Focus Image
To date, the feasibility of super-resolution microscopy for imaging live and thick samples is still limited. Stimulated emission depletion (STED) microscopy requires high-intensity illumination to achieve sub-diffraction resolution, potentially introducing photodamage to live specimens. Moreover, the out-of-focus background may degrade the signal stemming from the focal plane. Here, we propose a new method to mitigate these limitations without drawbacks. First, we enhance a STED microscope with a detector array, enabling image scanning microscopy (ISM). Therefore, we implement STED-ISM, a method that exploits the working principle of ISM to reduce the depletion intensity and achieve a target resolution. Later, we develop Focus-ISM, a strategy to improve the optical sectioning and remove the background of any ISM-based imaging technique, with or without a STED beam. The proposed approach requires minimal architectural changes to a conventional microscope but provides substantial advantages for live and thick sample imaging.
Focus image
In this section, we demonstrate how the concept of APR can be beneficial to STED microscopy, enabling a target resolution at lower STED beam intensity. To this end, we simulated at increasing depletion power the in-focus scanned PSFs (Fig. 2a), the relative shift vectors (Fig. 2b), and the PSFs of the conventional STED and of the STED-ISM image, reconstructed with the APR method (Fig. 2c). We obtained the conventional STED images by summing all the scanned images, thus discarding the micro-image information, as it would happen with a single-element detector. The shift vectors strongly depend on the STED beam intensity: the higher the depletion power (equivalently, the saturation factor), the smaller the effective fluorescent spot, and the shorter the shift vectors. Based on the APR concept, the shift vectors reflect the maximum position of the PSFs of each scanned image. Since the excitation PSF shrinks down to a single point for increasing STED beam intensity, its product with any detection PSF shrinks as well and its maximum position approaches the optical axis. In other words, in the case of high STED beam intensities, all scanned PSFs depend mainly on the effective excitation PSF and the influence of the detection PSF becomes negligible. Thus, the scanned images vary in SNR but no longer in position. This insight shows that the adaptive pixel reassignment operation is beneficial for STED microscopy mainly for a specific range of STED beam intensities. More precisely, the spatial resolution and SNR of the STED-ISM reconstruction are improved with respect to the conventional STED counterparts for mild STED beam intensities (Fig. 2c, Supplementary Fig. 2). High STED beam intensities lead to very short shift vectors and ultimately to negligible benefits from the APR method.
In (a), we compare raw images of fluorescent beads with the corresponding ISM reconstructions. In (b), we show the shift vectors and fingerprint calculated from an image of the beads. In (c), we show the resolution gain (left) and the signal gain (right) of STED-ISM with respect of STED, for increasing STED powers. We measure the resolution and the signal, respectively, as the FWHM and the peak value of the successful fit of the fluorescent bead to a Gaussian curve. The graphs report the average values of multiple beads with the corresponding standard errors. In (d), we show detailed regions of images of tubulin-labeled fixed cells. In (e), we show detailed images of living Hela cells with SIR tubulin labeling. More specifically, we compare raw STED (left), STED-ISM (right, upper corner) and the result of multi-image deconvolution STED-ISM+ (right, bottom corner). All results are shown with increasing STED power, from left to right. The full images are shown in Supplementary Figs. 3, 4. The values in white are the images' resolution, estimated using a fit to a Gaussian model (a) or Fourier ring correlation (d and e).
To investigate the concrete advantages of STED-ISM over conventional STED microscopy, we explored the case when one is concerned most about the number of stimulating photons delivered to the sample: live-cell imaging (Fig. 3d, Supplementary Fig. 4). Also in this case, we report an enhancement of SNR and spatial resolution of the resulting STED-ISM images with respect to raw STED counterparts. The results are further improved by applying our multi-image deconvolution algorithm11,40, here completely parameter-free thanks to the PSFs estimation via Fourier ring correlation (FRC) analysis41. Moreover, we were able to perform extended STED-ISM time lapses of live Hela cells without inducing any noticeable photo-bleaching effect, given the reduced STED beam intensity necessary to obtain the target resolution (Supplementary Fig. 5)
In the following, we present a naive approach to exploit the relation between the distribution of the signal on different detector elements and its origin on the optical axis. Later, we discuss a more precise approach to identify and remove the out-of-focus background from each micro-image, and consequently to the reconstructed STED-ISM image.
a STED image of tubulin-labeled fixed Hela cell at different pinhole sizes. The bottom/right corner is normalized to the f2-STED-ISM image. Even at the smallest pinhole size, the out-of-focus light hides some structure. b f-ISM reconstruction of the STED image. The f1-ISM may introduce some negative values, here coloured blue. The f2-ISM method does not produce negative values and maintains a higher photon count. The f+-ISM image has been produced with a Richardson-Lucy deconvolution algorithm manually stopped at the 5th step and using the background estimated from the f1-ISM method. c Zoomed details of the above images, corresponding to the region identified by the dashed white box.
