3D Network Research Articles

AFSegNet: few-shot 3D ankle-foot bone segmentation via hierarchical feature distillation and multi-scale attention and fusion

Accurate segmentation of ankle and foot bones from CT scans is essential for morphological analysis. Ankle and foot bone segmentation challenges due to the blurred bone boundaries, narrow inter-bone gaps, gaps in the cortical shell, and uneven spongy bone textures. Our study endeavors to create a deep learning framework that harnesses advantages of 3D deep learning and tackles the hurdles in accurately segmenting ankle and foot bones from clinical CT scans. A few-shot framework AFSegNet is proposed considering the computational cost, which comprises three 3D deep-learning networks adhering to the principles of progressing from simple to complex tasks and network structures. Specifically, a shallow network first over-segments the foreground, and along with the foreground ground truth are used to supervise a subsequent network to detect the over-segmented regions, which are overwhelmingly inter-bone gaps. The foreground and inter-bone gap probability map are then input into a network with multi-scale attentions and feature fusion, a loss function combining region-, boundary-, and topology-based terms to get the fine-level bone segmentation. AFSegNet is applied to the 16-class segmentation task utilizing 123 in-house CT scans, which only requires a GPU with 24 GB memory since the three sub-networks can be successively and individually trained. AFSegNet achieves a Dice of 0.953 and average surface distance of 0.207. The ablation study and comparison with two basic state-of-the-art networks indicates the effectiveness of the progressively distilled features, attention and feature fusion modules, and hybrid loss functions, with the mean surface distance error decreased up to 50%.

Membrane fouling mechanisms in the presence of microplastics and organic matter: The unexpected mitigating role of Ca2+

Ultrafiltration (UF) is demonstrated to be highly effective in the removal of microplastics (MPs), but the presence of coexisting foulants introduces significant uncertainties into the associated membrane fouling behaviors. In this study, membrane fouling mechanisms were investigated when MPs, represented by polystyrene (PS), coexisted with typical organic foulants (sodium alginate, SA) and inorganic ions (Ca2+). Fouling tests revealed that the order of Ca2+ addition significantly impacted the fouling behavior of the SA-PS combined foulants. Specifically, the specific filtration resistance (SFR) was reduced by 40.82 % in the SA-PS-Ca2+ foulants and by 90.92 % in the SA-Ca2+-PS foulants, compared to the SA-PS foulants. X-ray photoelectron spectroscopy and density functional theory calculations indicated that sufficient cross-linking of Ca2+ with SA molecular chains in the SA-Ca2+-PS foulants, forming a large-scale 3D network that encapsulated more PS particles and resulted in larger flocs than those found in the SA-PS-Ca2+ foulants. According to extended Flory-Huggins theory, the improved filtration performance of the SA-PS combined foulants was due to substantial changes in chemical potential during their transition from gel to flocs upon Ca2+ addition. Furthermore, interfacial thermodynamic analyses suggested that increased repulsion between SA-Ca2+-PS foulants and between them and the membrane led to a looser fouling layer, significantly mitigating membrane fouling. This study elucidates the fouling mechanisms in the presence of MPs and other foulants from the perspectives of energy changes and molecular structures, providing novel insights for developing strategies to mitigate membrane fouling.

Insights into the structure and polymerization mechanisms of CO molecules under pressure

High pressure technique can greatly enrich the chemistry research by innovating the traditional research paradigm. Recently, tremendous attentions have been paid to the high-pressure behavior of low-Z molecules, such as CO, CO2, N2, O2 and mixtures. These molecules tend to polymerize into extended solids at the pressure of 1−100 GPa, but the structures and polymerization mechanisms are still poorly understood. Herein, as a research model, high pressure polymerization process of carbon monoxide (CO) is studied in detail both experimentally and theoretically. The in-situ Raman spectra and angle- resolved X-ray diffraction experiments prove the successful synthesis of p-CO and its amorphous structure. The theoretical simulations reveal that two CO molecules dimerize into the ethylenedione (OCCO) diradical with spin-polarized singlet state firstly, then the OCCO diradical induces the subsequent chain elongation, ring closure and chain crosslinking reactions, leading to formation of the amorphous 3D network. The multiple basic units, hybrid coordination of C/O atoms and complex connecting styles in p-CO are revealed. Based on the polymerization mechanisms, the fundamental principles governing the character (amorphous or crystalline) of extended solids under high pressure are elucidated. Due to the small dipole moment and the head-to-tail disorder of CO molecules, it is reasonable to speculate that crystalline p-CO may exist under more rigorous conditions than 110 GPa and 2000 K, at which the isoelectronic nitrogen (N2) molecules polymerize into a single-bonded cubic form of nitrogen. Our study provides a profound insight into the polymerization mechanism and structures of low-Z CO molecules under compression, contributes to the diversified chemical researches and has a generally scientific implications for the interior dynamics of planets.

