Characterizing diffusion-controlled release of small-molecules using quantitative MRI in view of applications to orthopedic infection.

Calcium sulfate is an established carrier for localized drug delivery, but a means to non-invasively measure drug release, which would improve our understanding of localized delivery, remains an unmet need. We aim to quantitatively estimate the diffusion-controlled release of small molecules loaded into a calcium sulfate carrier through a gadobutrol-based contrast agent, which acts as a surrogate small molecule. A central cylindrical core made of calcium sulfate, either alone or within a metal scaffold, is loaded with contrast agents that release into agar. Multi-echo scans are acquired at multiple time points over 4 weeks and processed into R2* and quantitative susceptibility mapping (QSM) maps. Mean R2* values are fit to a known drug delivery model, which are then compared with the decrease in core QSM. Fitting R2* measurements of calcium sulfate core while constraining constants to a drug release model results in an R2-value of 0.991, yielding a diffusion constant of 4.59 × 10-11 m2 s-1. Incorporating the carrier within a metal scaffold results in a slower release. QSM shows the resulting loss of susceptibility in the non-metal core but is unreliable around metal. R2* characterizes the released gadobutrol, and QSM detects the resulting decrease in core susceptibility. The addition of a porous metal scaffold slows the release of gadobutrol, as expected.

Cascaded partitioned phase modulation for cross-connection of orbital angular momentum mode and polarization multiplexing channels.

Multi-dimensional orbital angular momentum (OAM) mode multiplexing provides a promising route for enlarging communication capacity and establishing comprehensive networks. While multi-dimensional multiplexing has gained advancements, the cross-connection of these multiplexed channels, especially involving modes and polarizations, remains challenging due to the needs for multi-mode interconversion and on-demand polarization control. Herein, we propose an OAM mode-polarization cross-transformation solution via cascaded partitioned phase modulation, which enables the divergently separated OAM modes to be independently phase-imposed within distinct spatial regions, leading to the synergistic conversion operation of mode and polarization channels. In demonstrations, we implemented the cross-connection of three OAM modes and two polarization multiplexed channels, achieving the mode purity that exceeds 0.951 and polarization contrast up to 0.947. The measured mode insertion losses and polarization conversion losses are below 3.42 and 3.54 dB, respectively. Consequently, 1.2 Tbit/s quadrature phase shift keying signals were successfully exchanged, yielding the bit-error-rates close to 10-6. Incorporating with increased partitioned phase treatments, this approach shows promise in accommodating massive mode-polarization multiplexed channels, which hold the potential to augment networking capability of large-scale OAM mode multiplexing communication networks.

Sharpness-aware Lookahead for Accelerating Convergence and Improving Generalization.

Lookahead is a popular stochastic optimizer that can accelerate the training process of deep neural networks. However, the solutions found by Lookahead often generalize worse than those found by its base optimizers, such as SGD and Adam. To address this issue, we propose Sharpness-Aware Lookahead (SALA), a novel optimizer that aims to identify flat minima that generalize well. SALA divides the training process into two stages. In the first stage, the direction towards flat regions is determined by leveraging a quadratic approximation of the optimization trajectory, without incurring any extra computational overhead. In the second stage, however, it is determined by Sharpness-Aware Minimization (SAM), which is particularly effective in improving generalization at the terminal phase of training. In contrast to Lookahead, SALA retains the benefits of accelerated convergence while also enjoying superior generalization performance compared to the base optimizer. Theoretical analysis of the expected excess risk, as well as empirical results on canonical neural network architectures and datasets, demonstrate the advantages of SALA over Lookahead. It is noteworthy that with approximately 25% more computational overhead than the base optimizer, SALA can achieve the same generalization performance as SAM which requires twice the training budget of the base optimizer.

Magneto-Acoustic Field-Induced Unstable Interface of Magnetic Microswarm.

Research on the interfacial instability of two-phase systems can help in gaining a better understanding of various hydrodynamic instabilities in nature. However, owing to the nonlinear and complex spatiotemporal dynamics of the unstable interface, the instability is challenging to control and suppress. This paper presents a novel interfacial instability of the magnetic microswarm induced by the competition between the destabilizing effect of magnetic field and the stabilizing effect of acoustic field. The physics underlying this novel phenomenon is discussed by analyzing the contributions of the external fields. Unlike previous studies, this study demonstrates that the instability is independent of the interfacial force or diffusion effect and can persist without dissipation over time. The manipulation of the unstable interface is further achieved by adjusting the configuration of the magneto-acoustic system. This approach can be used in thermal encoding metamaterials and has great potential applications in systems where the instability is detrimental.

