The utilization of generative artificial intelligence models, such as GAN and diffusion model, is increasingly pervasive in the field of image processing. Generative models effectively learn the distribution of photoacoustic data to optimize and enhance image quality, demonstrating superior performance in photoacoustic tomography (PAT). While most image optimization efforts in PAT occur within the image domain, improving the quality of image reconstruction could be better achieved through direct optimization in the data-domain. However, the majority of publicly available datasets are currently based on the image domain, and the high cost and complexity of PAT systems contribute to a shortage of publicly available datasets in the data-domain. To address this issue, this study has established a data-domain sharing database. The dataset consists of photoacoustic signals captured from various samples at different sparse views using a self-designed and constructed system. This work aims to mitigate the deficiencies in data-domain database for PAT, thereby fostering the development of generative artificial intelligence in data-domain of photoacoustic imaging.
Photoacoustic microscopy is a hybrid imaging technique that capitalizes on the photoacoustic effect to enhance imaging processes, achieving precise image reconstruction by eliminating noise within photoacoustic signals. This study introduces an innovative deep learning denoising algorithm based on score-based diffusion generative models. During the forward propagation process, the model acquires a score representation of the prior noise distribution resulting from the diffusion of the photoacoustic image. In the reverse reconstruction process, the noisy photoacoustic image serves as input. Following multiple iterations by the solver, a noise-free photoacoustic image is generated as the output. A predictor-corrector framework, trained during the forward propagation process, is employed to rectify the reverse evolution. This algorithm effectively reduces noise and demonstrates its efficacy in complex denoising challenges, thereby significantly improving the quality of photoacoustic imaging.
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.