浏览全部资源
扫码关注微信
纸质出版日期： 2024-03-16 ，
移动端阅览
候英飒， 孙正， 孙美晨. 2024. JIR-Net：用于光声层析图像重建的联合迭代重建网络. 中国图象图形学报， 29(03):0823-0838
Hou Yingsa， Sun Zheng， Sun Meichen. 2024. JIR-Net： a joint iterative reconstruction network forphotoacoustic tomography image reconstruction. Journal of Image and Graphics， 29(03):0823-0838
目的
2
高质量的图像重建是光声层析成像（photoacoustic tomography，PAT）技术的关键，有限角度稀疏测量和组织非均匀的声学特性都会影响重建图像质量。采用迭代重建技术可在一定程度上提高图像质量，但是其结果依赖于有关成像目标的先验假设模型。而且在迭代优化过程中需要反复计算前向成像算子及其伴随算子，因此计算成本较高，需要合理选择正则化方法及其参数。为了解决该问题，提出一种根据不完备光声测量信号联合重建光吸收能量分布图和声速分布图的深度学习方法。
方法
2
设计并搭建基于学习迭代策略的联合迭代重建网络（joint iterative reconstruction network，JIR-Net），网络由4个结构单元组成，每个单元包括特征提取、特征融合和重建3个模块。网络的输入是探测器在成像平面中采集的不完备光声信号和预设的常数声速，输出是重建的光吸收能量分布图和声速分布图。分别构建仿真、仿体和在体数据集，用于训练、验证和测试网络。在训练网络的过程中，将光吸收能量密度和声速的梯度下降信息整合到网络训练中，并利用反向传播梯度下降法求解非线性最小二乘问题。
结果
2
数值仿真、仿体和在体实验结果表明：与交替优化法、U-Net后处理法和深度梯度下降法相比，JIR-Net重建的光吸收能量分布图的结构相似度可分别提高约39.5%、26.4%和7.6%，峰值信噪比可分别提高约95.6%、71.4%和15.5%。与交替优化法相比，JIR-Net重建的声速分布图的结构相似度和峰值信噪比可分别提高约34.4%和22.6%。
结论
2
JIR-Net解决了由于有限角度稀疏测量和组织声速分布不均匀所致的光声图像质量下降问题，实现了从光声信号到高质量光吸收能量分布图和声速分布图的映射。
Objective
2
Photoacoustic tomography （PAT） is a hybrid functional imaging modality developed rapidly in recent years. PAT is physically based on the photoacoustic effect， where biological tissues are irradiated by short laser pulses， inducing broadband （～MHz） ultrasonic waves （i.e.， photoacoustic waves） due to optical absorption and thermoelastic expansion. Ultrasonic transducers deployed around the imaging target collect photoacoustic waves from which images are reconstructed to show the morphological structure and functional properties of tissues. High-quality image reconstruction is essential for PAT， which suffers from incomplete measurements and heterogeneous acoustic properties of tissues. Traditional image reconstruction methods include back projection， time reversal， Fourier transform-based reconstruction， and delay and sum. For simplicity， these methods are usually based on ideal assumptions about the imaging scenario， such as fixed speed of sound， a lossless acoustic media without attenuation， a point-like ultrasonic detector with sufficient bandwidth， and complete measurement. However， in real-world applications， these ideal scenarios often do not occur， leading to the degradation of the quality of images reconstructed using these methods. The model-based iterative reconstruction scheme is commonly used to improve image quality， where the inversion of a forward imaging model describing the generation of photoacoustic signal is iteratively solved. However， its real-time applications are limited by its high computational cost because the forward imaging operator and its adjoint operator need to be calculated repeatedly in the iterative process. Regularization tools with properly defined parameters are necessary to obtain stable optimization. In addition， the reconstruction quality highly depends on the prior assumptions of the imaging object. In recent years， deep learning has shown great potential in reconstructing high-quality images from photoacoustic measurements. This work aims to solve the problem of image quality degradation caused by incomplete measurement and heterogeneously distributed speed of sound.
Method
2
A deep learning method is proposed to reconstruct jointly images representing the distributions of optical absorption and speed of sound within the imaging domain from incomplete photoacoustic measurements. A convolutional neural network， named joint iterative reconstruction network （JIR-Net）， is constructed based on an iterative learning strategy. Incomplete photoacoustic measurements are fed into the network， and images representing absorbed optical energy density and speed of sound distributions are output. The network consists of four structural units， and each unit is composed of three modules： feature extraction， feature fusion， and reconstruction. The feature extraction module extracts features from four inputs via convolution. The feature fusion module combines the features extracted from the input. Finally， the reconstruction module recreates the distributions of absorbed optical energy density and speed of sound. The network is trained using simulation， phantom， and
in vivo
datasets， where the gradient descent information of the absorbed optical energy density and the speed of sound distributions is incorporated into the network training. The nonlinear least square problem is solved by using the back propagation gradient descent. The validity of the method is demonstrated by simulation， phantom， and
in vivo
studies. Compared with traditional nonlearning methods， non-iterative learning method， and learning iterative method based on depth gradient descent， JIR-Net is superior in reconstructing high-quality images from sparse data measured in acoustically heterogeneous media.