Design and Development of a 3D Network Hybrid Polymeric System for Enhanced Dielectric Properties through Selective γ-Crystal Growth of Poly(PVDF-CTFE) and Reduced High-Frequency Relaxation

Design and Development of a 3D Network Hybrid Polymeric System for Enhanced Dielectric Properties through Selective γ-Crystal Growth of Poly(PVDF-CTFE) and Reduced High-Frequency Relaxation

Unraveling the Band Structure and Orbital Character of a π-Conjugated 2D Graphdiyne-Based Organometallic Network.

Graphdiyne-based carbon systems generate intriguing layered sp-sp2 organometallic lattices, characterized by flexible acetylenic groups connecting planar carbon units through metal centers. At their thinnest limit, they can result in 2D organometallic networks exhibiting unique quantum properties and even confining the surface states of the substrate, which is of great importance for fundamental studies. In this work, the on-surface synthesis of a highly crystalline 2D organometallic network grown on Ag(111) is presented. The electronic structure of this mixed honeycomb-kagome arrangement - investigated by angle-resolved photoemission spectroscopy and scanning tunneling spectroscopy - reveals a strong electronic conjugation within the network, leading to the formation of two intense electronic band-manifolds. In comparison to theoretical density functional theory calculations, it is observed that these bands exhibit a well-defined orbital character that can be associated with distinct regions of the sp-sp2 monomers. Moreover, it is found that the halogen by-products resulting from the network formation locally affect the pore-confined states, causing a significant energy shift. This work contributes to the understanding of the growth and electronic structure of graphdiyne-like 2D networks, providing insights into the development of novel carbon materials beyond graphene with tailoredproperties.

Large 3D network of concave Fe-N-C nanoparticles by nano-welding for high-power fuel cells

Large 3D network of concave Fe-N-C nanoparticles by nano-welding for high-power fuel cells

Towards harmonization of SO(3)-equivariance and expressiveness: a hybrid deep learning framework for electronic-structure Hamiltonian prediction

Abstract Deep learning for predicting the electronic-structure Hamiltonian of quantum systems necessitates satisfying the covariance laws, among which achieving SO(3)-equivariance without sacrificing the non-linear expressive capability of networks remains unsolved. To navigate the harmonization between SO(3)-equivariance and expressiveness, we propose HarmoSE, a deep learning method synergizing two distinct categories of neural mechanisms as a two-stage encoding and regression framework. The first stage corresponds to group theory-based neural mechanisms with inherent SO(3)-equivariant properties prior to the parameter learning process, while the second stage is characterized by a non-linear 3D graph Transformer network we propose, featuring high capability on non-linear expressiveness. Their combination lies in the point that, the first stage predicts baseline Hamiltonians with abundant SO(3)-equivariant features extracted, assisting the second stage in empirical learning of equivariance; and in turn, the second stage refines the first stage’s output as a fine-grained prediction of Hamiltonians using powerful non-linear neural mappings, compensating for the intrinsic weakness on non-linear expressiveness capability of mechanisms in the first stage. Our method enables precise, generalizable predictions while capturing SO(3)-equivariance under rotational transformations, and achieves state-of-the-art performance in Hamiltonian prediction tasks under multiple mean absolute error (MAE) metrics, such as the average MAE across all samples and matrix elements, the MAE for challenging samples, the MAE for different Hamiltonian blocks, and the MAE for the challenging blocks. It also demonstrates significant improvements in accuracy for downstream quantities, such as occupied orbital energy and the electronic wavefunction, as measured by MAE and cosine similarity, respectively.

