Guidance-base Diffusion Models for Improving Photoacoustic Image Quality

2025-02-10

AI Summary

Poor image quality in single-shot photoacoustic (PA) imaging—typically mitigated via multi-frame averaging—incurs high computational cost and low acquisition efficiency. To address this, we propose a conditional diffusion model that jointly incorporates sensor-specific physical priors and imaging-condition guidance. Specifically, we embed the PA sensor’s impulse response, laser pulse energy, and trigger delay into the diffusion process as conditioning signals, and enforce physics-informed constraints during reverse sampling to synergistically integrate data-driven learning with physical modeling. Our method significantly enhances the signal-to-noise ratio and microvascular structural fidelity of single-frame reconstructions, achieving performance comparable to conventional multi-frame averaging while reducing imaging time by several-fold and substantially lowering system load. This work establishes a new paradigm for high-speed, low-cost, high-fidelity single-shot PA imaging.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Search and Optimization: Learning to Search
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding

Application Category

• Search and Retrieval-Augmented AI: Personalized, context-aware and across-device search
• Security and Privacy: Privacy-enhancing technologies
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation

Abstract

Photoacoustic(PA) imaging is a non-destructive and non-invasive technology for visualizing minute blood vessel structures in the body using ultrasonic sensors. In PA imaging, the image quality of a single-shot image is poor, and it is necessary to improve the image quality by averaging many single-shot images. Therefore, imaging the entire subject requires high imaging costs. In our study, we propose a method to improve the quality of PA images using diffusion models. In our method, we improve the reverse diffusion process using sensor information of PA imaging and introduce a guidance method using imaging condition information to generate high-quality images.

Problem

• Improving photoacoustic image quality
• Reducing high imaging costs
• Enhancing reverse diffusion process with sensor information

Innovation

• diffusion models enhance image quality
• sensor information optimizes reverse diffusion
• guidance method uses imaging conditions