Cross-connection of multiplexed cylindrical vector beams using off-axis spin-decoupled metasurfaces.

Cylindrical vector beam (CVB) multiplexing communication demands effective mode cross-connection techniques to establish communication networks. While methods like polarized grating and coordinate transformation have been developed for (de)multiplexing CVB modes, challenges persist in the cross-connection of these multiplexed mode channels, including multi-mode conversion and inhomogeneous polarization control. Herein, we present an independent off-axis spin-orbit interaction strategy utilizing spin-decoupled metasurfaces. Cross-connection is achieved by encoding conjugated Dammann optical vortex grating phases onto the two orthogonal circularly polarized components of CVBs. Experimental results demonstrate the successful interconversion of four CVB modes (CVB+1 and CVB-2, CVB+2 and CVB-4) using a Si-based metasurface with a polarization conversion efficiency exceeding 85%. This facilitates the cross-connection of 200 Gbit/s quadrature phase-shift keying signals with bit-error-rates below 10-6. Offering advantages such as ultra-compact device size, flexible control of CVB modes, and multi-mode parallel processing, this approach shows promise in advancing the networking capabilities of CVB mode multiplexing communication networks.

Multi-dimensional cylindrical vector beam (de)multiplexing through cascaded wavelength- and polarization-sensitive metasurfaces.

Cylindrical vector beams (CVBs) exhibit great potential for multiplexing communication, owing to their mode orthogonality and compatibility with conventional wavelength multiplexing techniques. However, the practical application of CVB multiplexing communication faces challenges due to the lack of effective spatial polarization manipulation technologies for (de)multiplexing multi-dimensional physical dimensions of CVBs. Herein, we introduce a wavelength- and polarization-sensitive cascaded phase modulation strategy that utilizes multiple coaxial metasurfaces for multi-dimensional modulation of CVBs. By leveraging the spin-dependent phase modulation mechanism, these metasurfaces enable the independent transformation of the two orthogonal polarization components of CVB modes. Combined with the wavelength sensitivity of Fresnel diffraction in progressive phase modulation, this approach establishes a high-dimensional mapping relationship among CVB modes, wavelengths, spatial positions, and Gaussian fundamental modes, thereby facilitating multi-dimensional (de)multiplexing involving CVB modes and wavelengths. As a proof of concept, we theoretically demonstrate a 9-channel multi-dimensional multiplexing system, successfully achieving joint (de)multiplexing of 3 CVB modes (1, 2, and 3) and 3 wavelengths (1550 nm, 1560 nm, and 1570 nm) with a diffraction efficiency exceeding 80%. Additionally, we show the transmission of 16-QAM signals across 9 channels with the bit-error-rates below 10-5. By combining the integrability of metasurfaces with the high-dimensional wavefront manipulation capabilities of multilevel modulation, our strategy can effectively address the diverse demands of different wavelengths and CVB modes in optical communication.

Ultramicroporous Tröger's Base Framework Membranes for pH-Neutral Aqueous Organic Redox Flow Batteries.

Processable polymers of intrinsic microporosity (PIMs) are emerging as promising candidates for next-generation ion exchange membranes (IEMs). However, especially with high ion exchange capacity (IEC), IEMs derived from PIMs suffer from severe swelling, thus, resulting in decreased selectivity. To solve this problem, we report ultramicroporous polymer framework membranes constructed with rigid Tröger's Base network chains, which are fabricated via an organic sol-gel process. These membranes demonstrate excellent antiswelling, with swelling ratios below 4.5% at a high IEC of 2.09 mmol g-1, outperforming currently reported PIM membranes. The rigid ultramicropore confinement and charged modification of pore channels endow membranes with both very high size-exclusion selectivity and competitive ion conductivity. The membranes thus enable the efficient and stable operation of pH-neutral aqueous organic redox flow batteries (AORFBs). This work presents the advantages of polymer framework materials as IEMs and calls for increasing attention to extending their varieties and utilization in other applications.