Advances in Electrically Conductive Hydrogels: Performance and Applications.

Electrically conductive hydrogels are highly hydrated 3D networks consisting of a hydrophilic polymer skeleton and electrically conductive materials. Conductive hydrogels have excellent mechanical and electrical properties and have further extensive application prospects in biomedical treatment and other fields. Whereas numerous electrically conductive hydrogels have been fabricated, a set of general principles, that can rationally guide the synthesis of conductive hydrogels using different substances and fabrication methods for various application scenarios, remain a central demand of electrically conductive hydrogels. This paper systematically summarizes the processing, performances, and applications of conductive hydrogels, and discusses the challenges and opportunities in this field. In view of the shortcomings of conductive hydrogels in high electrical conductivity, matchable mechanical properties, as well as integrated devices and machines, it is proposed to synergistically design and process conductive hydrogels with applications in complex surroundings. It is believed that this will present a fresh perspective for the research and development of conductive hydrogels, and further expand the application of conductive hydrogels.

Recent Development of Fibrous Hydrogels: Properties, Applications and Perspectives.

Fibrous hydrogels (FGs), characterized by a 3D network structure made from prefabricated fibers, fibrils and polymeric materials, have emerged as significant materials in numerous fields. However, the challenge of balancing mechanical properties and functions hinders their further development. This article reviews the main advantages of FGs, including enhanced mechanical properties, high conductivity, high antimicrobial and anti-inflammatory properties, stimulus responsiveness, and an extracellular matrix (ECM)-like structure. It also discusses the influence of assembly methods, such as fiber cross-linking, interfacial treatments of fibers with hydrogel matrices, and supramolecular assembly, on the diverse functionalities of FGs. Furthermore, the mechanisms for improving the performance of the above five aspects are discussed, such as creating ion carrier channels for conductivity, in situ gelation of drugs to enhance antibacterial and anti-inflammatory properties, and entanglement and hydrophobic interactions between fibers, resulting in ECM-like structured FGs. In addition, this review addresses the application of FGs in sensors, dressings, and tissue scaffolds based on the synergistic effects of optimizing the performance. Finally, challenges and future applications of FGs are discussed, providing a theoretical foundation and new insights for the design and application of cutting-edge FGs.

Defect Self-Elimination in Nanocube Superlattices Through the Interplay of Brownian, van der Waals, and Ligand-Based Forces and Torques.

Understanding defect healing is necessary to control the electronic and optoelectronic performance of devices based on nanoparticle (NP) superlattices. However, a key challenge remains to understand how NP interactions and the resulting dynamics are coupled to defect self-elimination during assembly processes. Additional degrees of freedom that account for the anisotropic nature of NPs associated with rotational dynamics and torques further complicate the challenge. Here, we investigate nanocube (NC) superlattices by employing liquid-phase transmission electron microscopy, continuum theory, and molecular dynamics simulations. Our detailed analyses reveal that interparticle forces and torques due to ligand interactions dominate those from Brownian motions and van der Waals interactions. More importantly, NC translations and rotations induced by unbalanced forces and torques are transmitted to neighboring NCs, prompting "chain interactions" in a two-dimensional (2D) network and expediting self-elimination. The mechanistic understanding will further enable the design and fabrication of defect-free superlattices as well as those with tailored defects via assembly of anisotropic particles.

Highly Persistent and Robust Emulsion Stabilized by Hierarchical Cellulose Microgel through Stereo-hindrance under Extreme Conditions.

Emulsion stabilization oil-in-water under harsh conditions (e.g., hypersaline, acid and alkali, and high temperature) is a great challenge for conventional emulsion stabilizers, including surfactants or particles. Herein, a persistent and robust emulsion under harsh conditions is achieved via a sustainable solution based on a hierarchical cellulose microgel (h-CMG). h-CMG in aqueous suspension establishes stable 3D networks by its nanotentacles and microbody, which can spatially hinder the approach and coalescence of droplets to each other and stabilize the emulsion. Unlike the emulsion stabilized by minimizing the interfacial energy and protecting the water/oil interface, the emulsion spatially stabilized by h-CMG exhibits extensive tolerance to extreme conditions and the feature of stable high internal phase emulsion (80%, v/v). The related droplet size and emulsion volume are almost unchanged in the pH range of 1 to 14, in NaCl aqueous solution of 2 mol/L, or even at 80 °C for 12 h. This genuine biobased emulsifier is highly favorable to the food, cosmetics, and pharmaceutical industries without the risk of nanotoxicity, and it also can provide a reliable emulsifier for the chemical and drilling industries involving harsh operation circumstances.

