Traditional deconvolution methods can improve the spatial resolution of photoacoustic computed tomography (PACT) systems but are often sensitive to noise. We propose a novel approach to enhance the resolution of PACT, by modeling the system’s point spread function (PSF) and performing deep-learning-based deconvolution. We train a robust deep learning model without the need for ground truth, using a self-supervised method on a mixed dataset of simulation, phantom, and in vivo data, in combination with various data augmentation techniques. We demonstrate that our deep learning deconvolution achieves superior spatial resolution, image contrast, and artifact suppression, when compared to traditional deconvolution methods.
The most clinically compatible PAT configuration usually employs a linear ultrasound array, which often has a limited detection view and poor image fidelity. Exogenous contrast agents such as nanoparticles can be employed but lacks clinical translation potential. We have developed a new methodology by using clinically-approved microbubble as virtual point sources that strongly scatter the local pressure waves from surround hemoglobin, preserving PAT’s functional capability and clinical translation potential. We can overcome the limited-detection-view problem and achieve high-fidelity functional PAT in deep tissue. We have investigated the working principle and demonstrated proof-of-concept applications using simulations, phantoms, and in vivo small-animal studies.
KEYWORDS: Photoacoustic tomography, Brain, Tissue optics, Monte Carlo methods, 3D modeling, Brain imaging, Blood vessels, Absorption, Spatial resolution, Signal attenuation
Photoacoustic computed tomography (PACT) has great potential in mouse brain imaging. Conventional PACT either assumes homogenous optical fluence or uses simplified attenuation model for optical fluence estimation, resulting in inaccurate estimation of absorption coefficient of the chromophore. To optimize the quantitative performance of PACT, we used MCX 3D Monte Carlo simulation to study the optical fluence distribution in a complete mouse brain model, which contains complete anatomy and blood vasculature information. Our results suggest that optical fluence decays five times globally due to strong scattering tissue and fluctuates locally due to additional optical heterogeneity introduced by blood vessels.
Photoacoustic imaging (PAI) is a promising imaging technique in preclinical study, which combines both merits of optical and ultrasound imaging. However, PA image quality is seriously suffered from various noises such as random white noise, intrinsic noise of devices and background noise from imaging environment, especially in in-vivo experiments. Regular linear filters like mean filters no longer provide a satisfactory performance. A boosted filter is necessary to degrade the noise level and enhance PA image contrast. In this paper, we applied a nonlinear de-noising filter based on mathematical morphology, which can help smooth the target boundary and well suppress impulse noise in PA images. Phantom and in-vivo experiments in this paper will show the feasibility and performance as a newly-used filter in PA image processing.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.