranda, Chem. Phys. Lett.332 (5),428-434 (2000).
[3]Y. Yang, X. Liu and S. Zhang, Appl. Phys. Lett.95 (20),3 (2009).
[4]N. Chigarev, J. Zakrzewski, V. Tournat and V. Gusev, J. Appl. Phys.106 (3), (2009).
[5]J. W. Wang, W. Zhang, L. R. Liang and Q. X. Yu, Sensor. Actuat. B-Chem.160 (1),1268-1272(2011).
[6]C. N. Wu and Y. J. Wu, Opt. Commun.185 (4-6),359-362 (2000).
[7]M. Xu and L. V. Wang, Phys. Rev. E 67 (5),15 (2003).
[8]L. Li, R. J. Zemp, G. Lungu, G. Stoica and L. V. Wang, J. Biomed. Opt.12 (2), 3 (2007).
[9]X. M. Yang and L. V. Wang, J. of Biomed. Opt.12 (6),060507 (2007).
[10]C. H. Li and L. V. Wang, Appl. Phys. Lett.93 (3),033902 (2008).
[11]L. V. Wang, Med. Phys.35 (12),5758 (2008).
[12]C. Li and L. V. Wang, Phys. Med. Biol.54,39 (2009).
[13]L. V. Wang and S. Hu, Science 335 (6075),1458-1462 (2012).
[14]C. H. Li, A. Aguirre, J. Gamelin, A. Maurudis, Q. Zhu and L. V. Wang, J. Biomed. Opt.15(1) (2010).
[15]X. Wang, Y. Pang, G. Ku, X. Xie, G. Stoica and L. V. Wang, Nat. Biotechnol. 21 (7),803-806 (2003).
[16]X. Wang, D. L. Chamberland, P. L. Carson, J. B. Fowlkes, R. O. Bude, D. A. Jamadar and B. J. Roessler, Med. Phys.33 (8),2691-2697 (2006).
[17]X. Wang, D. L. Chamberland and D. A. Jamadar, Opt. Lett.32 (20),3002-3004 (2007).
[18]X. Wang, Y. Pang, G. Ku, G. Stoica and L. V. Wang, Opt. Lett.28(19), 1739-1741 (2003).
[19]X. Wang, X. Xie, G. Ku, L. V. Wang and G. Stoica, J. Biomed. Opt.11(2), 024015(2006).
[20]R. I. Siphanto, K. K. Thumma, R. G. M. Kolkman, T. G. van Leeuwen, F. F. M. de Mul, J. W. van Neck, L. N. A. van Adrichem and W. Steenbergen, Opt. Express 13 (1),89-95 (2005).
[21]F. Ye, S. H. Yang and D. Xing, Appl. Phys. Lett.97 (21),213702 (2010).
[22]J. J. Niederhauser, M. Jaeger, R. Lemor, P. Weber and M. Frenz, IEEE T. Med. Imaging 24 (4),436-440 (2005).
[23]R. A. Kruger, P. Y. Liu, Y. R. Fang and C. R. Appledorn, Med.Phys.22 (10), 1605-1609 (1995).
[24]M. Xu and L. V. Wang, IEEE Trans. Med. Imaging 21 (7),814-822 (2002).
[25]M. Xu, Y. Xu and L. V. Wang, IEEE T. Biomed. Eng.50 (9),1086-1099 (2003).
[26]M. Xu and L. V. Wang, Phys. Rev. E 71 (1),016706 (2005).
[27]Y. Xu, L. V. Wang, G. Ambartsoumian and P. Kuchment, Med. Phys.31 (4), 724-733 (2004).
[28]Y. Xu and L. V. Wang, IEEE T. Ultrason. Ferr.50 (9),1134-1146 (2003).
[29]X. Jin and L. V. Wang, Phys. Med. Biol.51 (24),6437-6448 (2006).
[30]R. Courant, K. O. Friedrichs and H. Lewy, IBM J.11,215-234 (1956).
[31]X. C. Pan, Y. Zou and M. A. Anastasio, IEEE T. Image. Process.12 (7), 784-795 (2003).
[32]X. L. Dean-Ben, V. Ntziachristos and D. Razansky, Appl. Phys. Lett.98 (17), 171110(2011).
[33]M. Fink, F. Wu, J.-L. Thomas and M. Fink, IEEE T. Ultrason. Ferroelec. Fre. Contr.39 (5),24 (1992).
[34]J. L. Thomas, F. Wu and M. Fink, Ultrasonic Imaging 18 (2),16 (1996).
[35]M. Fink, D. Cassereau, A. Derode, C. Prada, P. Roux, M. Tanter, J. L. Thomas and F. Wu, Rep. Prog. Phys.63 (12),1933-1995 (2000).
[36]P. Blomgren, G. Papanicolaou and H. Zhao, J. Acoust. Soc. Am. 111 (1),19 (2001).
[37]G. Lerosey, J. de Rosny, A. Tourin, A. Derode, G. Montaldo and M. Fink, Phys. Rev. Lett.92 (19),193904 (2004).
[38]A. Sutin, A. Sarvazyan, P. A. Johnson and J. A. TenCate, J. Acoust. Soc. Am. 115 (5),2384-2384 (2004).
[39]Y. Xu and L. V. Wang, Phys. Rev. Lett.92 (3),033902 (2004).
[40]S. Cathelinea, M. Fink, N. Quieffin and R. K. Ing, Appl. Phys. Lett.90 (6),3 (2007).
[41]M. Fink, G. Lerosey, J. De Rosny and A. Tourin, Science 315 (5815), 1120-1122(2007).
[42]G. Bartal, G. Lerosey and X. Zhang, Phys. Rev. B 79 (20),201103 (2009).
[43]O. Couture, J.-F. Aubry, M. Tanter and M. Fink, Appl. Phys. Lett.94 (17), 173901 (2009).
[44]F. Lemoult, G. Lerosey, J. de Rosny and M. Fink, Phys. Rev. Lett 103 (17), 173902(2009).
[45]C. S. Larmat, R. A. Guyer and P. A. Johnson, Phys. Today 63 (8),31-35 (2010).
[46]C. Draeger and M. Fink, Phys. Rev. Lett.79 (3),407-410 (1997).
[47]B. T. Cox, S. R. Arridge and P. C. Beard, Inverse Probl.23 (6), S95-S112 (2007).
[48]D. Cassereau and M. Fink, IEEE T. Ultrason. Ferroelectr. Freq. Control.39 (5), 579-592(1992).
[49]M. Fink and C. Prada, Inverse Probl.17,38 (2001).
[50]C. Tao and X. Liu, Opt. Express 18 (3),2760 (2010).
[51]I. Vasconcelos, R. Snieder and H. Douma, Phys. Rev. E 80 (3),036605 (2009).
[52]A. Aubry and A. Derode, Phys. Rev. Lett 102 (8),084301 (2009).