3D Geometry Research Articles

A mathematical model for the interaction of anisotropic turbulence with porous surfaces

Leading-edge noise is a complex phenomenon that occurs when a turbulent fluid encounters a solid object, and is a notable concern in various engineering applications. This study enhances a mathematical leading-edge noise model (Hales et al., J. Fluid Mech., vol. 970, 2023, A29) for anisotropic flow and porous boundaries. The model has two key components. First, we adjust the velocity spectrum to account for the possibility of anisotropy in the flow. This paper rigorously introduces a third dimension for the turbulence spectrum that preserves the turbulence kinetic energy and mathematical definitions for integral length scales. Second, we adapt the fully analytical acoustic transfer function to account for different boundaries by implementing convective impedance boundary conditions when formulating the gust-diffraction problem. This problem is then solved using the Wiener–Hopf technique. We discuss important aspects of this method, including the factorisation of a non-trivial scalar kernel function and the application of suitable edge conditions for the problem. Each modification is inspired by experimental leading-edge noise data using a series of different porous leading edges and anisotropic turbulence generated by a cylinder upstream of the edge. Experimental data demonstrate the interplay between anisotropy and leading-edge modifications while achieving the characteristic mid-frequency noise reduction expected from porous leading edges. Our model is adapted to best fit the trends of the data via a tailored impedance function, leading to good agreement with all datasets across an extended frequency range. This tailored function is used to successfully validate the model against other datasets from a different set of experiments.

Development of Thermosensitive Hydrogels with Tailor-Made Geometries to Modulate Cell Harvesting of Non-Flat Cell Cultures

Considering the complexity in terms of design that characterizes the different tissues of the human body, it is necessary to study and develop more precise therapies. In this sense, this article presents the possibility of fabricating photocurable thermosensitive hydrogels with free geometry and based on N-Vinyl Caprolactam (VCL) with the aim of modulating the adhesion of non-planar cell cultures. The fabrication process is based on the use as a mold of two-layer thick water-soluble polyvinyl alcohol (PVA) previously printed by Extrusion Material (MatEx). From this technology it has been possible to obtain hydrogels with different 3D geometries and different crosslinking percentages (2, 4 and 6 mol%). Studies have shown that networks reduce their thermosensitivity not only when the percentage of crosslinking in the formulation increases, but also when the thickness of the hydrogel obtained increases. Based on this reduction in thermosensitivity, the less crosslinked (2 mol%) hydrogels have been evaluated to carry out a novel direct application in which hydrogels with curved geometry have allowed cell adhesion and proliferation at 37 °C with the endothelial cell line C166-GFP; likewise, non-aggressive cell detachment was observed when the hydrogel temperature was reduced to values of 20 °C. Therefore, the present manuscript shows a novel application for the synthesis of free-form thermosensitive hydrogels that allows modulation of non-planar cell cultures.

3D simulations of the RF wave propagation and power coupling in cold magnetized plasma near an ICRF antenna

Abstract The electromagnetic wave propagations and their coupling characteristics in magnetized plasma near the antenna of Ion Cyclotron Range of Frequencies (ICRF) is studied based on self-developed 3DFEM-IA code. This code effectively resolves the three-dimensional (3D) geometry and the electromagnetic field using the finite element method. Our findings reveal that the distributions of electromagnetic fields and energy flow density significantly depend on the antenna phases, surface current density on the antenna straps, and background plasma density. Notably, the non-uniform surface current density on the antenna straps, resulting from the presence of induced currents, contributes to a reduction in coupling power within the edge plasma. Furthermore, the calculated coupling impedance increases with plasma density, corroborating well with experimental measurements.

