Deep learning-based image deconstruction method with maintained saliency.

Visual properties that primarily attract bottom-up attention are collectively referred to as saliency. In this study, to understand the neural activity involved in top-down and bottom-up visual attention, we aim to prepare pairs of natural and unnatural images with common saliency. For this purpose, we propose an image transformation method based on deep neural networks that can generate new images while maintaining the consistent feature map, in particular the saliency map. This is an ill-posed problem because the transformation from an image to its corresponding feature map could be many-to-one, and in our particular case, the various images would share the same saliency map. Although stochastic image generation has the potential to solve such ill-posed problems, the most existing methods focus on adding diversity of the overall style/touch information while maintaining the naturalness of the generated images. To this end, we developed a new image transformation method that incorporates higher-dimensional latent variables so that the generated images appear unnatural with less context information but retain a high diversity of local image structures. Although such high-dimensional latent spaces are prone to collapse, we proposed a new regularization based on Kullback-Leibler divergence to avoid collapsing the latent distribution. We also conducted human experiments using our newly prepared natural and corresponding unnatural images to measure overt eye movements and functional magnetic resonance imaging, and found that those images induced distinctive neural activities related to top-down and bottom-up attentional processing.

Tri-view two-photon microscopic image registration and deblurring with convolutional neural networks.

Two-photon fluorescence microscopy has enabled the three-dimensional (3D) neural imaging of deep cortical regions. While it can capture the detailed neural structures in the x-y image space, the image quality along the depth direction is lower because of lens blur, which often makes it difficult to identify the neural connectivity. To address this problem, we propose a novel approach for restoring the isotropic image volume by estimating and fusing the intersection regions of the images captured from three orthogonal viewpoints using convolutional neural networks (CNNs). Because convolution on 3D images is computationally complex, the proposed method takes the form of cascaded CNN models consisting of rigid transformation, dense registration, and deblurring networks for more efficient processing. In addition, to enable self-supervised learning, we trained the CNN models with simulated synthetic images by considering the distortions of the microscopic imaging process. Through extensive experiments, the proposed method achieved substantial image quality improvements.

Generative and discriminative model-based approaches to microscopic image restoration and segmentation.

Image processing is one of the most important applications of recent machine learning (ML) technologies. Convolutional neural networks (CNNs), a popular deep learning-based ML architecture, have been developed for image processing applications. However, the application of ML to microscopic images is limited as microscopic images are often 3D/4D, that is, the image sizes can be very large, and the images may suffer from serious noise generated due to optics. In this review, three types of feature reconstruction applications to microscopic images are discussed, which fully utilize the recent advancements in ML technologies. First, multi-frame super-resolution is introduced, based on the formulation of statistical generative model-based techniques such as Bayesian inference. Second, data-driven image restoration is introduced, based on supervised discriminative model-based ML technique. In this application, CNNs are demonstrated to exhibit preferable restoration performance. Third, image segmentation based on data-driven CNNs is introduced. Image segmentation has become immensely popular in object segmentation based on electron microscopy (EM); therefore, we focus on EM image processing.

Mu-net: Multi-scale U-net for two-photon microscopy image denoising and restoration.

Advances in two two-photon microscopy (2PM) have made three-dimensional (3D) neural imaging of deep cortical regions possible. However, 2PM often suffers from poor image quality because of various noise factors, including blur, white noise, and photo bleaching. In addition, the effectiveness of the existing image processing methods is limited because of the special features of 2PM images such as deeper tissue penetration but higher image noises owing to rapid laser scanning. To address the denoising problems in 2PM 3D images, we present a new algorithm based on deep convolutional neural networks (CNNs). The proposed model consists of multiple U-nets in which an individual U-net removes noises at different scales and then yields a performance improvement based on a coarse-to-fine strategy. Moreover, the constituent CNNs employ fully 3D convolution operations. Such an architecture enables the proposed model to facilitate end-to-end learning without any pre/post processing. Based on the experiments on 2PM image denoising, we observed that our new algorithm demonstrates substantial performance improvements over other baseline methods.

In Vivo Effects of Preservative-free and Preserved Prostaglandin Analogs: Mouse Ocular Surface Study.

PURPOSE: Chronic use of topical hypotensive agents induces several side effects caused by preservatives. The purpose of this study was to evaluate the effects of prostaglandin analogs with varying concentrations of benzalkonium chloride (BAC), preservative-free (PF), and alternative preservatives on mouse corneal tissue. METHODS: Thirty-five, 8- to 10-week-old female C57BL/6 mice (five mice for each group) were used for this study. To the control group, we applied normal saline, and to each drug-treated group we applied 0.02% BAC, bimatoprost 0.01% (with BAC 0.02%), latanoprost 0.005% (with BAC 0.02%), travoprost 0.004% (with 0.001% polyquad) or tafluprost 0.0015% with/without 0.001% BAC, once a day (9 p.m.) for 4 weeks. Corneal fluorescein staining was evaluated in all groups. After harvest, the corneal tissues were embedded in paraffin and then Hematoxylin-Eosin stain was performed for histopathological examination. Immunofluorescence staining was done against TNF-α, IL-6, HLA DR, pJNK, and pAkt. RESULTS: In corneal fluorescein staining, severe punctate epithelial keratitis was seen in the groups of 0.02% BAC, 0.02% BAC containing bimatoprost 0.01% and latanoprost 0.005%. The surface desquamation, irregular surface, loss of cell borders, anisocytosis and stromal shrinkage were observed in the groups of BAC-containing eye drops. Moreover, the groups treated with BAC-containing eye drops have high inflammatory markers, significantly decreased cell viability-related signal, pAkt, and higher apoptosis-inducing signal, pJNK, than the control group. On the other hand, travoprost 0.004% and PF tafluprost 0.0015% have less cellular morphologic changes, lower inflammation, and higher cellular viability than BAC-containing formulations. CONCLUSIONS: Corneal damage, increased inflammation and apoptosis and low cell viability were observed in BAC-containing groups. PF or alternatively preserved glaucoma medications seem to be a reasonable and viable alternative to those preserved with BAC.

Deguelin inhibits vasculogenic function of endothelial progenitor cells in tumor progression and metastasis via suppression of focal adhesion.

Deguelin is a nature-derived chemopreventive drug. Endothelial progenitor cells (EPCs) are bone-marrow (BM)-derived key components to induce new blood vessels in early tumorigenesis and metastasis. Here we determined whether deguelin inhibits EPC function in vitro and in vivo at doses not affecting cancer cell apoptosis. Deguelin significantly reduced the number of EPC colony forming units of BM-derived c-kit+/sca-1+ mononuclear cells (MNCs), proliferation, migration, and adhesion to endothelial cell monolayers, and suppressed incorporation of EPC into tube-like vessel networks when co-cultured with endothelial cells. Deguelin caused cell cycle arrest at G1 without induction of apoptosis in EPC. In a mouse tumor xenograft model, tumor growth, lung metastasis and tumor-induced circulating EPCs were supressed by deguelin treatment (2 mg/kg). In mice tranplanted with GFP-expressing BM-MNCs, deguelin reduced the co-localization of CD31 and GFP, suggesting suppression of BM-derived EPC incoporation into tumor vessels. Interestingly, focal adhesion kinase (FAK)-integrin-linked kinase (ILK) activation and actin polymerization were repressed by deguelin. Decreased number of focal adhesions and a depolarized morphology was found in deguelin-treated EPCs. Taken together, our results suggest that the deguelin inhibits tumorigenesis and metastasis via EPC suppression and that suppression of focal adhesion by FAK-integrin-ILK-dependent actin remodeling is a key underlying molecular mechanism.