基于光强传输方程的无透镜定量相位显微与衍射层析成像研究

Early detection and accurate diagnosis are critical for the effective treatment and prevention of diseases. In many cases, diagnosis unfortunately requires complex and costly microscopes, which limit their applications and use to centralized settings with relatively advanced infrastructures and well-trained healthcare professionals. China is the largest developing country, and in many impoverished areas, access to adequate medical equipment may not even be available. Unfortunately, most diseases, especially the infectious diseases are quite endemic in those low-resource areas. Imagine, if we can reform the conventional microscopes to make them portable, compact and cost-effective, they can then become cheap and convenient point-of-case testing (POCT) tool resource-limited areas. In this context, this research focuses on establishing new theories, setups, and approaches for realizing high-resolution, lens-less, label-free microscopy. There are three main themes within the proposed research. The first is to retrieve the quantitative phase of the specimen in a lens-free, non-interferometric manner, based on the transport-of-equation phase retrieval and multi-wavelength partially coherent illumination. The second theme is to improve the transverse resolution of the lens-free imaging based on the new developments in the areas of image and signal processing, such as the subpixel super-resolution, image restoration and the compressive sensing. The PIs will develop a controllable micro-scanning mechanism based on tilting the flat optical component, establish the mathematical model of the pixel transfer function for lens-less imaging, and develop the accurate and efficient phase deconvolution algorithm for resolution enhancement based on compressive sensing. The third theme is to achieve the phase tomography and improve the axial resolution of the tomographic imaging by using multi-angle illuminations from a programmable two dimensional LED matrix. The multi-angle illuminations covers a finite volume of the three-dimensional spatial frequency spectrum of the object function, and then followed by an iterative constraint algorithm to fill the remaining part of the frequency support to enhance the axial resolution. The research in this direction leads towards new approaches to reconstruct high-precision high-resolution 3D phase or refractive index distribution of biological samples, such as cells and tissues, in a label-free, non-interferometric manner. In this way the fruits of the research will provide the theory foundation and technical support for next generation portable, compact and cost-effective microscopic devices with applications in low-resource settings and at the point-of-care.

实现高性能显微设备的体积小型化、成本低廉化、操作简便化，必然能够大大降低医疗检测的门槛，为资源条件有限的地区提供快捷、廉价的即时诊断（POCT）工具。在此背景下，本项目以建立高分辨、无透镜、无标记成像的新理论、新体制、新方法为科学目标：基于光强传输方程相位恢复理论，通过“多波长照明、固定物平面”的方式实现无透镜、非干涉定量相位的准确恢复；借鉴亚像元成像理论，采用倾斜平板光学元件实现可控微扫描，建立无透镜显微成像像素传递函数模型，并发展基于压缩感知的自适应高分辨率图像反卷积算法，以提高无透镜成像的横向分辨率；采用二维LED阵列多角度照明的方式覆盖物函数三维频谱空间内的支持域，并发展出基于“空域约束，频域填充”的迭代补偿算法，以提高相位层析的轴向分辨率。最终实现具有亚细胞分辨能力的无透镜、无标记成像。本项目的成功实施将为实现面向POCT应用的小型化、低成本的显微设备奠定理论基础，提供技术支撑。

实现高分辨显微设备的体积小型化、成本低廉化、操作简便化将有助于降低医疗检测的门槛，为资源稀缺地区急、重、恶性病症的早发现、早治疗提供有利条件。本项目以建立高分辨、无透镜、无标记成像的新理论、新体制、新方法为科学目标，开展基于光强传输方程的无透镜显微成像技术研究。.本项目的具体研究成果包括：.① 提出了基于最优化环形照明的光强传输方程定量相位成像与三维折射率层析成像方法，通过照明光场调制实现了相位传递函数响应与支持域的最优化，实现了横向分辨率达208nm的无标记非干涉定量动态定量相位成像以及，横向分辨率与轴向分辨率达到200 nm与645 nm的三维折射率层析。.② 提出了基于高数值孔径可编程照明的大视场超分辨显微成像技术，并搭建了满足空域频域最优采样率的高数值孔径可编程照明显微成像系统。在0.4数值孔径10倍物镜2.34mm2大视场的前提下，有效合成数值孔径达到了1.6，在波长435nm照明下的重构分辨率达到了154nm（半宽分辨率），最终的空间带宽积为0.985亿像素。相比于使用相同物镜的传统显微成像系统，空间带宽积提高了近50倍。.③ 提出了一种自适应松弛系数的无透镜超分辨显微成像技术，并搭建了基于轴向扫描的无透镜显微成像系统，实现了在29.85mm2大视场下770nm的横向半宽分辨率，达到了相机像素尺寸限制的采样分辨率的2.17倍。.上述研究成果实现高速、高分辨率的无透镜大视场高分辨显微成像，为细胞生物学，病理学，临床医学研究与诊断提供一种崭新的无标记影像学手段。.通过2016-2019四年攻关，发表期刊论文68篇（SCI-64篇），发表论文WOS引用次数447次，项目在研期间授权发明专利17件，申请发明专利46件，超额完成了预期目标。