3D Architecture Research Articles

New methods at the transition from research to routine diagnostics

Pathology, traditionally focused on classification and diagnosis, is continuously evolving through new technologies. Advances in proteomics, epigenetics, tissue staining, and 3D imaging expand the possibilities of classical morphology. The aim of this study was to investigate how modern technologies can improve diagnostic accuracy and therapy selection and how they can be integrated into pathologic routine diagnostics. Recent studies in proteomics, epigenetics, multiplex tissue staining, and 3D tissue imaging were analyzed to assess their application and the challenges of clinical implementation. The analysis shows significant potential for pathologic diagnostics. Proteomics provides adeeper understanding of the molecular architecture of tumors, while epigenetics and 3D genome architecture offer new insights into genetic regulation and tumor heterogeneity. Multiplex tissue staining and 3D tissue imaging improve spatial tissue analysis. Despite the potential to improve diagnostics, high costs, technical complexity, and lack of standardization hinder integration into clinical practice. Nevertheless, these technologies offer promising approaches for optimizing diagnostics and therapy selection. Research and interdisciplinary collaboration are crucial to successfully integrating these innovations into routine clinical practice.

Autonomous Self-Healing Magnetoelectric I-Skin from Self-Bonded Deep Eutectic Polymer.

Next-generation ionic skin (i-skin) should be self-healing and self-powered, promoting its development toward lightweight, miniaturization, compact, and portable designs. Previously reported self-powered i-skin mostly either lack the ability to self-repair damaged parts or only have self-healing capabilities some components, falling short of achieving complete device self-healability. In this work, a self-bonding strategy is presented to obtain an all-polymerizable deep eutectic solvent (PDES) magnetoelectric i-skin (MIS) that simultaneously achieves self-powering and full-device autonomous self-healability. The three-layered MIS can easily restore mechanical and electrochemical performance at the full-device level without requiring any external stimulus. The developed MIS can be easily configured into various 3D architectures with highly compatible magnetic and conductive components, offering promising potential for the advancement of embodied energy technologies. The present work provides a versatile and user-friendly platform for producing a wide range of intrinsic self-healing multi-layered devices made from soft materials, with potential applications extending beyond human-machine interfaces and artificial intelligence.

Blink Detection Using 3D Convolutional Neural Architectures and Analysis of Accumulated Frame Predictions.

Blink detection is considered a useful indicator both for clinical conditions and drowsiness state. In this work, we propose and compare deep learning architectures for the task of detecting blinks in video frame sequences. The first step is the training and application of an eye detector that extracts the eye regions from each video frame. The cropped eye regions are organized as three-dimensional (3D) input with the third dimension spanning time of 300 ms. Two different 3D convolutional neural networks are utilized (a simple 3D CNN and 3D ResNet), as well as a 3D autoencoder combined with a classifier coupled to the latent space. Finally, we propose the usage of a frame prediction accumulator combined with morphological processing and watershed segmentation to detect blinks and determine their start and stop frame in previously unseen videos. The proposed framework was trained on ten (9) different participants and tested on five (8) different ones, with a total of 162,400 frames and 1172 blinks for each eye. The start and end frame of each blink in the dataset has been annotate by specialized ophthalmologist. Quantitative comparison with state-of-the-art blink detection methodologies provide favorable results for the proposed neural architectures coupled with the prediction accumulator, with the 3D ResNet being the best as well as the fastest performer.

