Search

Abstract Silicon photonic signal processors promise a new generation of signal processing hardware with significant advancements in processing bandwidth, low power consumption, and minimal latency. Programmable silicon photonic signal processors, facilitated by tuning elements, can reduce hardware development cycles and costs. However, traditional programmable photonic signal processors based on optical switches face scalability and performance challenges due to control complexity and transmission losses. Here, we propose a scalable parallel signal processor on silicon for versatile applications by interleaving wavelength and temporal optical dimensions. Additionally, it incorporates ultra-low-loss waveguides and low-phase-error optical switch techniques, achieving an overall insertion loss of 10 dB. This design offers low loss, high scalability, and simplified control, enabling advanced functionalities such as accurate microwave reception, narrowband microwave photonic filtering, wide-bandwidth arbitrary waveform generation, and high-speed parallel optical computing without the need for tuning elements calibration. Our programmable parallel signal processor demonstrates advantages in both scale and performance, marking a significant advancement in large-scale, high-performance, multifunctional photonic systems.

Abstract Premature ovarian insufficiency (POI) is an essential cause of reduced fertility and quality of life in young women. Mesenchymal stem cells (MSCs) and MSCs-derived extracellular vesicles (EVs) have the ability to migrate to damaged tissues and are considered as promising therapeutic approaches for POI. However, the homing ability and therapeutic efficacy of MSCs administered in vivo are still insufficient, and their potential tumorigenicity and multi-differentiation potential also bring many doubts about their safety. The targeting ability and migration efficiency of MSCs can be improved by genetic engineering and surface modification, thereby maximizing their therapeutic efficacy. However, the use of viral vectors also has increased safety concerns. In addition, EVs, which seem to be the current therapeutic alternative to MSCs, are still poorly targeted for distribution, although they have improved in terms of safety. This paper reviews the comparative therapeutic effects of MSCs and their derived EVs on POI, their biodistribution after in vivo administration, and the most important possible ovarian targeting strategies. Difficulties such as homogeneity and yield before clinical application are also discussed. This article will provide new insights into precision therapy and targeted drug delivery for female ovarian diseases. Graphical Abstract

The coronavirus disease 2019 (COVID-19) pandemic poses an enormous threat to public health worldwide. Many retrospective studies and case reports to date have shown associations between severe COVID-19 and diseases of multi-organs. However, the research on the causal mechanisms behind this phenomenon is neither extensive nor comprehensive. We conducted a proteome-wide Mendelian randomization (MR) study using summary statistics from a Genome-Wide Association Studies (GWAS) of severe COVID-19 and diseases related to seven organs: lung, spleen, liver, heart, kidney, testis, and thyroid, based on the European ancestry. The primary analytical method used is the radial inverse variance-weighted (radial IVW) method, supplemented with the inverse variance-weighted (IVW), weighted-median (WM), MR-Egger methods. Our findings have confirmed the association between severe COVID-19 and multiple organ-related diseases, such as Hypothyroidism, strict autoimmune (HTCBSA), Thyroid disorders (TD), and Graves’ disease (GD). And we have also identified certain proteins that are associated with organ-related diseases, such as Superoxide Dismutase 2 (SOD2) and TEK Receptor Tyrosine Kinase (TEK), which are also considered potential drug targets. Phenotype scanning and sensitivity analyses were implemented to consolidate the results for Mendelian randomization. This study provides a compelling foundation for investigating COVID-19 caused diseases in future studies.

Deep learning offers a novel opportunity to achieve both high-quality and high-speed computer-generated holography (CGH). Current data-driven deep learning algorithms face the challenge that the labeled training datasets limit the training performance and generalization. The model-driven deep learning introduces the diffraction model into the neural network. It eliminates the need for the labeled training dataset and has been extensively applied to hologram generation. However, the existing model-driven deep learning algorithms face the problem of insufficient constraints. In this study, we propose a model-driven neural network capable of high-fidelity 4K computer-generated hologram generation, called 4K Diffraction Model-driven Network (4K-DMDNet). The constraint of the reconstructed images in the frequency domain is strengthened. And a network structure that combines the residual method and sub-pixel convolution method is built, which effectively enhances the fitting ability of the network for inverse problems. The generalization of the 4K-DMDNet is demonstrated with binary, grayscale and 3D images. High-quality full-color optical reconstructions of the 4K holograms have been achieved at the wavelengths of 450 nm, 520 nm, and 638 nm.

