Enhancing Light-Sheet Fluorescence Microscopy Illumination Beams through Deep Design Optimization.

Light sheet fluorescence microscopy (LSFM) provides the benefit of optical sectioning coupled with rapid acquisition times for imaging of tissue-cleared specimen. This allows for high-resolution 3D imaging of large tissue volumes. Inherently to LSFM, the quality of the imaging heavily relies on the characteristics of the illumination beam, with the notion that the illumination beam only illuminates a thin section that is being imaged. Therefore, substantial efforts are dedicated to identifying slender, non-diffracting beam profiles that can yield uniform and high-contrast images. An ongoing debate concerns the employment of the most optimal illumination beam; Gaussian, Bessel, Airy patterns and/or others. Comparisons among different beam profiles is challenging as their optimization objective is often different. Given that our large imaging datasets (~0.5TB images per sample) is already analyzed using deep learning models, we envisioned a different approach to this problem by hypothesizing that we can tailor the illumination beam to boost the deep learning models performance. We achieve this by integrating the physical LSFM illumination model after passing through a variable phase mask into the training of a cell detection network. Here we report that the joint optimization continuously updates the phase mask, improving the image quality for better cell detection. Our method's efficacy is demonstrated through both simulations and experiments, revealing substantial enhancements in imaging quality compared to traditional Gaussian light sheet. We offer valuable insights for designing microscopy systems through a computational approach that exhibits significant potential for advancing optics design that relies on deep learning models for analysis of imaging datasets.

Deep learning-based adaptive optics for light sheet fluorescence microscopy.

Light sheet fluorescence microscopy (LSFM) is a high-speed imaging technique that is often used to image intact tissue-cleared specimens with cellular or subcellular resolution. Like other optical imaging systems, LSFM suffers from sample-induced optical aberrations that decrement imaging quality. Optical aberrations become more severe when imaging a few millimeters deep into tissue-cleared specimens, complicating subsequent analyses. Adaptive optics are commonly used to correct sample-induced aberrations using a deformable mirror. However, routinely used sensorless adaptive optics techniques are slow, as they require multiple images of the same region of interest to iteratively estimate the aberrations. In addition to the fading of fluorescent signal, this is a major limitation as thousands of images are required to image a single intact organ even without adaptive optics. Thus, a fast and accurate aberration estimation method is needed. Here, we used deep-learning techniques to estimate sample-induced aberrations from only two images of the same region of interest in cleared tissues. We show that the application of correction using a deformable mirror greatly improves image quality. We also introduce a sampling technique that requires a minimum number of images to train the network. Two conceptually different network architectures are compared; one that shares convolutional features and another that estimates each aberration independently. Overall, we have presented an efficient way to correct aberrations in LSFM and to improve image quality.

Maternal organophosphate flame retardant exposure alters the developing mesencephalic dopamine system in fetal rat.

Organophosphate flame retardants (OPFRs) have become the predominant substitution for legacy brominated flame retardants but there is concern about their potential developmental neurotoxicity (DNT). OPFRs readily dissociate from the fireproofed substrate to the environment, and they (or their metabolites) have been detected in diverse matrices including air, water, soil, and biota, including human urine and breastmilk. Given this ubiquitous contamination, it becomes increasingly important to understand the potential effects of OPFRs on the developing nervous system. We have previously shown that maternal exposure to OPFRs results in neuroendocrine disruption, alterations to developmental metabolism of serotonin (5-HT) and axonal extension in male fetal rats, and potentiates adult anxiety-like behaviors. The development of the serotonin and dopamine systems occur in parallel and interact, therefore, we first sought to enhance our prior 5-HT work by first examining the ascending 5-HT system on embryonic day 14 using whole mount clearing of fetal heads and 3-dimensional (3D) brain imaging. We also investigated the effects of maternal OPFR exposure on the development of the mesocortical dopamine system in the same animals through 2-dimensional and 3D analysis following immunohistochemistry for tyrosine hydroxylase (TH). Maternal OPFR exposure induced morphological changes to the putative ventral tegmental area and substantia nigra in both sexes and reduced the overall volume of this structure in males, whereas 5-HT nuclei were unchanged. Additionally, dopaminergic axogenesis was disrupted in OPFR exposed animals, as the dorsoventral spread of ventral telencephalic TH afferents were greater at embryonic day 14, while sparing 5-HT fibers. These results indicate maternal exposure to OPFRs alters the development trajectory of the embryonic dopaminergic system and adds to growing evidence of OPFR DNT.

Quantitative analysis of illumination and detection corrections in adaptive light sheet fluorescence microscopy.