Prospective motion correction for brain MRI using spherical navigators.

PURPOSE: To demonstrate for the first time the feasibility of performing prospective motion correction using spherical navigators (SNAVs). METHODS: SNAVs were interleaved in a 3D FLASH sequence with an additional short baseline scan (6.8 s) for fast rotation estimation. Assessment of SNAV-based prospective motion correction was performed in six volunteers. Participant motion was guided using randomly generated stepwise prompts as well as prompts derived from real motion cases. Experiments were performed on a 3 T MRI scanner using a 32-channel head coil. RESULTS: When optimized for real-time application, SNAV-based motion estimates were computed in 25.8 ± 1.3 ms. Phantom-based quantification of rotation and translation accuracy indicated mean absolute errors of 0.10 ± 0.09° and 0.25 ± 0.14 mm, respectively. Implementing SNAV-based motion estimates for prospective motion correction led to a clear improvement in image quality with minimal increase in scan time (<5%). CONCLUSION: Optimization of SNAV processing for real-time application enables prospective motion correction with low latency and minimal scan time requirements.

Full-duplex cylindrical vector beam multiplexing communication using spin-dependent phase modulation metasurfaces.

Cylindrical vector beam (CVB) has recently gained attention as a promising carrier for signal multiplexing owing to its mode orthogonality. However, the full-duplex multiplexing communication has not been previously explored for the lack of effective technologies to parallelly couple and separate CVB modes. Herein, we present a full-duplex solution for CVB multiplexing communication that utilizes spin-dependent phase modulation metasurfaces. By independently phase-modulating the two spin eigenstates of CVBs with the metasurface via spin-dependent orbital interactions, and loading two binary Dammann vortex gratings, we enabled an independent and reciprocal wave vector manipulation of CVBs for full-duplex (de)multiplexing operation. To demonstrate this concept, we constructed a 16-channel (including 4 CVB modes and 4 wavelengths) full-duplex CVB multiplexing communication system and achieved the bidirectional transmission of 800 Gbit/s quadrature-phase shift-keying (QPSK) signals over a 5 km few-mode fiber. Our results demonstrate the successful multiplexing and demultiplexing of 2 radial CVB modes and 2 azimuthal CVB modes in full-duplex communication with the bit-error-rates approaching 1.87 × 10-5.

CACNN: Capsule Attention Convolutional Neural Networks for 3D Object Recognition.

Recently, view-based approaches, which recognize a 3D object through its projected 2-D images, have been extensively studied and have achieved considerable success in 3D object recognition. Nevertheless, most of them use a pooling operation to aggregate viewwise features, which usually leads to the visual information loss. To tackle this problem, we propose a novel layer called capsule attention layer (CAL) by using attention mechanism to fuse the features expressed by capsules. In detail, instead of dynamic routing algorithm, we use an attention module to transmit information from the lower level capsules to higher level capsules, which obviously improves the speed of capsule networks. In particular, the view pooling layer of multiview convolutional neural network (MVCNN) becomes a special case of our CAL when the trainable weights are chosen on some certain values. Furthermore, based on CAL, we propose a capsule attention convolutional neural network (CACNN) for 3D object recognition. Extensive experimental results on three benchmark datasets demonstrate the efficiency of our CACNN and show that it outperforms many state-of-the-art methods.

A magnetic field-driven multi-functional "medical ship" for intestinal tissue collection in vivo.

The incidence of intestinal cancer has risen significantly. Because of the many challenges posed by the complex environment of the intestine, it is difficult to diagnose accurately and painlessly using conventional methods, which requires the development of new body-friendly diagnostic methods. Micro- and nanomotors show great potential for biomedical applications in restricted environments. However, the difficulty of recycling has been a constraint in the collection of biological tissues for diagnostic purposes. Here, we propose a multi-functional "medical ship" (MFMS) that can be rapidly driven by a magnetic field and can reversibly "open" and "close" its internal storage space under NIR laser irradiation. It provides a transportation and recovery platform for micro- and nanomotors and cargoes. In addition, fast selection of the MFMS and magnetic nanoparticles (MNPs) can be realized through adjusting the strength and frequency of the external magnetic field. Rapid encapsulation of intestinal tissues by MNPs was achieved using a low-frequency rotating magnetic field. In addition, we demonstrated the controlled release of MNPs using the MFMS and the collection of intestinal tissues. The proposed MFMS is an intelligent and controllable transportation platform with a simple structure, which is expected to be a new tool for performing medical tasks within the digestive system.