Result
2
Numerical simulation， phantom， and
in vivo
experiment results show the trained JIR-Net is robust to data sparsity and insensitive to the initial iterative plan. Moreover， the superiority of JIR-Net in complex structure reconstruction is proven
in vivo
. Compared with the depth gradient descent method， the U-Net post-processing method， and the alternate optimization method， the structural similarity of the reconstructed images representing absorbed optical energy density distribution can be improved by 7.6%， 26.4%， and 39.5%， respectively， and the peak signal-to-noise ratio can be improved by 15.5%， 71.4%， and 95.6%， respectively. Compared with the alternate optimization method， the structural similarity and peak signal-to-noise ratio of the JIR-Net reconstructed speed of sound images are increased by 34.4% and 22.6%， respectively.
Conclusion
2
JIR-Net achieves the mapping from incomplete photoacoustic measurements to high-quality images representing the distributions of absorbed optical energy density and speed of sound. The method can be used for image reconstruction from limited-view sparse photoacoustic signals collected with any measuring geometry in acoustic media with inhomogeneously distributed speed of sound. The method eliminates the need for prior knowledge of the characteristics of the imaging target and reduces the need for scanning and detection equipment， enabling constructing more compact imaging systems.
图像重建技术光声层析成像（PAT）深度学习光吸收能量密度声速（SoS）联合重建梯度下降
image reconstruction techniquesphotoacoustic tomography（PAT）deep learningabsorbed optical energy densityspeed of sound（SoS）joint reconstructiongradient descent
Adler J and Öktem O. 2017. Solving ill-posed inverse problems using iterative deep neural networks. Inverse Problems， 33（12）： #124007 ［DOI： 10.1088/1361-6420/aa9581http://dx.doi.org/10.1088/1361-6420/aa9581］
Antholzer S， Schwab J， Bauer-Marschallinger J， Burgholzer P and Haltmeier M. 2019. NETT regularization for compressed sensing photoacoustic tomography//Proceedings Volume 10878， Photons Plus Ultrasound： Imaging and Sensing 2019. San Francisco， USA： SPIE： 272-282 ［DOI： 10.1117/12.2508486http://dx.doi.org/10.1117/12.2508486］
Awasthi N， Jain G， Kalva S K， Pramanik M and Yalavarthy P K. 2020. Deep neural network-based sinogram super-resolution and bandwidth enhancement for limited-data photoacoustic tomography. IEEE Transactions on Ultrasonics， Ferroelectrics， and Frequency Control， 67（12）： 2660-2673 ［DOI： 10.1109/TUFFC.2020.2977210http://dx.doi.org/10.1109/TUFFC.2020.2977210］
Boink Y E， Manohar S and Brune C. 2020. A partially-learned algorithm for joint photo-acoustic reconstruction and segmentation. IEEE Transactions on Medical Imaging， 39（1）： 129-139 ［DOI： 10.1109/TMI.2019.2922026http://dx.doi.org/10.1109/TMI.2019.2922026］
Farnia P， Mohammadi M， Najafzadeh E， Alimohamadi M， Makkiabadi B and Ahmadian A. 2020. High-quality photoacoustic image reconstruction based on deep convolutional neural network： towards intra-operative photoacoustic imaging. Biomedical Physics and Engineering Express， 6（4）： #045019 ［DOI： 10.1088/2057-1976/ab9a10http://dx.doi.org/10.1088/2057-1976/ab9a10］
Guan S， Khan A A， Sikdar S and Chitnis P V. 2020. Limited-view and sparse photoacoustic tomography for neuroimaging with deep learning. Scientific Reports， 10（1）： #8510 ［DOI： 10.1038/s41598-020-65235-2http://dx.doi.org/10.1038/s41598-020-65235-2］
Guo M J， Lan H R， Yang C C and Gao F. 2021. AS-Net： fast photoacoustic reconstruction with multi-feature fusion from sparse data ［EB/OL］. ［2022-12-16］. https://arxiv.org/pdf/2101.08934.pdfhttps://arxiv.org/pdf/2101.08934.pdf