4D Printing of Liquid Crystal Emulsions for Smart Structures with Multiple Functionalities

Abstract3D printing, and more recently 4D printing, has emerged as a transformative technology for fabricating structures with complex geometries and responsive properties. However, employing functional colloidal solutions as inks for printing remains unexplored. In this work, we present a novel and versatile 4D printing approach for fabricating functional and complex‐shaped objects using polymerizable liquid crystal (LC) emulsion droplets. Leveraging a digital light processing (DLP) 3D printing technique, we achieve rapid production of intricate 3D geometries with high resolution. The printed structures retain the LC ordering from the precursor droplets, imparting the final objects with shape memory properties, including shape fixation and recovery upon heating or light exposure. Light‐responsive behavior is introduced post‐printing by embedding an azo dye into the 3D structures. Additionally, we explore the potential to create intrinsically porous 3D structures by selectively removing non‐reactive components from the printed geometries, adding an extra level of functionality to the printed objects. Furthermore, we incorporate chiral nematic LCs into the emulsion droplets, producing 3D objects with tunable reflective properties. To our knowledge, this is the first example of DLP 3D printing with emulsions, offering an effective and versatile pathway for developing 4D‐printed materials with potential applications in optics, robotics, microfluidics, and biomedicine.

Periodic solutions, breather, lump and interaction solutions of a generalized (2+1)-dimensional Hirota bilinear equation via the bilinear neural network method

The main focus of this paper is to address a generalized (2+1)-dimensional Hirota bilinear equation utilizing the bilinear neural network method. The paper presents the periodic solutions through a single-layer model of [3-4-1], followed by breather, lump and their interaction solutions by using double-layer models of [3-3-2-1] and [3-3-3-1], respectively. A significant innovation introduced in this work is the computation of periodic cross-rational solutions through the design of a novel [3-(2+2)-4-1] model, where a specific hidden layer is partitioned into two segments for subsequent operations. Three-dimensional and density figures of the solutions are given alongside an analysis of the dynamics of these solutions.

NeCA: 3D Coronary Artery Tree Reconstruction from Two 2D Projections via Neural Implicit Representation

Cardiovascular diseases (CVDs) are the most common health threats worldwide. 2D X-ray invasive coronary angiography (ICA) remains the most widely adopted imaging modality for CVD assessment during real-time cardiac interventions. However, it is often difficult for the cardiologists to interpret the 3D geometry of coronary vessels based on 2D planes. Moreover, due to the radiation limit, often only two angiographic projections are acquired, providing limited information of the vessel geometry and necessitating 3D coronary tree reconstruction based only on two ICA projections. In this paper, we propose a self-supervised deep learning method called NeCA, which is based on neural implicit representation using the multiresolution hash encoder and differentiable cone-beam forward projector layer, in order to achieve 3D coronary artery tree reconstruction from two 2D projections. We validate our method using six different metrics on a dataset generated from coronary computed tomography angiography of right coronary artery and left anterior descending artery. The evaluation results demonstrate that our NeCA method, without requiring 3D ground truth for supervision or large datasets for training, achieves promising performance in both vessel topology and branch-connectivity preservation compared to the supervised deep learning model.

Displaced volume versus submerged volume in the floating of a solid

Abstract This paper describes a new theoretical approach accompanied by experiments that demonstrate that the volume of the displaced fluid cannot be used to calculate the magnitude of the buoyant force when a solid object is floated in a container of only slightly larger cross-section than the object. It is important to emphasize that the discussions conducted here are accessible to high school and first-year university physics students because the concept of hydrostatic pressure and easily understood geometric considerations are applied.