Light-sheet fluorescence microscopy (LSFM) is a high-speed, high-resolution and minimally phototoxic technique for 3D imaging of in vivo and in vitro specimens. LSFM exhibits optical sectioning and when combined with tissue clearing techniques, it facilitates imaging of centimeter scale specimens with micrometer resolution. Although LSFM is ubiquitous, it still faces two main challenges that effect image quality especially when imaging large volumes with high-resolution. First, the light-sheet illumination plane and detection lens focal plane need to be coplanar, however sample-induced aberrations can violate this requirement and degrade image quality. Second, introduction of sample-induced optical aberrations in the detection path. These challenges intensify when imaging whole organisms or structurally complex specimens like cochleae and bones that exhibit many transitions from soft to hard tissue or when imaging deep (> 2 mm). To resolve these challenges, various illumination and aberration correction methods have been developed, yet no adaptive correction in both the illumination and the detection path have been applied to improve LSFM imaging. Here, we bridge this gap, by implementing the two correction techniques on a custom built adaptive LSFM. The illumination beam angular properties are controlled by two galvanometer scanners, while a deformable mirror is positioned in the detection path to correct for aberrations. By imaging whole porcine cochlea, we compare and contrast these correction methods and their influence on the image quality. This knowledge will greatly contribute to the field of adaptive LSFM, and imaging of large volumes of tissue cleared specimens.

Illumination angle correction during image acquisition in light-sheet fluorescence microscopy using deep learning.

Light-sheet fluorescence microscopy (LSFM) is a high-speed imaging technique that provides optical sectioning with reduced photodamage. LSFM is routinely used in life sciences for live cell imaging and for capturing large volumes of cleared tissues. LSFM has a unique configuration, in which the illumination and detection paths are separated and perpendicular to each other. As such, the image quality, especially at high resolution, largely depends on the degree of overlap between the detection focal plane and the illuminating beam. However, spatial heterogeneity within the sample, curved specimen boundaries, and mismatch of refractive index between tissues and immersion media can refract the well-aligned illumination beam. This refraction can cause extensive blur and non-uniform image quality over the imaged field-of-view. To address these issues, we tested a deep learning-based approach to estimate the angular error of the illumination beam relative to the detection focal plane. The illumination beam was then corrected using a pair of galvo scanners, and the correction significantly improved the image quality across the entire field-of-view. The angular estimation was based on calculating the defocus level on a pixel level within the image using two defocused images. Overall, our study provides a framework that can correct the angle of the light-sheet and improve the overall image quality in high-resolution LSFM 3D image acquisition.

Recent progress in digital holography with dynamic diffractive phase apertures [Invited].

Digital holography with diffractive phase apertures is a hologram recording technique in which at least one of the interfering waves is modulated by a phase mask. In this review, we survey several main milestones on digital holography with dynamic diffractive phase apertures. We begin with Fresnel incoherent correlation holography (FINCH), a hologram recorder with an aperture of a diffractive lens. FINCH has been used for many applications such as 3D imaging, fluorescence microscopy, superresolution, image processing, and imaging with sectioning ability. FINCH has played an important role by inspiring other digital holography systems based on diffractive phase aperture, such as Fourier incoherent single-channel holography and coded aperture correlation holography, which also are described in this review.

Optical incoherent synthetic aperture imaging by superposition of phase-shifted optical transfer functions.

The concept of an optical incoherent synthetic aperture is widely used in astronomical interferometric telescopes. In this Letter, we propose a new, to the best of our knowledge, method to realize optical incoherent synthetic aperture imaging. The method is based on a superposition of optical transfer functions of incoherent imaging systems. Only two small sub-apertures, out of a much larger full synthetic aperture, are open at any given time, and they transfer light from the observed object to the image sensor. During the imaging process, the two sub-apertures move over the full synthetic aperture, where the gap between them starts from zero and grows with time. For every position of the pair of sub-apertures, two images are captured. In one of the images, the sub-apertures have the same phase value, and in the other image, one of the sub-apertures is phase shifted by π radian relative to the other one. The final image with the image resolution of the synthetic aperture is obtained as a superposition of the entire recorded images. Optical experiments are performed on reflective objects, and results of the synthetic aperture-based method demonstrate an imaging performance similar to that of direct imaging by a system with a single aperture of the size of the synthetic aperture.

Depth-of-field engineering in coded aperture imaging.

Extending the depth-of-field (DOF) of an optical imaging system without effecting the other imaging properties has been an important topic of research for a long time. In this work, we propose a new general technique of engineering the DOF of an imaging system beyond just a simple extension of the DOF. Engineering the DOF means in this study that the inherent DOF can be extended to one, or to several, separated different intervals of DOF, with controlled start and end points. Practically, because of the DOF engineering, entire objects in certain separated different input subvolumes are imaged with the same sharpness as if these objects are all in focus. Furthermore, the images from different subvolumes can be laterally shifted, each subvolume in a different shift, relative to their positions in the object space. By doing so, mutual hiding of images can be avoided. The proposed technique is introduced into a system of coded aperture imaging. In other words, the light from the object space is modulated by a coded aperture and recorded into the computer in which the desired image is reconstructed from the recorded pattern. The DOF engineering is done by designing the coded aperture composed of three diffractive elements. One element is a quadratic phase function dictating the start point of the in-focus axial interval and the second element is a quartic phase function which dictates the end point of this interval. Quasi-random coded phase mask is the third element, which enables the digital reconstruction. Multiplexing several sets of diffractive elements, each with different set of phase coefficients, can yield various axial reconstruction curves. The entire diffractive elements are displayed on a spatial light modulator such that real-time DOF engineering is enabled according to the user needs in the course of the observation. Experimental verifications of the proposed system with several examples of DOF engineering are presented, where the entire imaging of the observed scene is done by single camera shot.

