Nanyang Normal University

Dynamic covalent chemistry (DCC) has emerged as an important framework for designing sustainable polymeric materials, offering an accessible route to recyclability, self‐healing, and environmental compatibility. By integrating dynamic covalent bonds (DCBs) such as imine, disulfide, Diels–Alder linkages, and more, DCC‐based networks enable efficient reprocessing, extend material lifespan, and facilitate closed‐loop recycling. Recent advances have demonstrated improved mechanical strength retention, enhanced healing efficiency, and the incorporation of bio‐derived monomers to reduce reliance on petrochemical feedstocks. These developments align not only with the traditional 12 principles of green chemistry but also with their updated 2020 interpretation, which emphasizes circularity, systems thinking, and lifecycle analysis. As DCC matures, optimizing bond exchange kinetics, network architectures, and integration of renewable resources will be key to enabling scalable, high‐performance, and circular polymer technologies.

To meet the growing demands of ultrawideband and high‐performance polarization‐related applications, a 3D metamaterial‐based terahertz (THz) polarization converter has been introduced. This converter integrates two types of resonators—split‐ring and cut‐wire resonators, each with unique heights into a single unit cell. It demonstrates exceptional performance in achieving ultrawideband linear polarization conversion, operating across a frequency range of 0.53–1.53 THz, with a polarization conversion ratio (PCR) consistently exceeding 0.9 and a fractional bandwidth (FBW) of 97.1%. Simulation results demonstrate that the occurrence of the wideband polarization conversion can be attributed to the strong electric and magnetic resonances between the top and bottom metal layers. Furthermore, the ultrawideband THz polarization converter is manufactured through a combination of the micro‐stereolithography‐based 3D printing technology and the electron beam evaporation deposition method. Finally, the feasibility of the fabricated ultrawideband THz polarization converter has been validated experimentally. These discoveries are thought to hold great promise by potentially offering valuable prospects for diverse polarization‐related applications within the burgeoning field of THz technologies.

Background This study aimed to identify groups of risk perception of illicit drug consumption and associated sociodemographic and mental health factors in Singapore. Methods A representative sample of 6,509 Singapore citizens and permanent residents aged 15 to 65 years was randomly selected for participation over 14 months. Information on perceived risk related to illicit drug consumption, and correlates were collected via questionnaires. Data were analysed using latent class analysis to identify risk perception profiles. Logistic regression analyses were performed to determine the sociodemographic and mental health correlates across different risk perception profiles. Results Four profiles emerged; high risk perception across all substances and frequencies, low risk perception across all substances and frequencies, low risk perception of cannabis, and low risk perception of occasional illicit drug consumption. About 9.4% (n = 557) reported low risk perception of cannabis. Males (vs. females, RR: 2.15, 95% CI: 1.57–2.94) were more likely to have a low risk perception of cannabis. Those aged 35–65 years (35–49 vs. 15–34, RR: 0.37, 95% CI: 0.25–0.54; 50–65 vs. 15–34, RR: 0.19, 95% CI: 0.10–0.37), with secondary school (vs. degree and above, RR: 0.23, 95% CI: 0.13–0.39) or pre-tertiary (vs. degree and above, RR: 0.42, 95% CI: 0.29–0.61) education were less likely to have a low risk perception of cannabis. Ever-smokers (vs. non-smokers, RR: 1.86, 95% CI: 1.31–2.64), and those grouped as “safe drinking” (vs. abstainers, RR: 3.80, 95% CI: 2.50–5.77) or “hazardous drinking” (vs. abstainers, RR: 7.00, 95% CI: 3.69–13.28) were more likely to perceive low risk of cannabis. Individuals with a low risk perception across all substances and profiles were more likely to report symptoms of anxiety (OR: 7.17, 95% CI: 1.31–38.98) when compared to high risk perception across all substances and frequencies individuals. Conclusions These findings provide critical insights for tailoring prevention and education initiatives to address substance use behaviours in Singapore.