Data-driven reduced order surrogate modeling for coronary in-stent restenosis

Background:The intricate process of coronary in-stent restenosis (ISR) involves the interplay between different mediators, including platelet-derived growth factor, transforming growth factor-β, extracellular matrix, smooth muscle cells, endothelial cells, and drug elution from the stent. Modeling such complex multiphysics phenomena demands extensive computational resources and time. Methods:This paper proposes a novel non-intrusive data-driven reduced order modeling approach for the underlying multiphysics time-dependent parametrized problem. In the offline phase, a 3D convolutional autoencoder, comprising an encoder and decoder, is trained to achieve dimensionality reduction. The encoder condenses the full-order solution into a lower-dimensional latent space, while the decoder facilitates the reconstruction of the full solution from the latent space. To deal with the 5D input datasets (3D geometry + time series + multiple output channels), two ingredients are explored. The first approach incorporates time as an additional parameter and applies 3D convolution on individual time steps, encoding a distinct latent variable for each parameter instance within each time step. The second approach reshapes the 3D geometry into a 2D plane along a less interactive axis and stacks all time steps in the third direction for each parameter instance. This rearrangement generates a larger and complete dataset for one parameter instance, resulting in a singular latent variable across the entire discrete time-series. In both approaches, the multiple outputs are considered automatically in the convolutions. Moreover, Gaussian process regression is applied to establish correlations between the latent variable and the input parameter. Results:The constitutive model reveals a significant acceleration in neointimal growth between 30−60 days post percutaneous coronary intervention (PCI). The surrogate models applying both approaches exhibit high accuracy in pointwise error, with the first approach showcasing smaller errors across the entire evaluation period for all outputs. The parameter study on drug dosage against ISR rates provides noteworthy insights of neointimal growth, where the nonlinear dependence of ISR rates on the peak drug flux exhibits intriguing periodic patterns. Applying the trained model, the rate of ISR is effectively evaluated, and the optimal parameter range for drug dosage is identified. Conclusion:The demonstrated non-intrusive reduced order surrogate model proves to be a powerful tool for predicting ISR outcomes. Moreover, the proposed method lays the foundation for real-time simulations and optimization of PCI parameters.

Modeling condensation of Methane-Water vapor mixtures in the Hedbäck Nozzle: A comparative study of 2D and 3D geometries

This study investigated the condensation behavior of a binary mixture of methane (CH4) and water vapor (H2O) within the Hedbäck nozzle, employing both 2D and 3D modeling approaches. The 2D model was initially developed to analyze the effects of inlet parameters and mixture’s condensation characteristics. However, the inherent limitations of the 2D model in capturing the full complexity of the flow dynamics prompted a transition to a 3D modeling approach. The research explored two distinct 3D geometries: one created by extruding the 2D section and another by rotating the 2D cross-section around its axis. The latter, which more accurately represented real-world nozzle conditions, revealed differences in flow patterns, condensation behavior, and entropy generation, emphasizing the crucial role of 3D effects in the condensation process. The sensitivity of the numerical simulation results to various nucleation theories and droplet growth models was acknowledged, and the most appropriate theory and model were meticulously selected. This approach minimized discrepancies between numerical modeling outcomes and experimental data, enhancing the accuracy and reliability of the findings. The 2D results indicate a 40.5% increase in the inlet partial pressure with a constant temperature and mass fraction, resulting in a 56% rise in maximum liquid mass fraction. Similarly, a 3.6% decrease in inlet temperature under constant pressure and mass fraction increases the average radius of outlet droplets by more than 13%. In addition, entropy generation analysis revealed that less than 1% of entropy generation is attributed to mean flow, whereas 99% is linked to turbulent fluctuations. Moreover, the 3D models demonstrated markedly different entropy generation rates compared to the 2D model, highlighting the importance of 3D approach for a comprehensive understanding of thermodynamic irreversibility within the system.

Development Study of Direct Damage Stability Algorithm Based on 3D Geometry Data Processing for Onboard Loading Instrument

Abstract The maritime world experienced a very significant development such as the process of cargo planning is the existence of a system called loading instrument which is regulated by several regulations from both Classification and IACS. Damage stability calculation is also very important for loading instruments. Currently in Indonesia, there is no known algorithm to calculate direct damage stability from 3D models. And also there is still no loading instrument in Indonesia that applies the algorithm. Therefore, this research aims to develop a direct damage stability calculation algorithm for the development of loading instruments in Indonesia. Direct damage stability uses the lost buoyancy method which is calculated from 3D geometry model directly. The calculation process starts by dividing the ship geometry and compartment with the desired station. In general, the numerical integration used in the calculation process is the Simpson method. Then Newton’s iteration is used to find an equilibrium condition by predetermined conditions. Finally, the calculation results will be compared with the cases of LR Rules and IACS Rules UR L5 as a form of validation. The calculation algorithm can later be applied to the onboard loading software computer needed by the ship for the loading process with damage applications.