Automated 3D semantic segmentation of PCB X-ray CT images and netlist extraction

Printed Circuit Board (PCB) design reconstruction is essential for addressing part obsolescence, intellectual property recovery, compliance, quality assurance, and enhancing national capabilities. Traditional methods for PCB design extraction, both non-geometry-based and geometry-based, have limitations in accuracy, efficiency, and scalability. This paper presents an automated approach, combining image processing and machine learning, to achieve 3D semantic segmentation of PCB X-ray Computed Tomography (X-ray CT) images and subsequent netlist extraction. By employing a 3D U-Net architecture with a ResNet-18 backbone and training on synthetic data, we introduce a first-of-its-kind method for direct 3D semantic segmentation, significantly improving over previous efforts. Our approach eliminates the need for extensive labeled datasets by using inherently labeled synthetic data. Further, this method enhances ease of segmentation by significantly reducing or eliminating the preprocessing effort required for 2D image stacks. It also improves universality by expanding the scope of application beyond images with specific 2D stack criteria, segmenting the 3D image in its entirety. Additionally, this method enables the processing of images of PCBs that have undergone bending, which is common among PCBs with a thickness below a certain threshold. The implications of this approach extend beyond PCBs, finding applications in various physical and biological sciences where 3D image segmentation is crucial. This methodology includes high-resolution 3D imaging, watershed segmentation, machine learning-based semantic segmentation, and netlist extraction. Validation with both synthetic and real-world PCB datasets shows high accuracy and robustness, offering a scalable solution for PCB design reconstruction.

Efficient Quantum Circuit Design with a Standard Cell Approach, with an Application to Neutral Atom Quantum Computers

We design quantum circuits by using the standard cell approach borrowed from classical circuit design, which can speed up the layout of circuits with a regular structure. Our standard cells are general and can be used for all types of quantum circuits: error-corrected or not. The standard cell approach enables the formulation of layout-aware routing algorithms. Our method is directly applicable to neutral atom quantum computers supporting qubit shuttling. Such computers enable zoned architectures for memory, processing and measurement, and we design circuits using qubit storage (memory and measurement zones) and standard cells (processing zones). Herein, we use cubic standard cells for Toffoli gates and, starting from a 3D architecture, we design a multiplication circuit. We present evidence that, when compared with automatic routing methods, our layout-aware routers are significantly faster and achieve shallower 3D circuits (by at least 2.5×), while also reducing routing costs. Additionally, our co-design approach can be used to estimate the resources necessary for a quantum computation without using complex compilation methods. We conclude that standard cells, with the support of layout-aware routing, pave the way to very-large-scale methods for quantum circuit compilation.

Deep imaging of LepR+ stromal cells in optically cleared murine bone hemisections

Tissue clearing combined with high-resolution confocal imaging is a cutting-edge approach for dissecting the three-dimensional (3D) architecture of tissues and deciphering cellular spatial interactions under physiological and pathological conditions. Deciphering the spatial interaction of leptin receptor-expressing (LepR+) stromal cells with other compartments in the bone marrow is crucial for a deeper understanding of the stem cell niche and the skeletal tissue. In this study, we introduce an optimized protocol for the 3D analysis of skeletal tissues, enabling the visualization of hematopoietic and stromal cells, especially LepR+ stromal cells, within optically cleared bone hemisections. Our method preserves the 3D tissue architecture and is extendable to other hematopoietic sites such as calvaria and vertebrae. The protocol entails tissue fixation, decalcification, and cryosectioning to reveal the marrow cavity. Completed within approximately 12 days, this process yields highly transparent tissues that maintain genetically encoded or antibody-stained fluorescent signals. The bone hemisections are compatible with diverse antibody labeling strategies. Confocal microscopy of these transparent samples allows for qualitative and quantitative image analysis using Aivia or Bitplane Imaris software, assessing a spectrum of parameters. With proper storage, the fluorescent signal in the stained and cleared bone hemisections remains intact for at least 2–3 months. This protocol is robust, straightforward to implement, and highly reproducible, offering a valuable tool for tissue architecture and cellular interaction studies.

Two- and Three-Dimensional In Vitro Models of Parkinson's and Alzheimer's Diseases: State-of-the-Art and Applications.