Understanding Short-Range Memory Effects in Deep Neural Networks.

Stochastic gradient descent (SGD) is of fundamental importance in deep learning. Despite its simplicity, elucidating its efficacy remains challenging. Conventionally, the success of SGD is ascribed to the stochastic gradient noise (SGN) incurred in the training process. Based on this consensus, SGD is frequently treated and analyzed as the Euler-Maruyama discretization of stochastic differential equations (SDEs) driven by either Brownian or Lévy stable motion. In this study, we argue that SGN is neither Gaussian nor Lévy stable. Instead, inspired by the short-range correlation emerging in the SGN series, we propose that SGD can be viewed as a discretization of an SDE driven by fractional Brownian motion (FBM). Accordingly, the different convergence behavior of SGD dynamics is well-grounded. Moreover, the first passage time of an SDE driven by FBM is approximately derived. The result suggests a lower escaping rate for a larger Hurst parameter, and thus, SGD stays longer in flat minima. This happens to coincide with the well-known phenomenon that SGD favors flat minima that generalize well. Extensive experiments are conducted to validate our conjecture, and it is demonstrated that short-range memory effects persist across various model architectures, datasets, and training strategies. Our study opens up a new perspective and may contribute to a better understanding of SGD.

High-order orbital angular momentum mode-based phase shift-keying communication using phase difference modulation.

Orbital angular momentum (OAM) mode offers a promising modulation dimension for high-order shift-keying (SK) communication due to its mode orthogonality. However, the expansion of modulation order through superposing OAM modes is constrained by the mode-field mismatch resulting from the rapidly increased divergence with mode orders. Herein, we address this problem by propose a phase-difference modulation strategy that breaks the limitation of modulation orders via introducing a phase-difference degree of freedom (DoF) beyond OAM modes. Phase-difference modulation exploits the sensitivity of mode interference to phase differences, thereby providing distinct tunable parameters. This enables the generation of a series of codable spatial modes with continuous variation within the same superposed OAM modes by manipulating the interference state. Due to the inherent independence between OAM mode and phase-difference DoF, the number of codable modes increases exponentially, which facilitates establishing ultra-high-order phase shift-keying by discretizing the continuous phase difference and establishing a one-to-one mapping between coding symbols and constructed modes. We show that a phase shift-keying communication link with a modulation order of up to 4 × 104 is achieved by employing only 3 OAM modes (+1, + 2 and +3), and the decode accuracy reaches 99.9%. Since the modulation order is exponentially correlated with the OAM modes and phase differences, the order can be greatly improved by further increasing the superimposed OAM modes, which may provide new insight for high-order OAM-based SK communication.

Deep learning prediction of stroke thrombus red blood cell content from multiparametric MRI.

BACKGROUND AND PURPOSE: Thrombus red blood cell (RBC) content has been shown to be a significant factor influencing the efficacy of acute ischemic stroke treatment. In this study, our objective was to evaluate the ability of convolutional neural networks (CNNs) to predict ischemic stroke thrombus RBC content using multiparametric MR images. MATERIALS AND METHODS: Retrieved stroke thrombi were scanned ex vivo using a three-dimensional multi-echo gradient echo sequence and histologically analyzed. 188 thrombus R2*, quantitative susceptibility mapping and late-echo GRE magnitude image slices were used to train and test a 3-layer CNN through cross-validation. Data augmentation techniques involving input equalization and random image transformation were employed to improve network performance. The network was assessed for its ability to quantitatively predict RBC content and to classify thrombi into RBC-rich and RBC-poor groups. RESULTS: The CNN predicted thrombus RBC content with an accuracy of 62% (95% CI 48-76%) when trained on the original dataset and improved to 72% (95% CI 60-84%) on the augmented dataset. The network classified thrombi as RBC-rich or poor with an accuracy of 71% (95% CI 58-84%) and an area under the curve of 0.72 (95% CI 0.57-0.87) when trained on the original dataset and improved to 80% (95% CI 69-91%) and 0.84 (95% CI 0.73-0.95), respectively, on the augmented dataset. CONCLUSIONS: The CNN was able to accurately predict thrombus RBC content using multiparametric MR images, and could provide a means to guide treatment strategy in acute ischemic stroke.

