Generally applicable and experimentally validated hybrid models and inversion schemes for quantitative photoacoustic tomography

普遍适用且经过实验验证的定量光声层析成像混合模型和反演方案

• 批准号: 451574916
• 负责人: Professor Jan Laufer, Ph.D.
• 金额: --
• 依托单位: Institut für Physik
• 依托单位国家: 德国
• 项目类别: Research Grants
• 资助国家: 德国
• 项目状态: 未结题
• 来源: https://gepris.dfg.de/gepris/projekt/451574916?language=en
• 关键词: Generally; applicable; experimentally; validated; hybrid

Biomedical photoacoustic (PA) tomography is based on the generation of ultrasound in the tissue via the absorption of intensity-modulated light. Due to the dominant absorption of haemoglobin in the visible and near-infrared wavelength range, it provides high-resolution 3D images of the vasculature in soft tissue. It is also sensitive to exogenous and genetically expressed contrast agents. A major potential of this technology lies in the ability to recover 3D high resolution images of absolute chromophore concentrations (e.g. oxy- and deoxyhaemoglobin, contrast agents) as well as derived functional parameters, such as blood oxygen saturation. The development of methods for quantitative PA tomography is of crucial importance for the translation of this technology to routine applications in functional and molecular imaging. In this project, comprehensive forward models will be developed to predict multispectral PA images that closely agree with those acquired using different tomographic PA scanner platforms. The models will be used in adjoint form in new inversion algorithms, which will be experimentally validated on images of tissue phantoms acquired using novel experimental methods. The limitations of the new methods will be evaluated in silico. It is envisaged that the models developed in this project will enable true quantitative PA tomography using both model-based and data-driven inversions. The successful development of these models and methods will provide a basis for rapid translation of functional and molecular PA imaging to applications in biology, the life sciences and medicine.

生物医学光声（PA）断层成像是基于通过吸收强度调制光在组织中产生超声。由于血红蛋白在可见光和近红外波长范围内的主要吸收，它提供了软组织中脉管系统的高分辨率3D图像。它也对外源性和基因表达的造影剂敏感。该技术的主要潜力在于能够恢复绝对发色团浓度（例如氧合血红蛋白和脱氧血红蛋白、造影剂）的3D高分辨率图像以及导出的功能参数（例如血氧饱和度）。定量PA断层扫描的方法的发展是至关重要的翻译这项技术的常规应用程序中的功能和分子成像。在这个项目中，将开发全面的正演模型来预测多光谱PA图像，这些图像与使用不同的层析PA扫描仪平台获得的图像非常一致。该模型将被用于伴随的形式，在新的反演算法，这将是实验验证的图像上获得的组织幻影使用新的实验方法。新方法的局限性将在计算机上进行评估。据设想，在这个项目中开发的模型将使真正的定量PA断层扫描使用基于模型和数据驱动的反演。这些模型和方法的成功开发将为功能和分子PA成像在生物学、生命科学和医学中的应用提供快速翻译的基础。