Haltmeier M and Nguyen L V. 2017. Analysis of iterative methods in photoacoustic tomography with variable sound speed. SIAM Journal on Imaging Sciences， 10（2）： 751-781 ［DOI： 10.1137/16M1104822http://dx.doi.org/10.1137/16M1104822］
Huang C， Wang K， Nie L M， Wang L V and Anastasio M A. 2013. Full-wave iterative image reconstruction in photoacoustic tomography with acoustically inhomogeneous media. IEEE Transactions on Medical Imaging， 32（6）： 1097-1110 ［DOI： 10.1109/TMI.2013.2254496http://dx.doi.org/10.1109/TMI.2013.2254496］
Huang C， Wang K， Schoonover R W， Wang L V and Anastasio M A. 2016. Joint reconstruction of absorbed optical energy density and sound speed distributions in photoacoustic computed tomography： a numerical investigation. IEEE Transactions on Computational Imaging， 2（2）： 136-149 ［DOI： 10.1109/TCI.2016.2523427http://dx.doi.org/10.1109/TCI.2016.2523427］
Kim M W， Jeng G S， Pelivanov I and O'Donnell M. 2020. Deep-learning image reconstruction for real-time photoacoustic system. IEEE Transactions on Medical Imaging， 39（11）： 3379-3390 ［DOI： 10.1109/TMI.2020.2993835http://dx.doi.org/10.1109/TMI.2020.2993835］
Kingma D P and Ba J L. 2017. Adam： a method for stochastic optimization ［EB/OL］. ［2022-12-16］. https://arxiv.org/pdf/1412.6980.pdfhttps://arxiv.org/pdf/1412.6980.pdf
Lan H R， Jiang D H， Yang C C， Gao F and Gao F. 2020. Y-Net： hybrid deep learning image reconstruction for photoacoustic tomography in vivo. Photoacoustics， 20： #100197 ［DOI： 10.1016/j.pacs.2020.100197http://dx.doi.org/10.1016/j.pacs.2020.100197］
Lan H R， Yang C C， Gao F and Gao F. 2021. Better than ground-truth？ Beyond supervised learning for photoacoustic imaging reconstruction ［EB/OL］. ［2022-12-16］. https://arxiv.org/pdf/2012.02472v2.pdfhttps://arxiv.org/pdf/2012.02472v2.pdf
Lan H R， Yang C C， Jiang D H and Gao F. 2019a. Reconstruct the photoacoustic image based on deep learning with multi-frequency ring-shape transducer array//Proceedings of the 41st Annual International Conference of the IEEE Engineering in Medicine and Biology Society （EMBC）. Berlin， Germany： IEEE： 7115-7118 ［DOI： 10.1109/EMBC.2019.8856590http://dx.doi.org/10.1109/EMBC.2019.8856590］
Lan H R， Zhou K， Yang C C， Cheng J， Liu J， Gao S H and Gao F. 2019b. Ki-GAN： knowledge infusion generative adversarial network for photoacoustic image reconstruction in vivo//Proceedings of 2019 International Conference on Medical Image Computing and Computer-Assisted Intervention. Shenzhen， China： Springer： 273-281 ［DOI： 10.1007/978-3-030-32239-7_31http://dx.doi.org/10.1007/978-3-030-32239-7_31］
Li H S， Schwab J， Antholzer S and Haltmeier M. 2020. NETT： solving inverse problems with deep neural networks. Inverse Problems， 36（6）： #065005 ［DOI： 10.1088/1361-6420/ab6d57http://dx.doi.org/10.1088/1361-6420/ab6d57］
Li S L， Feng C L， Yu K， Liu X， Jiang X and Zhao D Z. 2022. Critical review of human cardiac magnetic resonance image super resolution reconstruction based on deep learning method. Journal of Image and Graphics， 27（3）： 704-721
李书林， 冯朝路， 于鲲， 刘鑫， 江鑫， 赵大哲. 2022. 基于深度学习的心脏磁共振影像超分辨率前沿进展. 中国图象图形学报， 27（3）： 704-721 ［DOI： 10.11834/jig.210150http://dx.doi.org/10.11834/jig.210150］
Lutzweiler C and Razansky D. 2013. Optoacoustic imaging and tomography： reconstruction approaches and outstanding challenges in image performance and quantification. Sensors， 13（6）： 7345-7384 ［DOI： 10.3390/s130607345http://dx.doi.org/10.3390/s130607345］
Matthews T P， Poudel J， Li L， Wang L V and Anastasio M A. 2018. Parameterized joint reconstruction of the initial pressure and sound speed distributions for photoacoustic computed tomography. Siam Journal on Imaging Sciences， 11（2）： 1560-1588 ［DOI： 10.1137/17M1153649http://dx.doi.org/10.1137/17M1153649］
