White-light diffraction tomography of unlabelled live cells

Taewoo Kim, Renjie Zhou, Mustafa Mir, S. Derin Babacan, P. Scott Carney, Lynford L. Goddard, Gabriel Popescu
2014 Nature Photonics  
We present a technique called white-light diffraction tomography (WDT) for imaging microscopic transparent objects such as live unlabelled cells. The approach extends diffraction tomography to white-light illumination and imaging rather than scattering plane measurements. Our experiments were performed using a conventional phase contrast microscope upgraded with a module to measure quantitative phase images. The axial dimension of the object was reconstructed by scanning the focus through the
more » ... ject and acquiring a stack of phase-resolved images. We reconstructed the threedimensional structures of live, unlabelled, red blood cells and compared the results with confocal and scanning electron microscopy images. The 350 nm transverse and 900 nm axial resolution achieved reveals subcellular structures at high resolution in Escherichia coli cells. The results establish WDT as a means for measuring three-dimensional subcellular structures in a non-invasive and label-free manner. A transparent object illuminated by an electromagnetic field generates a scattering pattern that carries specific information about its internal structure. Inferring this information from measurements of the scattered field, that is, solving the inverse scattering problem, is the fundamental principle that has allowed X-ray diffraction measurements to reveal the molecular-scale organization of crystals 1 and more recently, image cells with nanoscale resolution 2,3 . The scattered field is related to the spatially varying dielectric susceptibility of the scattering object by a transformation that simplifies considerably and, more importantly, becomes invertible, when the incident field is only weakly perturbed by the presence of the object. In this regime, the first-order Born approximation 4 and the Rytov approximation 5 have been used to unambiguously retrieve the three-dimensional spatial distribution of the dielectric constant. Implementation of inverse scattering requires knowledge of both the amplitude and phase of the scattered field. This obstacle, known as the phase problem, has been associated with X-ray diffraction measurement throughout its century-old history (for a review, see ref. 6). In 1969, Wolf proposed diffraction tomography as a reconstruction method combining the X-ray diffraction principle with optical holography 7 . Unlike X-rays, light at lower frequencies can be used in phase imaging measurements, as demonstrated by Gabor 8 . In recent years, as a result of new advances in light sources, detector arrays and computing power, quantitative phase imaging (QPI), in which optical path-length delays are measured at each point in the field of view, has become a very active field of study 9 . Whether involving holographic or non-holographic methods 10-16 , QPI presents new opportunities for studying cells and tissues non-invasively, quantitatively and without the need for staining or tagging 17-23 . Projection tomography using laser QPI has made use of ideas from X-ray imaging and enabled three-dimensional imaging of transparent structures 24-26 . More recently, this method has been applied to live cells [27] [28] [29] [30] . This type of reconstruction has a complex set-up because of the requirement to either scan the illumination angle or rotate the specimen about a fixed axis. As a result, this method is limited to shallow depths of field 31 . Importantly, without additional efforts such as synthetic aperture 30 and digital de-noising techniques 32 , laser light imaging is plagued by speckles, which ultimately limit the resolving power of the method 33 . To mitigate this problem, tomographic methods based on white light have also been proposed 34-36 . These approaches require a priori knowledge of the three-dimensional point spread function (PSF) of the instrument and ignore the physics of the light-specimen interaction. Despite these efforts, three-dimensional cell imaging is still largely restricted to confocal fluorescence microscopy, an invasive method 37 . Here, we report on a new approach for label-free tomography of live cells and other transparent specimens, which we refer to as white-light diffraction tomography (WDT). WDT offers a high-performance, simple design, as well as suitability for operation in a conventional microscopy setting. Its main features can be summarized as follows. First, WDT is a generalization of diffraction tomography to broadband illumination. Second, WDT operates in an imaging rather than a scattering geometry. Note that this is a departure from the far-zone, angular scattering that is traditionally used in X-ray diffraction. When dealing with transparent objects, measuring the complex field at the image plane yields higher sensitivity than measuring in the far-zone 38 . Third, WDT is implemented using an existing phase contrast microscope with white-light illumination, and the three-dimensional structure is recovered by simply translating the objective lens, which scans the focal plane axially through the specimen. Because phase contrast microscopes are commonly used, the method shown here could be adopted on a large scale by non-specialists.
doi:10.1038/nphoton.2013.350 fatcat:agk4wionv5bfzpcwxnvfexui74