Pavement distress detection is a crucial task when assessing pavement performance conditions. Here, a novel deep-learning method based on a transformer network, referred to as ISTD-DisNet, is proposed for multi-type pavement distress semantic segmentation. In this methodology, a mix transformer (MiT) based on a hierarchical transformer structure is chosen as the backbone to obtain multi-scale feature information on pavement distress, and a mixed attention module (MAM) is introduced at the decoding stage to capture the pavement distress features across different channels and spatial locations. A learnable transposed convolution upsampling module (TCUM) enhances the model’s ability to restore multi-scale distress details. Subsequently, a novel parameter—the distress pixel density ratio (PDR)—is introduced based on the segmentation results. Analyzing the intrinsic correlation between the PDR and the pavement condition index (PCI), a new pavement damage index prediction model is proposed. Finally, the experimental results reveal that the F1 and mIOU of the proposed method are 95.51% and 91.67%, respectively, and the segmentation performance is better than that of the other seven mainstream segmentation models. Further PCI prediction model validation experimental results also indicate that utilizing the PDR enables the quantitative evaluation of the pavement damage conditions for each assessment unit, holding promising engineering application potential.

Abstract Recent advances in label-free histology promise a new era for real-time diagnosis in neurosurgery. Deep learning using autofluorescence is promising for tumor classification without histochemical staining process. The high image resolution and minimally invasive diagnostics with negligible tissue damage is of great importance. The state of the art is raster scanning endoscopes, but the distal lens optics limits the size. Lensless fiber bundle endoscopy offers both small diameters of a few 100 microns and the suitability as single-use probes, which is beneficial in sterilization. The problem is the inherent honeycomb artifacts of coherent fiber bundles (CFB). For the first time, we demonstrate an end-to-end lensless fiber imaging with exploiting the near-field. The framework includes resolution enhancement and classification networks that use single-shot CFB images to provide both high-resolution imaging and tumor diagnosis. The well-trained resolution enhancement network not only recovers high-resolution features beyond the physical limitations of CFB, but also helps improving tumor recognition rate. Especially for glioblastoma, the resolution enhancement network helps increasing the classification accuracy from 90.8 to 95.6%. The novel technique enables histological real-time imaging with lensless fiber endoscopy and is promising for a quick and minimally invasive intraoperative treatment and cancer diagnosis in neurosurgery.

Abstract In this study, we successfully constructed the new graphene oxide/poly-L-lactic acid (GO/PLLA) nanofiber scaffolds with a hydrophilic surface and porous network structure that were highly favorable for cell infiltration. When employed these new nanofiber scaffolds for a wide range of tissue engineering applications, it was expected to promote graft tissue survival and angiogenesis. The new GO/PLLA nanofiber scaffold with an appropriate concentration of 1.0 wt% was applied for the restoration of ovarian function and reserve in mice with primary ovarian insufficiency (POI). After co-transplanting the normal ovarian cortex loaded on these new nanomaterials into the in situ ovarian tissue of POI mice, the fusion of transplanted ovarian cortex with damaged ovarian tissue was improved, as well as the ovarian function and the follicle numbers. Moreover, angiogenesis was observed clearly and proved to exist in the transplanted tissue and nanomaterials, with the most conspicuous effect after co-transplantation with 1.0 wt% GO/PLLA nanofiber scaffold. In addition, nitric oxide (NO) production by phosphorylated endothelial nitric oxide synthase (p-eNOS) in vivo was proven to be involved in the effect of GO and PLLA on the improved survival rate of the transplanted ovarian cortex. This study provides a new method for the fertility preservation of ovarian tissue cryopreservation and transplantation, as well as a new strategy for the transplantation of other organs.