In vitro models play a pivotal role in advancing our understanding of neurodegenerative diseases (NDs) such as Parkinson's and Alzheimer's disease (PD and AD). Traditionally, 2D cell cultures have been instrumental in elucidating the cellular mechanisms underlying these diseases. Cultured cells derived from patients or animal models provide valuable insights into the pathological processes at the cellular level. However, they often lack the native tissue environment complexity, limiting their ability to fully recapitulate their features. In contrast, 3D models offer a more physiologically relevant platform by mimicking the 3D brain tissue architecture. These models can incorporate multiple cell types, including neurons, astrocytes, and microglia, creating a microenvironment that closely resembles the brain's complexity. Bioengineering approaches allow researchers to better replicate cell-cell interactions, neuronal connectivity, and disease-related phenotypes. Both 2D and 3D models have their advantages and limitations. While 2D cultures provide simplicity and scalability for high-throughput screening and basic processes, 3D models offer enhanced physiological relevance and better replicate disease phenotypes. Integrating findings from both model systems can provide a better understanding of NDs, ultimately aiding in the development of novel therapeutic strategies. Here, we review existing 2D and 3D in vitro models for the study of PD and AD.

Memory engram synapse 3D molecular architecture visualized by cryoCLEM-guided cryoET.

Memory is incorporated into the brain as physicochemical changes to engram cells. These are neuronal populations that form complex neuroanatomical circuits, are modified by experiences to store information, and allow for memory recall. At the molecular level, learning modifies synaptic communication to rewire engram circuits, a mechanism known as synaptic plasticity. However, despite its functional role on memory formation, the 3D molecular architecture of synapses within engram circuits is unknown. Here, we demonstrate the use of engram labelling technology and cryogenic correlated light and electron microscopy (cryoCLEM)-guided cryogenic electron tomography (cryoET) to visualize the in-tissue 3D molecular architecture of engram synapses of a contextual fear memory within the CA1 region of the mouse hippocampus. Engram cells exhibited structural diversity of macromolecular constituents and organelles in both pre- and postsynaptic compartments and within the synaptic cleft, including in clusters of membrane proteins, synaptic vesicle occupancy, and F-actin copy number. This "engram to tomogram" approach, harnessing in vivo functional neuroscience and structural biology, provides a methodological framework for testing fundamental molecular plasticity mechanisms within engram circuits during memory encoding, storage and recall.

Deep learning MRI models for the differential diagnosis of tumefactive demyelination versus IDH-wildtype glioblastoma.

Diagnosis of tumefactive demyelination can be challenging. The diagnosis of indeterminate brain lesions on MRI often requires tissue confirmation via brain biopsy. Noninvasive methods for accurate diagnosis of tumor and non-tumor etiologies allows for tailored therapy, optimal tumor control, and a reduced risk of iatrogenic morbidity and mortality. Tumefactive demyelination has imaging features that mimic isocitrate dehydrogenase-wildtype glioblastoma (IDHwt GBM). We hypothesized that deep learning applied to postcontrast T1-weighted (T1C) and T2-weighted (T2) MRI images can discriminate tumefactive demyelination from IDHwt GBM. Patients with tumefactive demyelination (n=144) and IDHwt GBM (n=455) were identified by clinical registries. A 3D DenseNet121 architecture was used to develop models to differentiate tumefactive demyelination and IDHwt GBM using both T1C and T2 MRI images, as well as only T1C and only T2 images. A three-stage design was used: (i) model development and internal validation via five-fold cross validation using a sex-, age-, and MRI technology-matched set of tumefactive demyelination and IDHwt GBM, (ii) validation of model specificity on independent IDHwt GBM, and (iii) prospective validation on tumefactive demyelination and IDHwt GBM. Stratified AUCs were used to evaluate model performance stratified by sex, age at diagnosis, MRI scanner strength, and MRI acquisition. The deep learning model developed using both T1C and T2 images had a prospective validation area under the receiver operator characteristic curve (AUC) of 88% (95% CI: 0.82 - 0.95). In the prospective validation stage, a model score threshold of 0.28 resulted in 91% sensitivity of correctly classifying tumefactive demyelination and 80% specificity (correctly classifying IDHwt GBM). Stratified AUCs demonstrated that model performance may be improved if thresholds were chosen stratified by age and MRI acquisition. MRI images can provide the basis for applying deep learning models to aid in the differential diagnosis of brain lesions. Further validation is needed to evaluate how well the model generalizes across institutions, patient populations, and technology, and to evaluate optimal thresholds for classification. Next steps also should incorporate additional tumor etiologies such as CNS lymphoma and brain metastases. AUC = area under the receiver operator characteristic curve; CNS = central nervous system; CNSIDD = central nervous system inflammatory demyelinating disease; FeTS = federated tumor segmentation; GBM = glioblastoma; IDHwt = isocitrate dehydrogenase wildtype; IHC = immunohistochemistry; MOGAD = myelin oligodendrocyte glycoprotein antibody associated disorder; MS = multiple sclerosis; NMOSD = neuromyelitis optica spectrum disorder; wt = wildtype.