Constraining the Circumstellar Medium Structure and Progenitor Mass-loss History of Interacting Supernovae Through 3D Hydrodynamic Modeling: The Case of SN 2014C

We investigate SN 2014C using three-dimensional (3D) hydrodynamic modeling, focusing on its early interaction with a dense circumstellar medium (CSM). Our objective is to uncover the pre-supernova (SN) CSM structure and constrain the progenitor star’s mass-loss history prior to core collapse. Our comprehensive model traces the evolution from the progenitor star through the SN event and into the SN remnant phase. We simulate the remnant’s expansion over approximately 15 yr, incorporating a CSM derived from the progenitor star’s outflows through dedicated hydrodynamic simulations. Analysis reveals that the remnant interacted with a dense toroidal nebula extending from 4.3 × 1016 to 1.5 × 1017 cm in the equatorial plane, with a thickness of approximately 1.2 × 1017 cm. The nebula’s density peaks at approximately 3 × 106 cm−3 at the inner boundary, gradually decreasing as ≈r −2 at greater distances. This nebula formed due to intense mass loss from the progenitor star between approximately 5000 and 1000 yr before collapse. During this period, the maximum mass-loss rate reached about 8 × 10−4 M ⊙ yr−1, ejecting ≈2.5 M ⊙ of stellar material into the CSM. Our model accurately reproduces Chandra and NuSTAR spectra, including the iron (Fe) K line, throughout the remnant’s evolution. Notably, the Fe line is self-consistently reproduced, originating from shocked ejecta, with ≈0.05 M ⊙ of pure-Fe ejecta shocked during the remnant–nebula interaction. These findings suggest that the 3D geometry and density distribution of the CSM, as well as the progenitor star’s mass-loss history, align with a scenario where the star was stripped through binary interaction, specifically common-envelope evolution.

Novelle approach to simulating spinal cord stimulation during tSCS using CT images and FEM

Abstract Transcutaneous spinal cord stimulation (tSCS) offers non-invasive relief for chronic pain and improves motor function in spinal cord injured (SCI) patients. However, its mechanisms are not currently fully understood, and patientspecific factors, such as Body mass index (BMI) and age, complicate treatments. This paper aims to understand tSCS better by developing a novel Finite Element Model (FEM) of the human body using CT scans. Three subjects (sex, male, female, male. Age: 26, 27, 64. BMI: 38.9, 24.1, 28.4) underwent a CT scan, performed on a Cannon Aquilion Prime (Slice thickness [mm]: 0.8, Voxel size [mm3]: 0.564, 0.328, 0.527), which imaged the trunk of the body, from top of the abdomen to the bottom of the pelvis. The images were then used in Materialize Mimics Research 21.0 to create 3D images of individual organs, skin, fat, muscles, skeleton, and spinal cord. After pre-processing in Autodesk Meshmixer, the models were converted into solid CAD objects in Ansys SpaceClaim R2021 and combined into a single abdominal model. Ansys Maxwell R2021 was then used for simulations. Five different two-electrode configurations were tested in the prototype phase with a simplified model setup. The positive electrodes were placed over the (Thoracic) T10, T12, (Lumbar) L2 and L4 vertebrae sequentially, with the negative electrode over (Sacral) S2. The simulations took on average 47.4 hours. A marked decline in electrical current penetration depth was observed as the electrodes were placed closer together which was consistent with known current distribution patterns. A preliminary validation test was also performed using a lamb’s thigh. Two electrodes were placed a known distance apart and a stimulation was given, needle electrodes were then inserted in a grid-like pattern to obtain voltage values. The setup was recreated and simulated in Ansys Maxwell. The resulting average percent difference was 44.93 ± 33.3 % (5Vpp Square wave) and 35.02 ± 23.38 % (5V DC). In both instances, the highest difference was at the edges of the electrodes and the lowest difference in the midpoint between electrodes. Later versions of FEM models incorporated more organs and had improved on previous mesh generation flaws, but encountered new mesh generation errors which could not be rectified before the conclusion of the master’s project. Despite complications, this project has provided a pipeline for creating similar models and shown their usability. Future work will involve overcoming the current mesh generation errors, reducing calculation time, and performing a thorough validation test.