TMIC-47. HUMAN BLOOD-BRAIN TUMOR BARRIER ON A CHIP TO INVESTIGATE PERSONALIZED TREATMENT FOR GLIOBLASTOMA PATIENTS

Abstract The inherent characteristics of glioblastoma multiforme (GBM), including tumoral heterogeneity and invasive capacity, combined with the presence of the blood-brain tumor barrier (BBTB), present challenges in developing effective treatment for GBM. Especially, the margins of GBM, where GBM cells infiltrate normal brain tissue, exhibit high resistance to therapies. Despite the difficulties in controlling tumor progression from this region, the GBM margin remains a critical area to be studied. Here we report a microengineered model that mimics the BBTB within the GBM margin, incorporating a 3D network of normal astrocytes and GBM cells isolated from patients newly diagnosed with GBM. The interaction between GBM cells and stromal cells results in increased vascular permeability, reactive gliosis, and alterations in astrocyte behavior to foster tumor invasiveness and progression. We compare patient-specific tumor responses to conventional chemotherapy in our BBTB on a chip model with clinical outcomes, demonstrating the capability of the model to predict personalized drug responses. Our BBTB model may serve as a personalized tool to examine the interactions between tumors and normal brain tissue, ultimately facilitating the screening of personalized medicine for GBM treatment.

Exploring the relationship among Alzheimer’s disease, aging and cognitive scores through neuroimaging-based approach

Alzheimer’s disease (AD) is a fatal neurodegenerative disorder, with the Mini-Mental State Examination (MMSE) and Clinical Dementia Rating (CDR) serving significant roles in monitoring its progression. We hypothesize that while cognitive assessment scores can detect AD-related brain changes, the targeted brain regions may differ. Additionally, given AD’s strong association with aging, we propose that specific brain regions are influenced by both AD pathology and aging, exhibiting strong correlations with both. To test these hypotheses, we developed a 3D convolutional network with a mixed-attention mechanism to recognize AD subjects from structural magnetic resonance imaging (sMRI) data and utilize 3D convolutional methods to pinpoint brain regions significantly correlated with the AD, MMSE, CDR and age. All models were trained and internally validated on 417 samples from the Alzheimer’s Disease Neuroimaging Initiative (ADNI), and the classification model was externally validated on 382 samples from the Australian Imaging and Lifestyle flagship (AIBL). This approach provided robust support for using MMSE and CDR in assessing AD progression and visually illustrated the relationship between aging and AD. The analysis revealed correlations among the four identification tasks (AD, MMSE, CDR and age) and highlighted asymmetric brain lesions in both AD and aging. Notably, we found that AD can accelerate aging to some extent, and a significant correlation exists between the rate of aging and cognitive assessment scores. This offers new insights into the relationship between AD and aging.

Syntheses, Structures, and Photocatalytic and Sonocatalytic Degradations of Methyl Blue of Cu(II) and Mn(II) Coordination Polymers Based on Tri(triazole) and Dicarboxylate Ligands

Cu(II) and Mn(II) coordination polymers [Cu(ttpa)(sub)]n (Cuttpa or 1) and {[Mn2(ttpa)2(nip)2(H2O)2]·3H2O}n (Mnttpa or 2) (ttpa = tris(4-(1,2,4-triazol-1-yl)phenyl)amine, H2sub = suberic acid, nip = 5-nitroisophthalicate) were hydrothermally prepared and the structures were characterized. Cuttpa exhibited a 2D (4,4) network based on [Cu2(COO)4] dimers with upper and lower dangled ttpa ligands and a 2D → 3D polythreaded network. Mnttpa showed a 2D (4,4) network with dangled uncoordinated triazole rings from ttpa ligands and nitro groups from nip2− ligands and a 2D → 3D polythreaded network. Eg data of Cuttpa and Mnttpa were 1.88 eV and 2.11 eV. Cuttpa and Mnttpa exhibited good catalytic activity for the decomposition of methyl blue (MB) under visible light and supersound irradiation. The decomposition mechanism using Cuttpa was explored. The holes (h+) and •OH hydroxyl radicals played the main roles, and the •O2− superoxide radicals played certain auxiliary roles in the decomposition of MB within the Cuttpa catalyst.