Insights into the MOF-Based Classic Configuration for the Differences in Effective Dye Adsorption, Magnetic Properties, and Computational Analyses.

Two 3D/2D anionic metal-organic frameworks (MOFs), [Cu(HL)]n (1) and [Mn3(L)2(DMF)4]n (2) (DMF = N,N-dimethylformamide), were synthesized by the solvothermal reaction of metal salts and 5'-(4-carboxyphenyl)-2',4',6'-triethyl-[1,1':3',1″-terphenyl]-4,4″-dicarboxylic acid (H3L). Single-crystal X-ray diffraction analyses revealed that complex 1 shows three-dimensional (3D) frameworks with a (3,6)-connected 3-fold interpenetrated topology with the Schläfli symbols of {4.62}2{42.610.83}, whereas the topology of the two-dimensional (2D) architecture can be defined as 2-fold stacked layers with the Schläfli symbols of {43}2{46.66.83} for complex 2. In addition, density functional theory calculations, together with UV-vis adsorption spectroscopy, zeta potential, effective aperture size analysis, TEM, and SEM, were also performed to determine the accurate adsorption sites and significant differences in dye adsorption for complexes 1 and 2. Interestingly, UV-vis studies confirm that Mn-MOF displays remarkable adsorption efficiency for cationic rhodamine B, methylene blue, malachite green, and methyl green, and the removal rate reached 95.2, 95.0, 87.0, and 78.0%, respectively, while almost no adsorption capacity was detected for anionic cresol red and methyl orange. However, Cu-MOF failed to efficiently adsorb any selected dyes. Moreover, the magnetic properties were also investigated through experimental and theoretical calculations in detail, which revealed the weak and stronger antiferromagnetic interactions that occurred between Cu(II) and Mn(II) centers, respectively. Finally, this work provides the profound mechanisms for magnetism and dye adsorption.

Construction of Double-layered DNA Tiles and Arrays from Double Crossover Motifs.

DNA double crossover (DX) motifs including DAE (double crossover, antiparallel, even spacing) and DAO (double crossover, antiparallel, odd spacing) are well-known monolayered DNA building blocks for construction of 2D DNA arrays and tubes in nanoscale and microscale. Compared to the 3D architectures of DNA origami and single-stranded DNA bricks to build nanoscale 3D bundles, tessellations, gears, castles, etc., designs of double- and multi-layers of DX motifs for 3D architectures are still limited. Herein, we report two types of double-layered tiles derived from DAE motifs with single-stranded circular 42- and 64-nt oligonucleotides as scaffold strands. Further tiling of the tiles generated planar 3D crystalline domains and curved tubes, correspondingly. Finally, we applied the chiral index theory to derive the unit tube parameters of six E-tiling (inter-tile distance of even spacing) tubes and analyzed the causation of difference between these tubes.

Widespread 3D genome reorganization precedes programmed DNA rearrangement in Oxytricha trifallax.

Genome organization recapitulates function, yet ciliates like Oxytricha trifallax possess highly-specialized germline genomes, which are largely transcriptionally silent. During post-zygotic development, Oxytricha 's germline undergoes large-scale genome editing, rearranging precursor genome elements into a transcriptionally-active genome with thousands of gene-sized nanochromosomes. Transgenerationally-inherited RNAs, derived from the parental somatic genome, program the retention and reordering of germline fragments. Retained and eliminated DNA must be distinguished and processed separately, but the role of chromatin organization in this process is unknown. We developed tools for studying Oxytricha nuclei and apply them to map the 3D organization of precursor and developmental states using Hi-C. We find that the precursor conformation primes the germline for development, while a massive spatial reorganization during development differentiates retained from eliminated regions before DNA rearrangement. Further experiments suggest a role for RNA-DNA interactions and chromatin remodeling in this process, implying a critical role for 3D architecture in programmed genome rearrangement.