In this perspective, the Editorial Board of the J. Opt. reflects on the past 25 years of the journal. The advances reported in journal have shaped the progress of diverse fields, from fundamental advances in optics to applications with optics as a key ingredient. The journal’s scope has seen it capture progress in several emergent fields, for instance, structured light covering orbital angular momentum, spatio-temporal solitons, topologies in light, singular optics and nonparaxial light. Reports include advances in optical devices, such as digital micromirror devices, metasurfaces and integrated photonics, as well as novel photonic materials based on nanophotonics. Application-based research includes super-resolution imaging, digital holography and nonlinear optics. We select key papers from across diverse disciplines to showcase the scope of the journal and the impact it has had on the wider community.

Detection of tumor‐associated natural killer cells (TANKs) is crucial for evaluating cancer immunotherapy because they are found to be associated with improved overall survival rate. However, existing analytical methods relying on tissue biopsy are invasive and static, restricting their capacity to deliver dynamic information. Herein, we report an activatable near‐infrared fluorescence (NIRF)/photoacoustic (PA) macromolecular reporter (BhCyNK) for real‐time imaging of TANKs. To optimize the PA performance, the fluorophore scaffold of BhCyNK is screened. Among four hemicyanine derivatives, BhCyS with the longest absorption maximum (∼800 nm) as well as the highest photothermal conversion efficiency (71.13%) and PA brightness is constructed into BhCyNK, which specifically triggers its NIRF signal (by 10‐fold increase) and PA signal (by 8.3‐fold increase) in the presence of a TANK‐overexpressed protease. BhCyNK effectively distinguishes natural killer cells from other immune cells including T cells, neutrophils, and macrophages. The high specificity of BhCyNK enables real‐time monitoring of TANKs population in the tumor of living mice amid cancer immunotherapies. The imaging results reveal that the increasing intratumoral signal of BhCyNK after combination immunotherapy correlates well with the increasing population of TANKs. Thus, this study not only reports a molecular strategy to develop efficient PA fluorophores but also provides an ideal tool for noninvasive monitoring of immune cells.

Electrochemical nitrogen reduction (eNRR) offers a sustainable and energy‐efficient alternative to the conventional Haber‐Bosch process for ammonia (NH3) synthesis, operating under mild conditions with reduced environmental impact. Open framework materials (OFMs), encompassing covalent‐organic frameworks (COFs) and metal‐organic frameworks (MOFs), have emerged as highly promising candidates due to their modular structures, tunable porosity, and adaptable functionalities. This review summarizes recent advancements in OFMs for eNRR, focusing on strategies for selection and design of active centers, regulation of porous structure, and conductivity enhancement strategy, as well as surface functionalization and interface engineering. Key challenges, including structural instability, low intrinsic conductivity, and the complexity of scalable synthesis, are critically analyzed. Advanced characterization methods, theoretical modeling, and machine learning are proposed as innovative tools to overcome these obstacles. Lastly, the potential for industrial‐scale applications of OFMs in sustainable NH3 production is discussed, highlighting their transformative role in eNRR.

The solution of proof-of-work in blockchain requires a large amount of resources; the lack of computing power on mobile devices limits the development of blockchain in mobile applications. In order to cope with the problem of insufficient resources on mobile devices, the combination of blockchain and Mobile Edge Computing has attracted much attention. In this paper, we consider an edge-enabled blockchain system containing two types of miners and a single Edge Service Provider (ESP). Different types of miners have different resource requirements and corresponding prices. Each miner makes a computation offloading decision based on the resource price set by ESP. Different miners possess varying resource requirements and correspond to distinct prices. Miners formulate task offloading decisions based on the prices set by ESP for different resource types. The ESP dynamically adjusts resource price based on miners’ offloading requests and makes admission control according to resource usage in each time slot. The problem is a joint optimization problem of computation offloading, price strategy, and admission control, which is challenging to solve due to the interaction among three issues. In this paper, we model the problem as a sequential decision-making problem and use a dynamic programming algorithm to obtain the optimal solution. Through simulation experiments, our results demonstrate that the proposed algorithm can effectively improve the overall social utility.

