Ptychography over all wavelengths

Ptychography has emerged as a transformative coherent imaging technique for both fundamental and applied sciences. Its applications in optical microscopy, however, often fall short for its relatively low imaging throughput and limited resolution. In this talk, Dr Zheng will discuss a coded ptychography technique that achieves an imaging throughput order of magnitude greater than previous demonstrations. In this platform, we translate the samples across the disorder-engineered surfaces for lensless diffraction data acquisition. The engineered surface can be made by smearing a monolayer of blood on top of the image sensor. The entire system can be built using a modified Blu-ray player, where the 405 nm laser from the optical pickup head can be used as a coherent light source for sample illumination. By tracking the phase wraps of the recovered images, the group reports the direct observation of bacterial growth in a 15-s interval at a 120 mm^2 area, with a phase sensitivity like that obtained from interferometric measurements. They also characterize cell growth via longitudinal dry mass measurement and perform rapid bacterial detection at low concentrations. For drug-screening application, they demonstrate antibiotic susceptibility testing and perform single-cell analysis of antibiotic-induced filamentation. The combination of high phase sensitivity, high spatiotemporal resolution, and ultra-large field of view is unique among existing microscopy techniques. Dr Zheng will also discuss several extensions of the coded ptychography technique, including optofluidic ptychography, synthetic aperture ptychography, and depth-multiplexed on-chip imaging, thereby expanding its applicability in different fields.

Ptychography is a computational method of microscopic imaging that can generate high spatial resolution complex phase information of an object by processing coherent diffraction patterns scattered from the object. With the development of Ptychography, it has been extensively used to study materials in combination with other imaging techniques, which can give multiple contrast mechanisms of the measured sample – for example, in combination with X-ray fluorescence (XRF) microscopy. While Ptychography can achieve diffraction-limited spatial resolution, the used X-ray probe profile information usually limits the resolution of XRF images. In this talk, Dr Wu will present a machine learning based approach utilizing multimodal measurement (ie the combination of Ptychography and XRF microscopy) to enhance the resolution of the corresponding XRF microscopy by deconvolving the X-ray probe from the XRF signal. The enhanced spatial resolution was observed for both simulated and experimental XRF data, showing superior performance over the state-of-the-art scanning XRF method with different nano-sized X-ray probes. Enhanced spatial resolutions were also observed for the accompanying 3D XRF tomography reconstructions.

Due to the optically transparent nature of many biological specimens, high-resolution microscopic bio-imaging often requires the use of dyes or fluorescent labels to produce sufficient contrast that are not ideal for many in vivo experiments. By measuring the optical phase delay introduced by semi-transparent tissue, quantitative phase imaging methods like ptychography offer an effective label-free imaging platform. Many transparent specimens also exhibit alternative endogenous optical contrast mechanisms, including anisotropic properties such as material birefringence and orientation. Here, Dr Horstmeyer describes several relatively straightforward extensions of ptychography that allow us to measure this phase-sensitive anisotropic information in 3D. The only hardware modifications required to a standard microscope include two polarization-sensitive optical elements and an LED array for specimen illumination. Dr Horstmeyer then extends the standard Fourier ptychography model and inverse solver to account for the vectorial nature of light, the sample, and the microscope itself, which allows us to computationally remove polarization-dependent imaging system aberrations. He demonstrates and validate large-area, high-resolution volumetric reconstructions of specimen refractive index, birefringence, and orientation for muscle fibre sections and diseased heart tissue, the latter of which carries important information for detecting cardiac amyloidosis.

Regardless of wavelength, ptychography is conventionally configured as a far-field method, where dark-field signal is processed to realise super-resolution beyond the aperture of the probe-forming optics. Operating in this high-resolution, lens-free imaging mode, it has proven more-or-less unbeatable. However, ptychography has advantages beyond resolution gains, particularly as a high-accuracy, large field-of-view and experimentally undemanding form of quantitative phase imaging (QPI). Applications here range from unstained imaging of live cells at optical wavelengths to imaging electric and magnetic fields in the electron microscope. For these applications, ptychography in a near-field configuration is appealing because the data requirements are radically reduced: the experiment is easier and quicker, and the reconstruction process quicker and less memory-intensive than far-field ptychography. In this talk Dr Maiden will present his group's work on this near-field form of ptychography and multi-slice ptychography, spanning a huge range of wavelengths from the optical, through X-ray to the electron regimes.

High-harmonic generation (HHG) driven by ultrafast lasers is an effective way to produce broadband coherent extreme ultraviolet (EUV) radiation with a table-top setup. Such sources open up the prospect of spectroscopic nanoscale imaging of integrated circuits and biological structures. Given the limitations of high-resolution optical components for EUV radiation, lensless imaging concepts such as ptychography are essential for efficient HHG-based imaging. 

To use the full bandwidth of HHG sources effectively, ptychography is ideally performed with multiple wavelengths in parallel. As perhaps the main requirement for robust image reconstruction is diversity, an important question is how to control the HHG source properties to optimize diversity, among scan positions as well as between wavelengths. A unique property of ptychography is its ability to retrieve the probe field and object response separately, making ptychography a powerful wavefront sensing technique. This feature enables a detailed investigation of the properties of HHG beams, which can be used to study the physics of HHG. 

Dr Witte will present the group's recent work on multi-wavelength HHG-based ptychography, and its applications in EUV imaging and wavefront sensing. Specifically, he will discuss strategies towards efficient spectroscopic imaging, and how the group used ptychographic wavefront sensing to characterize complex HHG beams, and study how properties such as wavefronts and orbital angular momentum transfer from the fundamental beam to the harmonics. 

Electron microscopes use electrons with wavelengths of a few picometers, and are potentially capable of imaging individual atoms in solids at a resolution ultimately set by the intrinsic size of an atom. Even with the rapid advances in aberration-corrector technology, both residual aberrations in the electron lenses and multiple scattering of the incident beam inside the sample, the best resolution possible was an order of magnitude worse than this limit. However, with recent advances in detector technology and ptychographic algorithms to unscramble multiple scattering, the resolution of the electron microscope is now limited only by the dose to the sample, and thermal vibrations of the atoms themselves. At high doses, these approaches have allowed us to image the detailed vibrational envelopes of individual atom columns as well as locating interstitial dopant atoms that would be hidden by channelling of the probe with conventional imaging modes. Even the location of all atoms in thin amorphous films now seems within reach. However, as the dose is reduced, so is our ability to robustly reconstruct the object. Challenges in characterizing and predicting performance will be discussed.