scholarly journals Creating a Simulated Dataset for Training Deep Convolutional Neural Networks for Use in Cardiovascular Photoacoustic Tomography

2020 ◽  
Vol 3 ◽  
Author(s):  
Hayley Chan ◽  
Craig Goergen ◽  
Katherine Leyba
Keyword(s):  
Data Augmentation ◽  
Ex Vivo ◽  
Initial Pressure ◽  
Image Resolution ◽  
Sensor Data ◽  
Photoacoustic Tomography ◽  
Multidimensional Space ◽  
Imaging Method ◽  
Ultrasound Images ◽  
Carotid Ultrasound

Background/Objective: Photoacoustic tomography possesses increasing potential as a non-invasive imaging method that combines optical and acoustic imaging to maximize the visualization of tissue. Determining the composition, orientation, and location of anatomical structures in multidimensional space requires maximizing image resolution and differentiation from noise and reflection artifacts. Using simulations to develop and improve methods for image resolution allows for flexibility and variation of numerous variables.    Methods: Binary masks were created from mouse common carotid ultrasound images using a graphical user interface for MATLAB. With the k-Wave toolbox, we performed time-reversal photoacoustic simulations using the masks. Medium properties for the simulations were assigned for sound speed and density for connective tissue (1540 m/s, 1027 kg/m3) and arterial walls (1569 m/s, 1102 kg/m3). The dataset was augmented through rotational and mirrored transformations and the addition of noise and reflection artifacts via Python open-source software.    Results: A set of 87 binary masks was generated from common carotid ultrasound images. These masks were used to simulate initial pressure distributions through the k-Wave toolbox to reconstruct the structure of the common carotid. Each simulation yielded graphs for initial pressure and sensor distribution, simulated sensor data, reconstructed initial pressure, and a comparison profile between the original and reconstructed pressure. Data augmentation was implemented using the reconstructed pressure output from the 87 simulations, each producing 12 distinct images from rotations and mirroring with the addition of noise and reflection artifacts. The final dataset yielded 1044 images.    Conclusion and Potential Impact: Future work will involve applying this dataset to a neural network to improve photoacoustic quality such that transfer learning can be applied on ex vivo and in vivo datasets. Thus, there is potential for use in diagnostic applications in patients with cardiovascular disease states like atherosclerosis and aneurysms that require high resolution visualization of tissue structure and composition. 

scholarly journals Image Reconstruction with Reliability Assessment in Quantitative Photoacoustic Tomography

2018 ◽  
Vol 4 (12) ◽  
pp. 148 ◽  
Author(s):  
Niko Hänninen ◽  
Aki Pulkkinen ◽  
Tanja Tarvainen
Keyword(s):  
Approximation Error ◽  
Initial Pressure ◽  
Photoacoustic Tomography ◽  
Imaging Method ◽  
Optical Parameters ◽  
Reconstruction Method ◽  
Reconstruction Procedure ◽  
Modeling Errors ◽  
Bayesian Approximation ◽  
Error Method

Quantitative photoacoustic tomography is a novel imaging method which aims to reconstruct optical parameters of an imaged target based on initial pressure distribution, which can be obtained from ultrasound measurements. In this paper, a method for reconstructing the optical parameters in a Bayesian framework is presented. In addition, evaluating the credibility of the estimates is studied. Furthermore, a Bayesian approximation error method is utilized to compensate the modeling errors caused by coarse discretization of the forward model. The reconstruction method and the reliability of the credibility estimates are investigated with two-dimensional numerical simulations. The results suggest that the Bayesian approach can be used to obtain accurate estimates of the optical parameters and the credibility estimates of these parameters. Furthermore, the Bayesian approximation error method can be used to compensate for the modeling errors caused by a coarse discretization, which can be used to reduce the computational costs of the reconstruction procedure. In addition, taking the modeling errors into account can increase the reliability of the credibility estimates.