Accurate predictions of the Earth's near‐term (1 year and 2–5 year lead‐time) climate are crucial for informed decision‐making in various sectors such as agriculture, energy, public health, and infrastructure planning. Using the yearly initialized decadal hindcasts of the sea surface temperature (SST) from the CMIP5 and CMIP6 datasets, we evaluated their prediction skills over the North Pacific, North Atlantic, Indian Ocean, and tropical eastern Pacific for the next 5 years. In terms of spatial patterns, only the CMIP5 CanCM4 model exhibits a Pacific decadal oscillation (PDO) pattern that aligns with the observations. Although CMIP6 models cannot precisely replicate the spatial pattern of the PDO, their accurate prediction of North Pacific mid‐latitude SST indicates reliable regional forecasting. In the North Atlantic, all the models except the CMIP6 CanESM5 can reproduce a spatial pattern for the Atlantic multi‐decadal oscillation closely resembling observations. Evaluation of prediction skill over the Indian Ocean and tropical eastern Pacific was performed on an interannual scale, focusing on predictions at the first forecast lead year in terms of seasonal phase locking and prediction accuracy. Notably, the CMIP6 dataset exhibited superior performance compared to CMIP5 in the Indian Ocean. Across seasonality, all models effectively captured the seasonal peak of the Indian Ocean Dipole (IOD), occurring in September–November. CMIP6 demonstrated superior performance to CMIP5 in predicting IOD intensity, RMSE, and correlation coefficients. Evaluation over the tropical eastern Pacific revealed no significant improvement in prediction skill from CMIP5 to CMIP6. The heightened prediction skill for the IOD in the CMIP6 dataset, relative to the CMIP5 dataset, is primarily evident in the eastern tropical Indian Ocean. Additionally, models in CMIP6 could simulate a robust correlation between the dipole mode index and Niño 3.4 index, whereas those in CMIP5 could not, underscoring an advancement in predictive capabilities.

Using water droplets to generate electricity is an attractive approach for addressing the energy crisis. However, achieving high charge transfer and power output in such systems remains a major challenge. Here, a tribovoltaic nanogenerator (TVNG) is developed based on a specially designed Schottky metal‐semiconductor‐metal (MSM) structure. This device is capable of efficiently converting the kinetic energy of water droplets into electricity. To improve performance, a patterned interface layer between the metal and semiconductor is introduced, which helps guide charge flow and control surface conductivity. Upon droplet impact, the mechanical friction between the liquid and the surface generates a potential that activates charge transport across the Schottky barrier. This breaks the equilibrium state and enhances carrier movement. As a result, the device achieves a record‐high charge output of 25500 nC from a single droplet, along with an output energy of 5.8 × 10⁻⁶ J. To showcase scalability, a TVNG module with 60 cells on a 3‐inch wafer delivers milliamp‐level current and charges a 220 µF capacitor to 0.6 V within 2 s. The effects of processing, materials, structure, and droplet properties are studied to guide the future design of high‐efficiency Schottky MSM‐based TVNG.

Objective Growth differentiation factor-15 (GDF-15) is a stress-reactive cytokine which is implicated in the pathogenesis of several diabetic complications. GDF-15 may induce endothelial dysfunction which is a determinant of arterial stiffness. We aimed to study the association between GDF-15 with pulse wave velocity (PWV) and endothelial function in patients with T2D, and the potential mediating role of albuminuria. Methods This was a prospective cohort study of 1784 patients with T2D. Baseline plasma GDF-15 was measured using multiplex immunoassay. Carotid-femoral PWV was measured using applanation method. We used laser Doppler imaging with iontophoresis to measure forearm endothelium-independent vasodilation (EIV) and endothelium-dependent vasodilation (EDV). Results The mean age of patients was 57.0 ± 10.8 years. Cross-sectionally, GDF-15 was associated with higher PWV and lower EDV at baseline. Baseline GDF-15 was also associated with higher PWV and lower EIV at follow-up with adjusted coefficients 0.33 (95%CI 0.15–0.52; p < 0.001) and −2.09 (95%CI −4.14–−0.05; p = 0.045) respectively. Albuminuria mediated 29.5% and 44.9% of the association between baseline GDF-15 with follow-up PWV and EDV respectively. Conclusion Baseline GDF-15 was independently associated with higher PWV and lower EDV over time, with mediation by albuminuria in patients with T2D. GDF-15 is a potential biomarker of arterial stiffness and impaired endothelial function in T2D.