Graphical abstract for Quantitative phase imaging through an ultra-thin lensless fiber endoscope.

Abstract The generation of tailored complex light fields with multi-core fiber (MCF) lensless microendoscopes is widely used in biomedicine. However, the computer-generated holograms (CGHs) used for such applications are typically generated by iterative algorithms, which demand high computation effort, limiting advanced applications like fiber-optic cell manipulation. The random and discrete distribution of the fiber cores in an MCF induces strong spatial aliasing to the CGHs, hence, an approach that can rapidly generate tailored CGHs for MCFs is highly demanded. We demonstrate a novel deep neural network—CoreNet, providing accurate tailored CGHs generation for MCFs at a near video rate. The CoreNet is trained by unsupervised learning and speeds up the computation time by two magnitudes with high fidelity light field generation compared to the previously reported CGH algorithms for MCFs. Real-time generated tailored CGHs are on-the-fly loaded to the phase-only spatial light modulator (SLM) for near video-rate complex light fields generation through the MCF microendoscope. This paves the avenue for real-time cell rotation and several further applications that require real-time high-fidelity light delivery in biomedicine.

CO2 reduction under simulated sunlight over photocatalysts has become an attractive researcher area recently. In this work, carbon nitride compounds modified by TiO2 nanoparticles (TNPs) have been used for the photoreduction of CO2 in the presence of CH4 at room temperature. Briefly, a series of noble-metal-free TNP-graphitic-carbon nitride (g-C3N4, also abbreviated CN) photocatalysts with different TNPs loadings and calcination temperatures have been synthesized by a wet-chemical method. The characterization results of XRD, FTIR, SEM, TEM, BET, XPS, CO2 Adsorption, UV-vis, and PL demonstrate that the BET surface area and CO2 adsorption capacity have been improved after the calcination. Besides, the g-C3N4 has been successfully coupled with the TNPs and a heterojunction has formed at their interface. These characters contribute to increase the photocatalytic activity of TNPs-CN toward reducing CO2 in the presence of CH4, and its’ performance is better than bare g-C3N4, Titania (P25)–CN, MgO–CN, or Cu2O–CN. Orthogonal experiments are then carried out to investigate the sensitivity factors and optimum conditions. The sensitivity results show that the reaction pressure makes little difference on the photocatalysis results, which verifies the photoinduced CO2–CH4 reaction has a tiny change in gas volume. In addition, under the optimum conditions, the turnover frequency (TOF) of CO after 4 h reaction can reach 9.98 μmol g-cat.−1 h−1, and traces of ethane and ethylene have been detected during the reactions. In addition, surface acetate and carbonaceous deposit are found on the (20)TNPs-CN/450 surface after continuous 24 h irradiation under the optimum conditions, which resulting in the inactivation of the catalyst. Finally, possible reaction mechanisms have been proposed based on the results.

Progressive loss of physiological integrity and accumulation of degenerative changes leading to functional impairment and increased susceptibility to diseases are the main features of aging. The ovary, the key organ that maintains female reproductive and endocrine function, enters aging earlier and faster than other organs and has attracted extensive attention from society. Ovarian aging is mainly characterized by the progressive decline in the number and quality of oocytes, the regulatory mechanisms of which have yet to be systematically elucidated. This review discusses the hallmarks of aging to further highlight the main characteristics of ovarian aging and attempt to explore its clinical symptoms and underlying mechanisms. Finally, the intervention strategies related to aging are elaborated, especially the potential role of stem cells and cryopreservation of embryos, oocytes, or ovarian tissue in the delay of ovarian aging.