Orbital angular momentum mode diversity gain in optical communication.

Vortex beams carrying orbital angular momentum (OAM) modes show superior multiplexing abilities in enhancing communication capacity. However, the signal fading induced by turbulence noise severely degrades the communication performance and even leads to communication interruption. Herein, we propose a diversity gain strategy to mitigate signal fading in OAM multiplexing communication and investigate the gain combination and channel assignment to optimize the diversity efficiency and communication capacity. Endowing signals with distinct channel matrices and superposing them with designed channel weights, we perform the diversity gain with an optimal gain efficiency, and the signal fading is mitigated by equalizing the turbulence noise. For the tradeoff between turbulence noise tolerance and communication capacity, multiplexed channels are algorithm-free assigned for diversity and multiplexing according to bit-error-rate and outage probability. As a proof of concept, we demonstrate a 6-channel multiplexing communication, where 3 OAM modes are assigned for diversity gain and 24 Gbit/s QPSK-OFDM signals are transmitted. After diversity gain, the bit-error-rate decreases from 1.41 × 10-2 to 1.63 × 10-4 at -14 dBm, and the outage probability of 86.7% is almost completely suppressed.

[Main components in different varieties of Xihuangcao by UPLC-Q-TOF-MS and UPLC-DAD].

This study established an ultra-high performance liquid chromatography coupled with quadrupole-time-of-flight mass spectrometry(UPLC-Q-TOF-MS) method to analyze the main components in different varieties of Xihuangcao and established a UPLC-DAD method to simultaneously determine the five active components(caffeic acid, rosmarinic acid, schaftoside, isoschaftoside, and oridonin).The chromatographic separation was performed on a Waters ACQUITY UPLC BEH C_(18) column(2.1 mm×100 mm, 1.7 µm) with a gradient elution of methanol(B)-water containing 0.1% formic acid(A) at a flow rate of 0.3 mL·min~(-1).The column temperature was 30 ℃.The Q-TOF-MS discriminant analysis was performed under positive electrospray ion mode and the split ratio was 1∶1. Quantitative analysis was carried out by UPLC-DAD.The determination wavelength was set at 245 nm.Thirty-two main components of Xihuangcao were separated and identified by UPLC-Q-TOF-MS, where 19 were identified in Rabdosia serra, nine in R.nervosa, 10 in R.lophanthoides, 15 in R.lophanthoides var.graciliflora, 10 in R.lophanthoides var.gerardianus, and seven in R.stracheyi.The UPLC-DVD method was developed for simultaneously determining five active components in different varieties of Xihuangcao.The standard curves for five compounds showed good linearity with correlation coefficients higher than 0.999 0.The precision, repeatability, and stability were good.The average recoveries(n=6) were between 97.01% and 102.7% with RSD<3.0%.The results of UPLC-Q-TOF-MS analysis provided a scientific basis for the use of R.stracheyi as a medicinal material of Xihuangcao and the equivalent use of R.lophanthoides var.gerardianus with R.lophanthoides var.graciliflora to some extent.The UPLC-DAD method for simultaneously determining five active components is simple, rapid, and accurate.This study can provide the basis for the quality control of different varieties of Xihuangcao.

Direct Z-Scheme Heterojunction Catalysts Constructed by Graphitic-C3N4 and Photosensitive Metal-Organic Cages for Efficient Photocatalytic Hydrogen Evolution.

The demand for improving the activity, durability, and recyclability of metal-organic cages (MOCs) that work as photocatalytic molecular devices in a homogeneous system has promoted research to combine them with other solid materials. An M2L4 type photosensitive metal-organic cage MOC-Q2 with light-harvesting ligands and catalytic Pd2+ centers has been synthesized and further heterogenized with graphitic carbon nitride to prepare a robust direct Z-scheme heterojunction photocatalyst for visible-light-driven hydrogen generation. The optimized g-C3N4/MOC-Q2 (0.7 wt%) sample exhibits a high H2 evolution activity of 6423 µmol g-1 h-1 in 5 h, and a total turnover number of 39,695 after 10 h, significantly superior to the bare MOC-Q2 used in the homogeneous solution and the comparison sample Pd/g-C3N4/L-4. The enhanced performances of g-C3N4/MOC-Q2 can be ascribed to its direct Z-scheme heterostructure, which effectively improves the charge separation and transfer efficiency. This work presents a rational approach of designing a binary photocatalytic system through combing micromolecular MOCs with heterogeneous semiconductors for water splitting.