Resolution-enhanced imaging using interferenceless coded aperture correlation holography with sparse point response.

Interferenceless coded aperture correlation holography (I-COACH) is a non-scanning, motionless, incoherent digital holography technique. In this study we use a special type of I-COACH in which its point spread hologram (PSH) is ensemble of sparse dots. With this PSH an imaging resolution beyond the classic diffraction limit is demonstrated. This resolution improvement is achieved due to the position of the coded aperture between the object and the lens-based imaging system. The coded aperture scatters part of the light, that otherwise is blocked by the system aperture, into the optical system, and by doing that, extends the effective numerical aperture of the system. The use of sparse PSH increases the signal-to-noise ratio of the entire imaging system. A lateral resolution enhancement by a factor of about 1.6 was noted in the case of I-COACH compared to direct imaging.

Noise suppression by controlling the sparsity of the point spread function in interferenceless coded aperture correlation holography (I-COACH).

Interferenceless coded aperture correlation holography (I-COACH) is an incoherent opto-digital technique for imaging 3D objects. In I-COACH, the light scattered from an object is modulated by a coded phase mask (CPM) and then recorded by a digital camera as an object digital hologram. To reconstruct the image, the object hologram is cross-correlated with the point spread function (PSF)-the intensity response to a point at the same object's axial location recorded with the same CPM. So far in I-COACH systems, the light from each object point has scattered over the whole camera area. Hence, the signal-to-noise ratio per camera pixel is lower in comparison to the direct imaging in which each point is imaged to a single image point. In this work, we consider the midway between the camera responses of a single point and of a continuous pattern over the entire camera area. The light in this study is focused onto a set of dots randomly distributed over the camera plane. With this technique, we show that there is a PSF with a best number of dots, yielding an image with a maximum product of the signal-to-noise ratio and the image visibility and a maximum value of structural similarity.

Superresolution beyond the diffraction limit using phase spatial light modulator between incoherently illuminated objects and the entrance of an imaging system.

We present a superresolution technique for imaging objects beyond the diffraction limit imposed by the limited numerical aperture (NA) of a general optical system. A coded phase mask (CPM) displayed on a spatial light modulator is introduced between the object and the input aperture of an ordinary lens-based imaging system. Consequently, the effective NA is increased beyond the inherent NA of the optical imaging system. Unlike conventional systems, the imaging in our proposed method is not direct from an object to a sensor, and the system requires a one-time calibration. In the calibration mode, a point object is mounted in the object plane, and the point spread intensity pattern is recorded. Following the calibration, the system is ready for imaging an arbitrary number of 2D objects. The intensity pattern from any object placed at the same axial location of the point object, and modulated by the same CPM, is recorded once by a digital camera. The superresolved image of the object is reconstructed by a nonlinear cross-correlation between the abovementioned two intensity patterns. The effective NA and the new resolution limit can be tuned by changing the scattering degree of the CPM.

Resolution enhancement in nonlinear interferenceless COACH with point response of subdiffraction limit patterns.

Interferenceless coded aperture correlation holography (I-COACH) is a non-scanning, motionless, incoherent digital holography technique for 3D imaging. The lateral and axial resolutions of I-COACH are equivalent to those of conventional direct imaging with the same numerical aperture. The main component of I-COACH is a coded phase mask (CPM) used as the system aperture. In this study, the CPM has been engineered using a modified Gerchberg-Saxton algorithm to generate a random distribution of subdiffraction spot arrays on the digital camera as a system response to a point source illumination. A library of point object holograms is created to calibrate the system for imaging different lateral sections of a 3D object. An object is placed within the calibrated 3D space and an object hologram is recorded with the same CPM. The various planes of the object are reconstructed by a non-linear cross-correlation between the object hologram and the point object hologram library. A lateral resolution enhancement of about 25% was noted in the case of I-COACH compared to direct imaging.

Extending the field of view by a scattering window in an I-COACH system.

Interferenceless coded aperture correlation holography (I-COACH) is an incoherent digital holography technique developed to record and reconstruct 3D images of objects without two-wave interference. Herein, we introduce a novel technique to extend the field of view (FOV) of I-COACH beyond the limit imposed by the ratio between the finite area of the image sensor and the magnification of the optical system. Light diffracted from a point object located on the optical axis is modulated by a pseudorandom coded phase mask, and the central part of the point spread hologram (PSH) on the image sensor is recorded. The point object is shifted laterally to predetermined lateral locations in order to collect the exterior parts of the PSH. The recorded PSHs are stitched together to produce a synthetic PSH (SPSH) with an area nine times that of any individual PSH recorded by the image sensor. An object with a lateral extent beyond the FOV limit of the image sensor is placed at the same axial location as the point object, and the object hologram is recorded. The object is reconstructed by a cross-correlation between the zero-padded object hologram and the SPSH. Hence, the object parts beyond the FOV limit of the image sensor are recovered. An SPSH library is created for different axial planes, and the corresponding axial planes of the object are reconstructed.