Dynamic Neural Radiance Fields (NeRF) have been successful in high-fidelity 3D modeling of talking portraits. However, slow training and inference speed have obstructed their potential usage. This paper proposes an efficient NeRF-based framework, which enables faster convergence and real-time synthesizing of stable talking portraits, by utilizing the recent success of grid-based NeRF. This is accomplished by decomposing the inherently high-dimensional talking portrait representation into three low-dimensional feature grids. Specifically, a Decomposed Audio-Spatial Encoding Module models the dynamic head with a 3D spatial grid and a 2D audio grid, where audio dynamics are modeled in a spatial-dependent manner to avoid undesirable flickering. The torso is handled with another 2D grid in a lightweight Pseudo-3D Deformable Module. Extensive experiments demonstrate that our method can generate realistic and audio-lips synchronized talking portrait videos, while also being highly efficient. Our project page is available at https://me.kiui.moe/radnerf/.

Photonic fibers have been developed into microlasers for various applications for its outstanding optical properties serving as a whispering‐gallery‐mode (WGM) resonator. Meanwhile, photonic fibers made from polymers are gaining attention due to their strong flexibility and cost‐effectiveness. However, their inherently low refractive index typically results in lower performance for polymer fiber‐based WGM microlasers in aqueous environments, limiting their use in biological applications. In this study, microbubbles are embedded within PMMA fiber, creating the flexible PMMA fiber microbubble laser. This structure led to a remarkable improvement in the microlaser's Q‐factor, increasing it ≈200‐fold and enabling stable functionality within aqueous biological environments. The fiber microbubble laser is further integrated into a microfluidic chip for multiplexed biomarker analysis in serum, achieving a detection limit for Immunoglobulin G (IgG) as low as 6.11 attomolar (aM). This microlaser can have promising applications in fundamental biological research is believed, marking a significant advancement in biosensing technology.

Quantum sensing has witnessed rapid development and transition from laboratories to practical applications in the past decade. Applications of quantum sensors, ranging from nanotechnologies to biosensing, are expected to benefit from quantum sensors' unprecedented spatial resolution and sensitivity. Solid-state spin systems are particularly attractive platforms for quantum sensing technologies because room temperature operation is viable while reaching the quantum limits of sensitivity. Among various solid-state spin systems, nitrogen-vacancy (NV) centers in diamond demonstrated high-fidelity initialization, coherent control, and high contrast readout of the electron spin state. However, electron spin coherence due to noise from the surrounding spin bath and this environment effect limits the sensitivity of NV centers. Thus, a critical task in NV center-based quantum sensing is sensitivity enhancement through coherence protection. Several strategies, such as dynamical decoupling techniques, feedback control, and nuclear spin-assisted sensing protocols, have been developed and realized for this task. Among these strategies, nuclear spin-assisted protocols have demonstrated greater enhancement of electron spin coherence. In addition, the electron and nuclear spin pair of an NV center in diamond naturally allows the application of the nuclear spin-assisted sensitivity enhancement protocol. Owing to long nuclear coherence times, further enhancement of sensitivity can be achieved by exploiting active nuclear spins (e.g., ¹⁴N, ¹³C) in the proximity of an NV center as memory ancillas when coupled with the NV center. Here, we review the spin properties of NV centers, mechanisms of the nuclear spin-assisted protocol and its gate variation, and the status of quantum sensing applications in high-resolution nuclear spin spectroscopy, atomic imaging, and magnetic field sensing. We discuss the potential for commercialization, current challenges in sensitivity enhancement, and conclude with future research directions for promoting the development of nuclear spin-assisted protocol and its integration into industrial applications.