SSR-2D: Semantic 3D Scene Reconstruction From 2D Images.

Most deep learning approaches to comprehensive semantic modeling of 3D indoor spaces require costly dense annotations in the 3D domain. In this work, we explore a central 3D scene modeling task, namely, semantic scene reconstruction without using any 3D annotations. The key idea of our approach is to design a trainable model that employs both incomplete 3D reconstructions and their corresponding source RGB-D images, fusing cross-domain features into volumetric embeddings to predict complete 3D geometry, color, and semantics with only 2D labeling which can be either manual or machine-generated. Our key technical innovation is to leverage differentiable rendering of color and semantics to bridge 2D observations and unknown 3D space, using the observed RGB images and 2D semantics as supervision, respectively. We additionally develop a learning pipeline and corresponding method to enable learning from imperfect predicted 2D labels, which could be additionally acquired by synthesizing in an augmented set of virtual training views complementing the original real captures, enabling more efficient self-supervision loop for semantics. As a result, our end-to-end trainable solution jointly addresses geometry completion, colorization, and semantic mapping from limited RGB-D images, without relying on any 3D ground-truth information. Our method achieves state-of-the-art performance of semantic scene completion on two large-scale benchmark datasets MatterPort3D and ScanNet, surpasses baselines even with costly 3D annotations in predicting both geometry and semantics. To our knowledge, our method is also the first 2D-driven method addressing completion and semantic segmentation of real-world 3D scans simultaneously.

3D City Modeling for Flood Inundation Analysis in Mayjen Sungkono Area, Surabaya

Abstract Floods in urban areas are primarily caused by high rainfall combined with inadequate drainage systems, inappropriate land use changes, and insufficient flood mitigation measures. Surabaya, characterized as a seasonal zone with periodic rainfall, frequently experiences flood inundation due to the loss of water retention areas and an inadequate drainage infrastructure. Three-dimensional (3D) city models provide a detailed representation of urban environments, capturing the three-dimensional geometry of objects and general urban structures. This 3D modeling approach can enhance the understanding of flood depths impacting surrounding buildings and infrastructure and assist in estimating the effects on drainage capacity and catchment areas for long-term planning. In this research, 3D flood inundation modeling was conducted using LiDAR data and aerial photography, integrated with other GIS data. The flood inundation simulation was performed using HEC-RAS, while the 3D modeling was carried out using GIS-based software employing semi-automatic methods. The results of the 3D modeling and visualization indicate potential flood inundation depths reaching up to 5 meters, affecting 1,100 buildings and 30 roads. Further research could focus on improving the accuracy of terrain and channel data used in flood modeling to enhance the precision of the results.

PROSTHESIS REPAIR OF ORAL IMPLANTS BASED ON ARTIFICIAL INTELLIGENC`E FINITE ELEMENT ANALYSIS

Dentists often suggest dental implants to replace missing teeth; nevertheless, mechanical issues can develop with these implants, which could lead to prosthesis replacement or repairs. When investigating implant systems' mechanical characteristics and stress distribution, finite element analysis (FEA) is a popular computational tool. In biomechanical investigations, this strategy is widely used. However, traditional FEA methods can be tedious and require expert expertise for accurate simulation and translation of results. To automate and simplify the process of mending oral implant prostheses, the article suggests a new framework called AI-FEA. The three primary parts that make up the suggested AI-FEA framework are 1. An AI-powered model creation module that utilizes data from medical imaging to autonomously construct 3D finite element designs that are unique to each patient. Utilizing deep learning approaches, this module segments and reconstructs three-dimensional geometries from computed tomography (CT) or cone-beam CT data using material properties and boundary conditions. 2. A FEA solver that runs simulations to test the way the implant system handles different loads. This component uses state-of-the-art numerical methods to model the implant and bone interface and determine stress distributions. 3. An AI-based decision support system that takes all that data and recommends the best way to fix the prosthesis. Combining FEA findings with patient-specific variables, this decision support system uses machine learning algorithms educated on an extensive dataset of implant failure instances and repair results to provide the optimal repair strategy. For patients experiencing issues with oral implants, the suggested AI-FEA framework might mean huge time and skill savings in prosthesis repair planning, leading to better, more individualized care.

