Application of a Virtual Imaging Framework for Investigating a Deep Learning-Based Reconstruction Method for 3D Quantitative Photoacoustic Computed Tomography

2025-10-03

AI Summary

Quantitative photoacoustic computed tomography (qPACT) reconstruction faces challenges including high computational cost, physical model inaccuracies, and experimental uncertainties; existing learning-based methods lack systematic validation. This work introduces the first high-fidelity virtual imaging framework for qPACT based on stochastically generated digital phantoms, integrating 3D photoacoustic physics modeling, Monte Carlo optical transport simulation, and acoustic distortion modeling to establish an end-to-end differentiable deep learning reconstruction pipeline for quantitative sO₂ mapping in breast tissue. The method demonstrates robustness across diverse anatomical configurations, physiological variations, and noise conditions, achieving significantly reduced sO₂ estimation error compared to conventional approaches. Key contributions are: (1) the first reproducible, photorealistic benchmark platform for validating learning-based qPACT methods; and (2) seamless integration of physics-driven and data-driven paradigms. This framework provides a verifiable paradigm and performance benchmark for preclinical translation of qPACT.

Technology Category

• Computer Vision: Learning & Optimization for CV
• Search and Optimization: Learning to Search
• Machine Learning: Calibration & Uncertainty Quantification

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Systems and Infrastructure for Web, Mobile and WoT: Virtualization and resource management in Web systems and infrastructures
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets

Abstract

Quantitative photoacoustic computed tomography (qPACT) is a promising imaging modality for estimating physiological parameters such as blood oxygen saturation. However, developing robust qPACT reconstruction methods remains challenging due to computational demands, modeling difficulties, and experimental uncertainties. Learning-based methods have been proposed to address these issues but remain largely unvalidated. Virtual imaging (VI) studies are essential for validating such methods early in development, before proceeding to less-controlled phantom or in vivo studies. Effective VI studies must employ ensembles of stochastically generated numerical phantoms that accurately reflect relevant anatomy and physiology. Yet, most prior VI studies for qPACT relied on overly simplified phantoms. In this work, a realistic VI testbed is employed for the first time to assess a representative 3D learning-based qPACT reconstruction method for breast imaging. The method is evaluated across subject variability and physical factors such as measurement noise and acoustic aberrations, offering insights into its strengths and limitations.

Problem

• Validating deep learning reconstruction for 3D quantitative photoacoustic tomography
• Addressing computational demands and modeling difficulties in qPACT imaging
• Evaluating reconstruction robustness across patient variability and physical factors

Innovation

• Realistic virtual imaging testbed for 3D reconstruction
• Evaluated deep learning method across subject variability
• Assessed performance under noise and acoustic aberrations