Photoacoustic tomography of ex vivo mouse hearts with myocardial infarction

2011 ◽  
Vol 16 (3) ◽  
pp. 036007 ◽  
Author(s):  
Markus Holotta ◽  
Harald Grossauer ◽  
Christian Kremser ◽  
Pavle Torbica ◽  
Jakob Völkl ◽  
...  
Keyword(s):  
Myocardial Infarction ◽  
Ex Vivo ◽  
Photoacoustic Tomography

scholarly journals Tracking Transplanted Stem Cells Using Magnetic Resonance Imaging and the Nanoparticle Labeling Method in Urology

2015 ◽  
Vol 2015 ◽  
pp. 1-9 ◽  
Author(s):  
Jae Heon Kim ◽  
Hong J. Lee ◽  
Yun Seob Song
Keyword(s):  
Magnetic Resonance Imaging ◽  
Stem Cells ◽  
Molecular Imaging ◽  
Magnetic Resonance ◽  
Imaging Modality ◽  
Imaging Techniques ◽  
Image Resolution ◽  
Imaging Method ◽  
Resonance Imaging

A reliablein vivoimaging method to localize transplanted cells and monitor their viability would enable a systematic investigation of cell therapy. Most stem cell transplantation studies have used immunohistological staining, which does not provide information about the migration of transplanted cellsin vivoin the same host. Molecular imaging visualizes targeted cells in a living host, which enables determining the biological processes occurring in transplanted stem cells. Molecular imaging with labeled nanoparticles provides the opportunity to monitor transplanted cells noninvasively without sacrifice and to repeatedly evaluate them. Among several molecular imaging techniques, magnetic resonance imaging (MRI) provides high resolution and sensitivity of transplanted cells. MRI is a powerful noninvasive imaging modality with excellent image resolution for studying cellular dynamics. Several types of nanoparticles including superparamagnetic iron oxide nanoparticles and magnetic nanoparticles have been used to magnetically label stem cells and monitor viability by MRI in the urologic field. This review focuses on the current role and limitations of MRI with labeled nanoparticles for tracking transplanted stem cells in urology.

scholarly journals Should clinicians be given hard copies of duplex carotid ultrasound images?

1991 ◽  
Vol 54 (10) ◽  
pp. 936-937
Author(s):  
S Gunatilake ◽  
P Sandercock ◽  
J Slattery
Keyword(s):  
Ultrasound Images ◽  
Carotid Ultrasound

Early-stage atherosclerosis detection using deep learning over carotid ultrasound images

2016 ◽  
Vol 49 ◽  
pp. 616-628 ◽  
Author(s):  
Rosa-María Menchón-Lara ◽  
José-Luis Sancho-Gómez ◽  
Andrés Bueno-Crespo
Keyword(s):  
Deep Learning ◽  
Early Stage ◽  
Ultrasound Images ◽  
Carotid Ultrasound

scholarly journals Synthetic aperture-based linear-array photoacoustic tomography considering the aperture orientation effect

2018 ◽  
Vol 11 (04) ◽  
pp. 1850015 ◽  
Author(s):  
Xiangwei Lin ◽  
Jaesok Yu ◽  
Naizhang Feng ◽  
Mingjian Sun
Keyword(s):  
Linear Array ◽  
Great Influence ◽  
Image Resolution ◽  
Synthetic Aperture ◽  
Orientation Effect ◽  
Photoacoustic Tomography ◽  
Optimal Detection ◽  
Phantom Experiment ◽  
Full View ◽  
Array Transducer

The synthetic aperture-based linear-array photoacoustic tomography (PAT) was proposed to address the limited-view shortcomings of the single aperture, but the detection field of view (FOV) determined by the aperture orientation effect was not fully considered yet, leading to the limited-view observation and image resolution degradation. Herein, the aperture orientation effect was proposed from the theoretical model and then it was verified via both the numerical simulation and phantom experiment. Different orientations were enumerated sequentially in the simulation to approximate the ideal full-view case for the optimal detection FOV, considering the detection pattern of the linear-array transducer. As a result, the corresponding optimal aperture orientation was 60[Formula: see text] if the synthetic aperture was seamlessly established by three single linear arrays, where the overlapped detection pattern was optimized from the individual linear-array transducer at the adjacent positions. Therefore, the limited-view artifacts were minimized and the image resolution was enhanced in this aperture orientation. This study showed that the aperture orientation had great influence on the optimal detection FOV in the synthetic aperture configuration, where the full-view imaging quality and enhanced image resolution could be achieved.

Sign in / Sign up

Export Citation Format

Share Document