Mapping the 3D genome architecture.

The spatial organization of the genome plays a critical role in regulating gene expression, cellular differentiation, and genome stability. This review provides an in-depth examination of the methodologies, computational tools, and frameworks developed to map the three-dimensional (3D) architecture of the genome, focusing on both ligation-based and ligation-free techniques. We also explore the limitations of these methods, including biases introduced by restriction enzyme digestion and ligation inefficiencies, and compare them to more recent ligation-free approaches such as Genome Architecture Mapping (GAM) and Split-Pool Recognition of Interactions by Tag Extension (SPRITE). These techniques offer unique insights into higher-order chromatin structures by bypassing ligation steps, thus enabling the capture of complex multi-way interactions that are often challenging to resolve with traditional methods. Furthermore, we discuss the integration of chromatin interaction data with other genomic layers through multimodal approaches, including recent advances in single-cell technologies like sci-HiC and scSPRITE, which help unravel the heterogeneity of chromatin architecture in development and disease.

Multiscale Three-Dimensional Evaluation and Analysis of Murine Lung Vasculature From Macro- to Micro-Structural Level.

The pulmonary vasculature plays a pivotal role in the development and progress of chronic lung diseases. Due to limitations of conventional two-dimensional histological methods, the complexity and the detailed anatomy of the lung blood circulation might be overlooked. In this study, we demonstrate the practical use of optical serial block face imaging (SBFI), ex vivo microcomputed tomography (micro-CT), and nondestructive optical tomography for visualization and quantification of the pulmonary circulation's 3D architecture from macro- to micro-structural levels in murine lung samples. We demonstrate that SBFI can provide rapid, cost-effective, and label-free visualization of mouse lung macrostructures and large pulmonary vessels. Micro-CT offers high-resolution imaging and captures microvascular and (pre)capillary structures, with microstructural quantification. Optical microscopy techniques such as optical projection tomography (OPT) and light sheet fluorescence microscopy, allows noninvasive, mesoscopic imaging of optically cleared mouse lungs, still enabling 3D microscopic reconstruction down to the precapillary level. By integrating SBFI, micro-CT, and nondestructive optical microscopy, we provide a framework for detailed and 3D understanding of the pulmonary circulation, with emphasis on vascular pruning and rarefaction. Our study showcases the applicability and complementarity of these techniques for organ-level vascular imaging, offering researchers flexibility in selecting the optimal approach based on their specific requirements. In conclusion, we propose 3D-directed approaches for a detailed whole-organ view on the pulmonary vasculature in health and disease, to advance our current knowledge of diseases affecting the pulmonary vasculature.

Charge noise in low Schottky barrier multilayer tellurium field-effect transistors.

Creating van der Waals (vdW) homojunction devices requires materials with narrow bandgaps and high carrier mobilities for bipolar transport, which are crucial for constructing fundamental building blocks like diodes and transistors in a 2D architecture. Following the recent discovery of elemental 2D tellurium, here, we systematically investigate the electrical transport and flicker noise of hydrothermally grown multilayer tellurium field effect transistors. While the devices exhibit a dominant p-type behavior with high hole mobilities up to ∼242 cm2 V-1 s-1 at room temperature and almost linear current-voltage characteristics down to 77 K, ambipolar behavior was observed in some cases. Through a detailed temperature dependent transport characterization, we estimated the Schottky barrier height as low as ∼20 meV. The good electrical contacts further facilitate the observation of metal-to-insulator transition at low temperature, being an intrinsic property of the tellurium channel rather than the contacts. Finally, detailed low frequency noise spectroscopy shows dominant 1/f type behavior across the entire gate-voltage range. The origin of the observed noise can be described by Hooge's mobility fluctuation model, rather than the carrier number fluctuations due to interfacial traps. We anticipate that such analysis will contribute to the development of futuristic low-noise devices using tellurium.