Wheel-like Magnetic-Driven Microswarm with a Band-Aid Imitation for Patching Up Microscale Intestinal Perforation.

Microscale intestinal perforation can cause considerable mortality and is very difficult to treat using conventional methods owing to the numerous challenges associated with microscale operations, which require the development of new body-friendly and effective treatment methods. Swarming micro- and nanomotors have shown great potential in biomedical applications in complex and hard-to-reach environments. Herein, we present a wheel-like magnetic-driven microswarm (WLM) with a band-aid imitation to patch microscale intestinal perforations by pasting on the perforation point in mucus-filled environments. A method called "packing under rolling" was applied to make the formed microswarms denser and rounder. Microswarms with variable aspect ratios can be fabricated by tuning the frequency and strength of the external magnetic field. Actuation and navigation in a confined complex environment, locomotion on three-dimensional surfaces, and multiple switchable motion modes have been realized by combining AC and DC magnetic fields. Moreover, we demonstrated WLM controllable navigation, movement, and microscale perforation patching in the chicken intestines ex vivo. The proposed strategy will contribute to the treatment of microscale intestinal perforation and may be applicable to novel, precise topical medication and microsurgery.

Orbital angular momentum deep multiplexing holography via an optical diffractive neural network.

Orbital angular momentum (OAM) mode multiplexing provides a new strategy for reconstructing multiple holograms, which is compatible with other physical dimensions involving wavelength and polarization to enlarge information capacity. Conventional OAM multiplexing holography usually relies on the independence of physical dimensions, and the deep holography involving spatial depth is always limited for the lack of spatiotemporal evolution modulation technologies. Herein, we introduce a depth-controllable imaging technology in OAM deep multiplexing holography via designing a prototype of five-layer optical diffractive neural network (ODNN). Since the optical propagation with dimensional-independent spatiotemporal evolution offers a unique linear modulation to light, it is possible to combine OAM modes with spatial depths to realize OAM deep multiplexing holography. Exploiting the multi-plane light conversion and in-situ optical propagation principles, we simultaneously modulate both the OAM mode and spatial depth of incident light via unitary transformation and linear modulations, where OAM modes are encoded independently for conversions among holograms. Results show that the ODNN realized light field conversion and evolution of five multiplexed OAM modes in deep multiplexing holography, where the mean square error and structural similarity index measure are 0.03 and 86%, respectively. Our demonstration explores a depth-controllable spatiotemporal evolution technology in OAM deep multiplexing holography, which is expected to promote the development of OAM mode-based optical holography and storage.

Effective magnetic susceptibility of 3D-printed porous metal scaffolds.

PURPOSE: 3D-printed porous metal scaffolds are a promising emerging technology in orthopedic implant design. Compared to solid metal implants, porous metal implants have lower magnetic susceptibility values, which have a direct impact on imaging time and image quality. The purpose of this study is to determine the relationship between porosity and effective susceptibility through quantitative estimates informed by comparing coregistered scanned and simulated field maps. METHODS: Five porous scaffold cylinders were designed and 3D-printed in titanium alloy (Ti-6Al-4V) with nominal porosities ranging from 60% to 90% using a cellular sheet-based gyroid design. The effective susceptibility of each cylinder was estimated by comparing acquired B0 field maps against simulations of a solid cylinder of varying assigned magnetic susceptibility, where the orientation and volume of interest of the simulations was informed by a custom alignment phantom. RESULTS: Magnitude images and field maps showed obvious decreases in artifact size and field inhomogeneity with increasing porosity. The effective susceptibility was found to be linearly correlated with porosity (R2  = 0.9993). The extrapolated 100% porous (no metal) magnetic susceptibility was -9.9 ppm, closely matching the expected value of pure water (-9 ppm), indicating a reliable estimation of susceptibility. CONCLUSION: Effective susceptibility of porous metal scaffolds is linearly correlated with porosity. Highly porous implants have sufficiently low effective susceptibilities to be more amenable to routine imaging with MRI.