In computer graphics, simplifying a polygonal mesh surface $\mathcal {M}$ into a geometric proxy that maintains close conformity to $\mathcal {M}$ is crucial, as it can significantly reduce computational demands in various applications. In this paper, we introduce the implicit shell (ImS), a concept designed to implicitly represent the sandwich-walled space surrounding $\mathcal {M}$, defined as $\{{\textbf {x}}\in \mathbb {R}^3|\epsilon _1\le f({\textbf {x}}) \le \epsilon _2, \epsilon _1< 0, \epsilon _2>0\}$. Here, f is an approximation of the signed distance function (SDF) of $\mathcal {M}$, and we aim to minimize the thickness $\epsilon _2-\epsilon _1$. To achieve a balance between mathematical simplicity and expressive capability in f, we employ a first-degree tri-variate tensor-product B-spline to represent f. This representation is coupled with adaptive knot grids that adapt to the inherent shape variations of $\mathcal {M}$. In this manner, the analytical form of f can be rapidly determined by solving a sparse linear system. Moreover, the process of identifying the extreme values of f among the infinitely many points on $\mathcal {M}$ can be simplified to seeking extremes among a finite set of candidate points. By exhausting the candidate points, we find the extreme values $\epsilon _1<0$ and $\epsilon _2>0$ that define the thickness. The constructed ImS is guaranteed to wrap $\mathcal {M}$ strictly, without any intersections between the bounding surfaces and $\mathcal {M}$. ImS offers numerous potential applications thanks to its rigorousness, tightness, expressiveness, and computational efficiency. We demonstrate the efficacy of ImS in mesh simplification through the control of global error.

Transitioning from horizontal surfaces to vertical walls is crucial for terrestrial robots to navigate complex environments. Replicating such impressive surface transitions in artificial insect‐scale robots has been particularly challenging. Here, innovative control schemes are introduced that enable ZoBorg (a cyborg beetle from Zophobas morio) to successfully climb walls from horizontal planes. The flex‐rigid structure, flexible footpads, sharp claws, and embedded sensors of the living insect enable ZoBorg to achieve agile locomotion with exceptional adaptability, all at low power and low cost. ZoBorg crosses low‐profile obstacles (5 and 8 mm steps) with a success rate exceeding 92% in less than one second. Most importantly, electrical stimulation of the elytron enables Zoborg to transition onto vertical walls with a success rate of 71.2% within 5 s. ZoBorg has potential applications for search and rescue missions due to its ability to traverse complex environments by crossing various obstacles, including low‐profile steps, inclines, and vertical walls.

We extend the Hybrid Unet Transformer (HUT) foundation model, which combines the advantages of the CNN and Transformer architectures with a noisy self-supervised approach, and demonstrate it in an ischemic stroke lesion segmentation task. We introduce a self-supervised approach using a noise anchor and show that it can perform better than a supervised approach under a limited amount of annotated data. We supplement our pre-training process with an additional unannotated CT perfusion dataset to validate our approach. Compared to the supervised version, the noisy self-supervised HUT (HUT-NSS) outperforms its counterpart by a margin of 2.4% in terms of dice score. HUT-NSS, on average, gained a further margin of 7.2% dice score and 28.1% Hausdorff Distance score over the state-of-the-art network USSLNet on the CT perfusion scans of the Ischemic Stroke Lesion Segmentation (ISLES2018) dataset. In limited annotated data sets, we show that HUT-NSS gained 7.87% of the dice score over USSLNet when we used 50% of the annotated data sets for training. HUT-NSS gained 7.47% of the dice score over USSLNet when we used 10% of the annotated datasets, and HUT-NSS gained 5.34% of the dice score over USSLNet when we used 1% of the annotated datasets for training. The code is available at https://github.com/vicsohntu/HUTNSS_CT.