Recent advances of versatile fluorophores for multifunctional biomedical imaging in the NIR-II region.

Fluorescence imaging in the second near-infrared region (NIR-II, 1000-1700 nm) enables high-resolution visualization of deep-tissue biological architecture and physiopathological events, due to the reduced light absorption, scattering and tissue autofluorescence. Numerous versatile NIR-II fluorescent probes have been reported over the past decades. In this review, we first provide a detailed account of the advantages of fluorescence imaging in the NIR-II region. Following this, the classification, design and performance optimization strategies of NIR-II fluorescent probes are systematically discussed, along with a broad range of biomedical applications in vivo. Finally, the discussion extends to the next generation of fluorescent probes for in vivo imaging and the challenges and perspectives for the clinical translation of fluorescence imaging technology in the NIR-II region.

Systematic Topology Synthesis and Power Density Visualization of Partial Power Processing Architecture

Systematic Topology Synthesis and Power Density Visualization of Partial Power Processing Architecture

Membrana preformativa: Unveiling the unexplored facets of dental development.

Odontogenesis is a complex and highly regulated biological process that involves a range of molecular mechanisms. Among these, Ki67 and Cyclin D1 are crucial cell cycle regulators that play pivotal roles in controlling cell proliferation during tooth development. This study aims to provide detailed insights into the expression patterns and functional significance of Ki67 and Cyclin D1 in tooth development. Through rigorous analysis, we seek to elucidate the intricate mechanisms underlying tooth development, helping to advance our understanding of this vital biological process. The procurement of rabbit tooth germ was performed only after obtaining the requisite ethical clearance. Subsequently, the tissues were processed and subjected to Hematoxylin and Eosin staining to facilitate enhanced visualization of the overall tissue architecture and organization & immunohistochemical staining of Ki-67 and Cyclin D1 was performed. In the tooth germ, cyclin D1 demonstrated intense staining in the dental papilla, especially in the membrana preformativa, with this intensity decreasing following predentin formation. Odontoblasts showed mild staining as they transitioned from pre-odontoblasts, which further diminished after dentin formation. Both the dental papilla and differentiating odontoblasts were positive for Ki67, though Ki67 staining in the odontoblasts reduced after dentin formation. In conclusion, the membrana preformativa plays a key role in odontogenesis, as indicated by its involvement in cellular proliferation and differentiation during tooth development.

3D neural architecture search to optimize segmentation of plant parts

3D neural architecture search to optimize segmentation of plant parts

Rapid prediction of conformationally-dependent DFT-level descriptors using graph neural networks for carboxylic acids and alkyl amines.

Data-driven reaction discovery and development is a growing field that relies on the use of molecular descriptors to capture key information about substrates, ligands, and targets. Broad adaptation of this strategy is hindered by the associated computational cost of descriptor calculation, especially when considering conformational flexibility. Descriptor libraries can be precomputed agnostic of application to reduce the computational burden of data-driven reaction development. However, as one often applies these models to evaluate novel hypothetical structures, it would be ideal to predict the descriptors of compounds on-the-fly. Herein, we report DFT-level descriptor libraries for conformational ensembles of 8528 carboxylic acids and 8172 alkyl amines towards this goal. Employing 2D and 3D graph neural network architectures trained on these libraries culminated in the development of predictive models for molecule-level descriptors, as well as the bond- and atom-level descriptors for the conserved reactive site (carboxylic acid or amine). The predictions were confirmed to be robust for an external validation set of medicinally-relevant carboxylic acids and alkyl amines. Additionally, a retrospective study correlating the rate of amide coupling reactions demonstrated the suitability of the predicted DFT-level descriptors for downstream applications. Ultimately, these models enable high-fidelity predictions for a vast number of potential substrates, greatly increasing accessibility to the field of data-driven reaction development.