Life cycle assessment of additively manufactured elastocaloric Ni-Ti heat pump/refrigeration components and the effect of design geometry on the environmental impact categories

Elastocaloric Ni-Ti alloys offer solid-state heat pump and cooling applications. The environmental impact of elastocaloric Ni-Ti parts manufactured via powder bed fusion was assessed using life cycle assessment with the ReCiPe method. These solid-state heat pumps can mitigate the environmental impacts of current air conditioning/refrigeration methods that use harmful chemicals like CFCs, HCFCs, and HFCs. Measurements of electricity, Ar gas, and compressed air consumption were considered within the gate-to-gate boundary, covering powder transportation, printing, and use-phase. Nickel was identified as the most damaging input, reducible through geometry optimization. Optimized geometries (given the same number of printing layers) require more energy during powder bed fusion however, the environmental benefit from material saving outweighs the damage from increased electrical usage by far. The powder bed fusion machine's standby mode consumes more energy and Ar gas than the printing process, presenting an opportunity for mitigation. Additively manufactured Ni-Ti elastocaloric structures were used as heat generators/sinks in flowing water during compression cycles, achieving 11 °C and 10 °C water temperature change per cycle for the solid geometry and the optimized geometry respectively. A 10 x 10 x 55 mm Ni-Ti structure, providing >2 °C temperature span under compressive cycles, served as the functional unit.

Analisis Kemampuan Pemahaman dan Pemecahan Masalah pada Siswa Kelas VIII UPT SMP Negeri 27 Medan Terhadap Materi Bangun Datar dan Ruang

This study is a descriptive study with a qualitative approach that aims to analyze the errors made by students in solving mathematics problems, especially on flat and solid geometry materials using the Newman Theory approach. The subjects in this study were 30 seventh grade students at SMP Negeri 27 Medan. The method of collecting data or information was carried out through a written test of 3 (three) descriptive questions and interviews with selected subjects. The information obtained was then analyzed based on the Newman analysis procedure which is divided into 5 (five) stages of errors, namely reading, understanding, transformation, process skills, and writing answers. The results of the analysis showed that the most significant errors occurred at the transformation stage, with a percentage of 83.3%, categorized as very high. On the other hand, errors in process skills were classified as very low, namely 16.6%. This finding shows the importance of developing more effective learning methods to help students understand and solve mathematics problems, especially those related to geometric concepts.

Improved weighted sum estimation of total radiated power at W7-X

AbstractAs magnetic confinement devices move toward higher fusion powers, moderating the heat load to the plasma-facing components becomes increasingly challenging. Efficient power dissipation can be achieved through control of the plasma radiation. However, defining a reliable proxy for the total radiated power is particularly challenging for non-axisymmetric devices such as stellarators. To address this problem, the radiated power can be estimated through a sum of the individual line-integrated bolometer measurements with weights properly calculated to account for the three-dimensional magnetic geometry. The present contribution aims to apply this weighted sum approach to Wendelstein 7-X (W7-X) and quantitatively validate it. First, we generate synthetic radiated power phantoms with characteristic W7-X radiation features to derive a set of optimized line-of-sight weights. Then, we test the weights on mock-ups and EMC3-EIRENE radiation patterns, including acquisition and analysis errors such as random noise fluctuations, camera misalignments, and field errors. Compared to other methods, the optimized weighted sum technique exhibited the best performance in all the presented synthetic test cases. When applied to experimental bolometer data, the optimized weights provided a proxy that is both reliable and real-time capable. Further validation is foreseen for the next experimental campaign.