Laser powder bed fusion (LPBF) is the dominant metal additive manufacturing technique due to its advantages in near-net-shape production of complex parts with high resolution. However, conventional quality control of LPBF-fabricated parts, including microstructure characterisation, often relies on trial-and-error experiments. These methods can be time-consuming, resource-intensive, and potentially destructive to specimens. This study introduces an image-to-image translation Cycle-consistent Generative Adversarial Network (CycleGAN)-based framework for generating statistically equivalent microstructures of LPBF-fabricated samples directly from corresponding as-printed surface inputs. The results demonstrate that the framework can effectively generate crystallographic and morphological features across 22 different process parameters for LPBF-fabricated pure nickel. The distribution of microstructural descriptors, such as grain size, grain shape, and even grain boundary misorientation angles, shows no significant difference from that measured by experiments. The generated microstructural mapping using image inputs with CycleGAN outperforms those from other generation methods on both qualitative and quantitative evaluations. The developed framework is material-agnostic and can be fine-tuned for other LPBF materials via transfer learning, providing potential applications in in-situ process optimisation and microstructure design in LPBF manufacturing.

Introduction The transmission and storage of unencrypted images over networks can lead to privacy breaches, exposing sensitive visual content. Image encryption techniques provide an effective solution to safeguard private data. Conventional encryption algorithms often rely on one-dimensional chaotic maps, which suffer from limitations such as restricted chaotic orbits and narrow parameter ranges. To address these shortcomings, this study introduces a novel two-dimensional chaotic map, 2D-CICM (Two-Dimensional Chebyshev and Infinite Collapse Map), combining the Chebyshev map and the Infinite Collapse map. The randomness of 2D-CICM is rigorously validated through attractor analysis, Lyapunov exponent evaluation, and the 0-1 test. Additionally, an image encryption algorithm leveraging 2D-CICM is proposed. Method The initial values and parameters of the chaotic map are dynamically generated based on the plaintext image's dimensions and pixel values. The scrambling phase employs dual-level permutation, targeting both pixel positions and bit-level data. In the diffusion phase, two sets of chaotic sequences, along with plaintext image information, are utilized to alter pixel values, producing the ciphertext image. The proposed 2D-CICM algorithm is currently under patent application and incorporates technological innovations designed to broaden its applicability. Results Simulations demonstrate that 2D-CICM exhibits robust performance, including an expansive key space (>2309), high key sensitivity (failure to decrypt with 10-14 key perturbations), nearideal entropy (7.9994 for Lena image), and resistance to differential attacks (NPCR 99.6126%, UACI 33.4681%). Discussion By integrating the Chebyshev map and the Infinite Collapse map, the 2D-CICM algorithm addresses several limitations of traditional chaotic maps, such as narrow parameter ranges and discontinuous chaotic orbits. It effectively withstands diverse cryptographic attacks, ensuring superior security Conclusion The 2D-CICM algorithm offers strong security guarantees and holds significant potential for applications in network information security and secure image communication.

Handwriting identification is widely accepted as scientific evidence. However, its authenticity is questioned because it depends on the appraiser's professional skills and susceptibility to deliberate false identification by expert witnesses. Consequently, there is an urgent need for an effective handwriting identification system (HWIS) that reduces reliance on the appraiser's skills and mitigates the risk of international false identification. Here, we report a HWIS that integrates a self‐powered handwriting signal data acquisition device with an advanced deep learning architecture possessing powerful feature extraction ability and one‐class classification function. The device successfully captures the characteristic differences in handwriting behavior between genuine writers and forgers, and the handwriting identification results demonstrate the excellent performance of our system, showcasing its powerful potential to solve the longstanding challenge of handwriting identification that has perplexed humans for a considerable period. Moreover, this work exhibits the system's capability for remote access and downloading the handwriting signal data through the data cloud, highlighting its practical value for fulfilling the requirements of handwriting recognition and identification applications, and it can effectively advance signature information security and ensure the protection of private information. image