In (a), we show the STED image of a tubulin-labelled fixed Hela cell (left). The f2-ISM reconstruction (right) show negligible loss of in-focus photons and greatly enhanced contrast, obtained by removing the out-of-focus content. In (b), we show the confocal image of the same cell (left). Even in this case, the f2-ISM reconstruction (right) shows negligible loss of in-focus photons and greatly enhanced contrast. The axial cross-sections are obtained from the x-axis highlighted with a white dashed line.
So far, we have discussed only the out-of-focus fluorescence background generated by the excitation of the sample at positions different from the focal plane. However, the depletion beam can generate non-negligible anti-Stokes fluorescence background possibly originating from any axial plane. Because of the annular-shaped distribution of the STED beam intensity, anti-Stokes fluorescence originates mainly from the periphery of the effective excitation region. Thus, the anti-Stokes background localizes in the fingerprint and micro-images similarly to the conventional out-of-focus background (Supplementary Fig. 8), and our focus-ISM approach is also able to remove the background from this source. This feature is especially beneficial in STED implementations where the anti-Stokes background is potentially dominant, such as single-wavelength two-photon excitation STED43.
To reconstruct the high-resolution STED-ISM image we used either the simple adaptive pixel-reassignment (APR) method or a multi-image deconvolution algorithm, which are fully described in Castello et al.11. Here, we briefly review the two methods.
If, as a result, some pixels have negative values, they are trimmed to zero. The second version, named f2-ISM, is more computationally demanding but is more accurate and cannot generate non-physical results. First, a region of the image containing only in-focus emitters is manually selected to calculate the in-focus fingerprint. The latter is fitted to a single Gaussian function and its standard deviation is recorded as σsig. If it is not possible to identify a region that contains only in-focus emitters, then an additional calibration measurement is needed. Alternatively, it is possible to estimate \(\sigma _\det \) theoretically. Then the adaptive pixel reassignment method is applied to the full dataset. Subsequently, each reassigned micro-image of each pixel is normalized and fitted to the following model
The results of our method demonstrate the capabilities to obtain high-resolution and high SNR STED-ISM images and to remove the out-of-focus light from ISM images. As such, the results do not depend on the statistical variations or the properties of the used samples. Thus, each result presented in the manuscript has been reproduced a limited number of times. More in detail
Provide an onChange callback that will receive a Focus object that has x and y properties for the newly selected coordinates. Optionally supply a focus to initialize with, or a retina src to use instead of the default white ring SVG.
The element is being set to position: absolute; and having its top and left properties adjusted based on some calculations using the image and parent containers' aspect ratios and dimensions. The 's parent container gets set to position: relative; and overflow: hidden; to create the effect. There are a few other inline styles that get applied, so if anything appears to be behaving unexpectedly, be sure to check that the inline styles on both the and its parent aren't being overridden by CSS on your page (especially from rules using !important).
Additionally, because the FocusedImage is positioned absolutely so it can shift as needed, its container needs to manage its own height and width. If you aren't seeing an image appear at all, it is likely that the parent div's height is fully collapsed.
Methods: An adaptive optics IOL metrology system comprising a model eye, wavefront sensor, deformable mirror, and an image-capturing device acquired through-focus images of a letter chart with 3.0 mm and 5.0 mm pupil diameters. The system was used to induce corneal astigmatism and higher-order aberrations (HOAs) in previously measured pseudophakic presbyopic eyes. A single-optic accommodating IOL (Crystalens HD (HD500), an apodized (Restor +3.0 diopter [D] SN6AD1) and full-aperture (Tecnis ZM900) diffractive multifocal IOL, and a monofocal IOL (Acrysof SN60AT) were evaluated. Image quality was quantified using the correlation-coefficient image-quality metric. 041b061a72