Ultrafast filtered back-projection for photoacoustic computed tomography DOI Creative Commons
Songde Liu, Zhijian Tan, Pengfei Shao

et al.

Biomedical Optics Express, Journal Year: 2024, Volume and Issue: 16(2), P. 362 - 362

Published: Nov. 11, 2024

The filtered back-projection (FBP) algorithm is widely used in photoacoustic computed tomography (PACT) for image reconstruction due to its simplicity and efficiency. Yet, the real-time processing of high-speed PACT data remains challenging regular FBP implementations. To enhance efficiency algorithm, researchers have developed implementations based on graphics units (GPUs). However, existing GPU-accelerated algorithms either sacrifice accuracy or are still inefficient high-speed, imaging. Herein, we report an ultrafast GPU acceleration-based implementation without sacrificing accuracy. Firstly, computation complexity filtering part significantly simplified with a pre-computed matrix Secondly, dramatically increased through parallel programming acceleration. As result, proposed takes only 0.38 ms reconstruct two-dimensional 512 × pixels, which 439 times faster than Numerical experimental results show that outperforms GPU-based best our knowledge, this fastest ever reported PACT. This work expected provide accurate solution

Language: Английский

Image Restoration for Ring-Array Photoacoustic Tomography Based on an Attention Mechanism Driven Conditional Generative Adversarial Network DOI Creative Commons
Wende Dong, Yanli Zhang,

Luqi Hu

et al.

Photoacoustics, Journal Year: 2025, Volume and Issue: unknown, P. 100714 - 100714

Published: April 1, 2025

Ring-Array photoacoustic tomography (PAT) systems have shown great promise in non-invasive biomedical imaging. However, images produced by these often suffer from quality degradation due to non-ideal imaging conditions, with common issues including blurring and streak artifacts. To address challenges, we propose an image restoration method based on a conditional generative adversarial network (CGAN) framework. Our approach integrates hybrid spatial channel attention mechanism within Residual Shifted Window Transformer Module (RSTM) enhance the generator's performance. Additionally, developed comprehensive loss function balance pixel-level accuracy, detail preservation, perceptual quality. We further incorporate gamma correction module contrast of network's output. Experimental results both simulated vivo data demonstrate that our significantly improves resolution restores overall

Language: Английский

Citations

0
Zero-Shot Artifact2Artifact: Self-incentive artifact removal for photoacoustic imaging DOI Creative Commons

Shuang Li,

Qian Chen, Chulhong Kim

et al.

Photoacoustics, Journal Year: 2025, Volume and Issue: 43, P. 100723 - 100723

Published: April 18, 2025

Three-dimensional (3D) photoacoustic imaging (PAI) with detector arrays has shown superior capabilities in biomedical applications. However, the quality of 3D PAI is often degraded due to reconstruction artifacts caused by sparse detectors. Existing iterative or deep learning-based methods are either time-consuming require large training datasets, limiting their practical application. Here, we propose Zero-Shot Artifact2Artifact (ZS-A2A), a zero-shot self-supervised artifact removal method based on super-lightweight network, which leverages fact that patterns more sensitive sensor data loss. By randomly dropping acquired PA data, it spontaneously generates subset reconstruct images, turn stimulates network learn results, thus enabling removal. This approach requires neither nor prior knowledge artifacts, making suitable for arbitrary array configurations. We validated ZS-A2A both simulation study and invivo animal experiments. Results demonstrate achieves high performance compared existing methods.

Language: Английский

Citations

0
Adaptively spatial PSF removal enables contrast enhancement for multi-layer image fusion in photoacoustic microscopy DOI
Ting Feng, H. Harold Li, Haigang Ma

et al.

Optics Letters, Journal Year: 2024, Volume and Issue: 49(24), P. 7146 - 7146

Published: Nov. 6, 2024

Optical-resolution photoacoustic microscopy enables cellular-level biological imaging in deep tissues. However, acquiring high-quality spatial images without knowing the point spread function (PSF) at multiple depths or physically improving system performance is challenging. We propose an adaptive multi-layer image fusion (AMPIF) approach based on blind deconvolution and registration. Our findings indicate that AMPIF method rapidly achieves optimized focused fused with superior resolution contrast relying prior knowledge of PSF. This holds significant potential for fast living tissues enhanced depths.

Language: Английский

Citations

0
Ultrafast filtered back-projection for photoacoustic computed tomography DOI Creative Commons
Songde Liu, Zhijian Tan, Pengfei Shao

et al.

Biomedical Optics Express, Journal Year: 2024, Volume and Issue: 16(2), P. 362 - 362

Published: Nov. 11, 2024

The filtered back-projection (FBP) algorithm is widely used in photoacoustic computed tomography (PACT) for image reconstruction due to its simplicity and efficiency. Yet, the real-time processing of high-speed PACT data remains challenging regular FBP implementations. To enhance efficiency algorithm, researchers have developed implementations based on graphics units (GPUs). However, existing GPU-accelerated algorithms either sacrifice accuracy or are still inefficient high-speed, imaging. Herein, we report an ultrafast GPU acceleration-based implementation without sacrificing accuracy. Firstly, computation complexity filtering part significantly simplified with a pre-computed matrix Secondly, dramatically increased through parallel programming acceleration. As result, proposed takes only 0.38 ms reconstruct two-dimensional 512 × pixels, which 439 times faster than Numerical experimental results show that outperforms GPU-based best our knowledge, this fastest ever reported PACT. This work expected provide accurate solution

Language: Английский

Citations

0