Action‐based video classification (or video‐based action recognition) is an active research area in computer vision. However, all currently utilised action‐based video classification approaches take spatial and temporal components into consideration while acoustic features (e.g. sound and speech) are neglected. In this study, the authors propose a novel approach to combine multiple cues (i.e. both visual and acoustic information) for action‐based video classification. Additionally, they introduce dense connections into their three‐stream network to address the gradient vanishing problem. Experimental results in the Kinetics Human Action Video data set and the Kinetics‐Sounds data set shows that their approach can effectively improve the accuracy in action‐based video classification.

Aging tests were used to investigate the long-term effects of BC on the immobilization of Cu, and the soil silicon dissolution of three types soils (black soil, (BS), vegetable garden soil (VS) and red soil (RS)). Litchi branch biochars (BC) at 10% (w/w) were incubated with three Cu (400 mg/kg) contaminated soils. The effect on soil properties of pH, soil organic carbon (SOC), dissolved organic carbon (DOC) and available silicon content were investigated, along with the speciation distribution of Cu. The results indicated that SOC, DOC, and available silicon content (except, BC300) increased with the application of BCs. On the other hand, the DTPA (diethylenetriaminepentaacetic acid) extractable Cu content in BS, VS and RS soils were reduced by 4–12%, 18–25%, and 12–19%, respectively. The Cu availability in all soils first increased, and then decreased during the aging process. The sum of the other four fractions, including the carbonate fraction and the inert component increased by 4–4.5% (BS), 1.4–2.1% (VS), and 0.5–1% (RS) respectively, over the long-term process. Moreover, during the whole aging process, the soil properties (such as pH, SOC, DOC and available silicon content) were almost stable. This study demonstrates that BCs, especially those produced at a higher temperature, are superior to those been produced at 300 °C in immobilizing Cu and releasing available silicon in soils. However, the remediation efficiencies were restricted by the soil type contamination status and remediation time.

This study aimed to investigate the relationship between pregnancy and myopia onset or progression in young adult women, utilizing a longitudinal cohort design with a substantial sample size. Myopia is a prevalent refractive error with significant public health and economic implications. There are genetic and environmental risk factors for myopia; however, the role of pregnancy in the development of myopia is unknown. No longitudinal studies with extensive sample sizes have clarified the relationship between pregnancy and myopia. This study included 100 women aged 20-50 years, who were followed up biennially for up to 14 years. Pregnancy occurrences and myopia assessments were recorded. A comparison of individuals with and without pregnancies revealed differences in the incidence of myopia and amblyopia, their ages, and body mass index. Hazard ratios for myopia onset or progression were significantly lower in individuals with at least one pregnancy, particularly during the early follow-up periods. Sensitivity analyses and adjustments for confounders supported these findings. This study demonstrated that pregnancy may prevent myopia onset or progression, particularly during early adulthood. Future research should confirm these findings and investigate the underlying mechanisms, informing clinical decisions, including refractive surgery timing in women of childbearing age. [ABSTRACT FROM AUTHOR]

Action-based video classification (or video-based action recognition) is an active research area in computer vision. However, all currently utilised action-based video classification approaches take spatial and temporal components into consideration while acoustic features (e.g. sound and speech) are neglected. In this study, the authors propose a novel approach to combine multiple cues (i.e. both visual and acoustic information) for action-based video classification. Additionally, they introduce dense connections into their three-stream network to address the gradient vanishing problem. Experimental results in the Kinetics Human Action Video data set and the Kinetics-Sounds data set shows that their approach can effectively improve the accuracy in action-based video classification. [ABSTRACT FROM AUTHOR]

Recent research on dense captioning based on the recurrent neural network and the convolutional neural network has made a great progress. However, mapping from an image feature space to a description space is a nonlinear and multimodel task, which makes it difficult for the current methods to get accurate results. In this paper, we put forward a novel approach for dense captioning based on hourglass-structured residual learning. Discriminant feature maps are obtained by incorporating dense connected networks and residual learning in our model. Finally, the performance of the approach on the Visual Genome V1.0 dataset and the region labelled MS-COCO (Microsoft Common Objects in Context) dataset are demonstrated. The experimental results have shown that our approach outperforms most current methods. [ABSTRACT FROM AUTHOR]