Unique cortical and subcortical activation patterns for different conspecific calls in marmosets.

The common marmoset (Callithrix jacchus) is known for its highly vocal nature, displaying a diverse range of calls. Functional imaging in marmosets has shown that the processing of conspecific calls activates a brain network that includes fronto-temporal areas. It is currently unknown whether different call types activate the same or different networks. In this study, nine adult marmosets (four females) were exposed to four common vocalizations (phee, chatter, trill, and twitter), and their brain responses were recorded using event-related fMRI at 9.4T. We found robust activations in the auditory cortices, encompassing core, belt, and parabelt regions, and in subcortical areas like the inferior colliculus, medial geniculate nucleus, and amygdala in response to these calls. Although a common network was engaged, distinct activity patterns were evident for different vocalizations that could be distinguished by a 3D convolution neural network, indicating unique neural processing for each vocalization. Our findings also indicate the involvement of the cerebellum and medial prefrontal cortex (mPFC) in distinguishing particular vocalizations from others.Significance Statement This study investigates the neural processing of vocal communications in the common marmoset (Callithrix jacchus). Utilizing event-related fMRI at 9.4T, we demonstrate that different calls (phee, chatter, trill, and twitter) elicit distinct brain activation patterns, challenging the notion of a uniform neural network for all vocalizations. Each call type distinctly engages various regions within the auditory cortices and subcortical areas. These findings offer insights into the evolutionary mechanisms of primate vocal perception and provide a foundation for understanding the origins of human speech and language processing.

Bamboo-derived aerogel with atomically dispersed Co as a high-performance bifunctional electrocatalyst for durable zinc-air batteries

The development of highly efficient and durable bifunctional electrocatalysts is crucial for rechargeable zinc-air batteries, which have garnered significant attention as a promising sustainable energy storage technology. Herein, a novel Co-N-C-1050 catalyst is synthesized using a high-temperature gas transport method, where high purity cobalt powder and bamboo biomass aerogel are treated with ammonia gas at 1050 °C. The resulting catalyst exhibits a 3D interconnected porous network composed of dense carbon fibers, which atomically dispersed Co, N, and C elements are uniformly distributed. The synergistic integration of highly active Co single atoms and N-doped 3D carbon fiber aerogel enables Co-N-C-1050 to demonstrate excellent bifunctional electrocatalytic performance for oxygen reduction reaction (ORR) and oxygen evolution reaction (OER), comparable to commercial Pt/C and RuO2 catalysts. This catalyst has a half-wave potential of 0.842 V for ORR and a potential of 1.472 V at 10 mA cm−2 for OER, outperforming N-C-1050 and benchmark catalysts. A zinc-air battery is driven by Co-N-C-1050 achieves a high power density of 135 mW/cm2 and exceptional stability over 138 h of charge/discharge cycling, surpassing the higher-cost Pt/C + RuO2 catalyst. This biomass aerogel based single-site Co-N-C catalyst offers a promising approach for developing high-efficiency and long-cycle-life zinc-air batteries.

Breathing effect in a Tetrapodal Lanthanide‐Phosphonate Organic Framework

The highly flexible organic linker hexamethylenediamine‐N,N,N’,N’‐tetrakis(methylphosphonic acid) (H8htp) was used in the preparation of a new 3D MOF formulated as [La2(SO4)2(H6htp)(H2O)4]∙2H2O (UAV‐15) by using hydrothermal synthesis. The flexibility of the organic linker posed a serious challenge in the preparation of crystalline materials: for this, the judicious selection of sulfuric acid to act as a mineralizing agent proved to be instrumental, acting in the retardation of the coordination of the phosphonic acid groups, as well as being an integral part the UAV‐15 by coordinating to the metal center. The material is a 3D network having channels filled with crystallization water molecules that can be easily removed by increasing of temperature (120ºC for 24h), leading to a single‐crystal‐to‐single‐crystal transformation originating [La2(SO4)2(H6htp)(H2O)3] (UAV‐15d). This transformation maintains the overall topology of UAV‐15 with only a small distortion of the framework. The reversibility of this transformation was studied, with the material reverting naturally to its original structure at ambient conditions for a period of 3 months or, in a faster way, when immersed in water at 120 ºC for 24h.