Park E Y， Lee H， Han S， Kim C and Kim J. 2022. Photoacoustic imaging systems based on clinical ultrasound platform. Experimental Biology and Medicine， 247（7）： 551-560 ［DOI： 10.1177/15353702211073684http://dx.doi.org/10.1177/15353702211073684］
Rosenthal A， Ntziachristos V and Razansky D. 2013. Acoustic inversion in optoacoustic tomography： a review. Current Medical Imaging， 9（4）： 318-336 ［DOI： 10.2174/15734056113096660006http://dx.doi.org/10.2174/15734056113096660006］
Shan H M， Wiedeman C， Wang G and Yang Y. 2019. Simultaneous reconstruction of the initial pressure and sound speed in photoacoustic tomography using a deep-learning approach//Proceedings Volume #11105， Novel Optical Systems， Methods， and Applications XXII. San Diego， USA： SPIE： 18-27 ［DOI： 10.1117/12.2529984http://dx.doi.org/10.1117/12.2529984］
Sun Z， Han D D and Yuan Y. 2016. 2-D image reconstruction of photoacoustic endoscopic imaging based on time-reversal. Computers in Biology and Medicine， 76： 60-68 ［DOI： 10.1016/j.compbiomed.2016.06.028http://dx.doi.org/10.1016/j.compbiomed.2016.06.028］
Sun Z and Sun L S. 2021. Simultaneous reconstruction of optical absorption property and speed of sound in intravascular photoacoustic tomography. Inverse Problems in Science and Engineering， 29（12）： 1764-1788 ［DOI： 10.1080/17415977.2021.1879805http://dx.doi.org/10.1080/17415977.2021.1879805］
Sun Z， Wang X Y and Yan X Y. 2021. An iterative gradient convolutional neural network and its application in endoscopic photoacoustic image formation from incomplete acoustic measurement. Neural Computing and Applications， 33（14）： 8555-8574 ［DOI： 10.1007/s00521-020-05607-xhttp://dx.doi.org/10.1007/s00521-020-05607-x］
Sun Z and Yan X Y. 2020. A deep learning method for limited-view intravascular photoacoustic image reconstruction. Journal of Medical Imaging and Health Informatics， 10（11）： 2707-2713 ［DOI： 10.1166/jmihi.2020.3204http://dx.doi.org/10.1166/jmihi.2020.3204］
Tang C S， Hu C C， Sun J D and Sima H F. 2021. Deep learning-based medical images analysis evolved from convolution to graph convolution. Journal of Image and Graphics， 26（9）： 2078-2093
唐朝生， 胡超超， 孙君顶， 司马海峰. 2021. 医学图像深度学习技术： 从卷积到图卷积的发展. 中国图象图形学报， 26（9）： 2078-2093 ［DOI： 10.11834/jig.200666http://dx.doi.org/10.11834/jig.200666］
Treeby B E and Cox B T. 2010. K-wave： MATLAB toolbox for the simulation and reconstruction of photoacoustic wave fields. Journal of Biomedical Optics， 15（2）： #021314 ［DOI： 10.1117/1.3360308http://dx.doi.org/10.1117/1.3360308］
Vu T， Li M C， Humayun H， Zhou Y and Yao J J. 2020. A generative adversarial network for artifact removal in photoacoustic computed tomography with a linear-array transducer. Experimental Biology and Medicine， 245（7）： 597-605 ［DOI： 10.1177/1535370220914285http://dx.doi.org/10.1177/1535370220914285］
Wang L H， Jacques S L and Zheng L Q. 1995. MCML——Monte Carlo modeling of light transport in multi-layered tissues. Computer Methods and Programs in Biomedicine， 47（2）： 131-146 ［DOI： 10.1016/0169-2607（95）01640-fhttp://dx.doi.org/10.1016/0169-2607（95）01640-f］
Xu M H and Wang L V. 2005. Universal back-projection algorithm for photoacoustic computed tomography//Proceedings Volume 5697， Photons Plus Ultrasound： Imaging and Sensing 2005. San Jose， USA： SPIE： 251-254 ［DOI： 10.1117/12.589146http://dx.doi.org/10.1117/12.589146］
Zhang J， Wang K， Yang Y Y and Anastasio M A. 2008. Simultaneous reconstruction of speed-of-sound and optical absorption properties in photoacoustic tomography via a time-domain iterative algorithm//Proceedings Volume #6856， Photons Plus Ultrasound： Imaging and Sensing 2008. San Jose， USA： SPIE： 427-434 ［DOI： 10.1117/12.764171http://dx.doi.org/10.1117/12.764171］
相关作者
相关机构
京公网安备11010802024621