Comparison of stable carotid plaques in patients with mild-to-moderate carotid stenosis with vulnerable plaques in patients with significant carotid stenosis.

To compare the characteristics of stable and vulnerable carotid plaques, and investigate the diagnostic performance of wall shear stress (WSS) based on magnetic resonance plaque imaging in carotid plaques. Retrospectively analyzed and divided 64 atherosclerotic plaques into stable carotid plaque groups with mild-to-moderate stenosis and vulnerable carotid plaque groups with significant stenosis. Computational fluid dynamics simulations were performed to calculate WSS parameters by using three-dimensional wall geometry based on high-resolution magnetic resonance plaque imaging of carotid bifurcation and patient specific boundary conditions obtained through color Doppler ultrasound. WSS parameters including upstream (WSSup), downstream (WSSdown), and core (WSScore) of plaque. The WSS parameters values were compared between the stable and vulnerable carotid plaque groups. Receiver operating characteristic curves and area under the curve (ROC-AUC) and Python were used to evaluate discriminative efficacy of WSS. WSSdown exhibited significant decrease in the vulnerable carotid plaque group (2.88 ± 0.41 Pa) compared to the stable carotid plaque group (4.47 ± 0.84 Pa) (P = .003). The difference of WSSup (3.28 ± 0.85 Pa vs 4.02 ± 0.74 Pa) and WSScore (1.12 ± 0.18 Pa vs 1.38 ± 0.38 Pa) between the two groups were also pronounced (P = .02, 0.01, respectively). The ROC-AUC values for WSSup, WSSdown, WSScore were 0.75 (95% CI, 0.58-0.93), 0.96 (95% CI, 0.79-1.14), 0.69 (95% CI, 0.56-0.83) respectively. When the value of WSSdown was 3.5 Pa, the sensitivity was 93.7% (95% CI, 76.1-111), specificity and accuracy was 87.5% (95% CI, 70.0-105), 88.4% (95% CI, 70.6-105) respectively. Notably, among these parameters, WSSdown demonstrated the highest discriminative efficiency with a F1 Score of 0.90, Diagnostic Odds Ratio of 105.0 and Matthews Correlation Coefficient of 0.81. Vulnerable carotid plaques with significant stenosis have lower WSS compared to stable plaques with mild-to-moderate stenosis, and downstream WSS showing the highest diagnostic efficacy.

Back to geometry: Efficient indoor space segmentation from point clouds by 2D–3D geometry constrains

This paper addresses the challenge of indoor space segmentation from 3D point clouds, which is essential for understanding interior layouts, reconstructing 3D structures, and developing indoor navigation maps. While current deep learning-based methods rely on projecting 3D point clouds into 2D for instance extraction, they often fail to capture the local and global 3D features necessary for effectively segmenting complex indoor spaces, such as multi-ring nested structures. These methods also struggle with generalization across different scenes. In response, this paper proposes an efficient indoor space segmentation method that integrates both 2D and 3D geometric constraints. By leveraging the distribution characteristics of point clouds in 2D and the local and global features in 3D, the method achieves reliable extraction of vertical structural information in complex indoor environments. To address under-segmentation in small spaces due to varying scales, the paper introduces an adaptive extraction method for space partition anchors, guided by local features. During instance-level space segmentation, a hierarchical contour tree structure is employed to precisely partition complex indoor spaces, effectively handling circular and composite structures. The proposed approach was tested on 96 RGB-D scans from the Beike dataset and 6 large-scale indoor scenes from the S3DIS dataset, covering a range of complexities, sizes, and structures. The experimental section includes ablation studies and thorough comparisons with existing state-of-the-art spatial partitioning algorithms based on morphology and deep learning. Results demonstrate that the proposed method significantly outperforms existing approaches in terms of accuracy, robustness, and generalization ability, providing a solid foundation for indoor space modeling and robotic navigation. The source code and datasets will be made publicly available via the “EISPGeo” link.