Hyaluronic Acid-Based 3D Bioprinted Hydrogel Structure for Directed Axonal Guidance and Modeling Innervation In Vitro.

Neurons form predefined connections and innervate target tissues through elongating axons, which are crucial for the development, maturation, and function of these tissues. However, innervation is often overlooked in tissue engineering (TE) applications. Here, multimaterial 3D bioprinting is used to develop a novel 3D axonal guidance structure in vitro. The approach uses the stiffness difference of acellular hyaluronic acid-based bioink printed as two alternating, parallel-aligned filaments. The structure has soft passages incorporated with guidance cues for axonal elongation while the stiff bioink acts as a structural support and contact guidance. The mechanical properties and viscosity differences of the bioinks are confirmed. Additionally, human pluripotent stem cell (hPSC) -derived neurons form a 3D neuronal network in the softer bioink supplemented with guidance cues whereas the stiffer restricts the network formation. Successful 3D multimaterial bioprinting of the axonal structure enables complete innervation by peripheral neurons via soft passages within 14 days of culture. This model provides a novel, stable, and long-term platform for studies of 3D innervation and axonal dynamics in health and disease.

Standalone single- and bi-layered human skin 3D models supported by recombinant silk feature native spatial organization

Physiologically relevant human skin models that include key skin cell types can be used forin vitrodrug testing, skin pathology studies, or clinical applications such as skin grafts. However, there is still no golden standard for such a model. We investigated the potential of a recombinant functionalized spider silk protein, FN-silk, for the construction of a dermal, an epidermal, and a bilayered skin equivalent (BSE). Specifically, two formats of FN-silk (i.e. 3D network and nanomembrane) were evaluated. The 3D network was used as an elastic ECM-like support for the dermis, and the thin, permeable nanomembrane was used as a basement membrane to support the epidermal epithelium. Immunofluorescence microscopy and spatially resolved transcriptomics analysis demonstrated the secretion of key ECM components and the formation of microvascular-like structures. Furthermore, the epidermal layer exhibited clear stratification and the formation of a cornified layer, resulting in a tight physiologic epithelial barrier. Our findings indicate that the presented FN-silk-based skin models can be proposed as physiologically relevant standalone epidermal or dermal models, as well as a combined BSE.

Nickel(II) and Copper(II) Dicyanoargentate Complexes with Ethylenediamine and 4,4´-Bipyridyl Ligands

The reactions of an aqueous solution of potassium dicyanoargentate with a mixture of nickel(II) or copper(II) chloride and ethylenediamine or 4,4´-bipyridyl in ethanol afford coordination polymers [Ni(En)2(Ag(CN)2)][Ag(CN)2] (I), [Cu(En)2(Ag(CN)2)][Ag(CN)2] (II), and [Cu(4,4´-Bipy)2(Ag(CN)2)2] (III) characterized by XRD (CIF files CCDC nos. 2225984 (I), 2214320 (II), and 2229270 (III)) and IR spectroscopy. According to the XRD data, the crystals of complexes I and II are formed by 1D chains {··NC– Ag–CN–M(En)2··}n (M = Ni (I), Cu (II)) linked with each other by the dicyanoargentate anions via argentophilic contacts (Ag···Ag 3.288(8) Å (I), 3.1616(14) Å (II)). The crystal of compound III consists of independent interpenetrating 3D networks built of polymer layers {Cu[Ag(CN)2]2}n bound to each other by the 4,4´-bipyridyl molecules. The bipyridyl linkers connect the Cu centers with the Ag centers of the [Ag(CN)2]– anions thus providing the tridentate coordination of the silver atoms. No Ag···Ag interactions are observed in the crystal of complex III.