Deep learning, especially through convolutional neural networks (CNN) such as the U-Net 3D model, has revolutionized fault identification from seismic data, representing a significant leap over traditional methods. Ou...Deep learning, especially through convolutional neural networks (CNN) such as the U-Net 3D model, has revolutionized fault identification from seismic data, representing a significant leap over traditional methods. Our review traces the evolution of CNN, emphasizing the adaptation and capabilities of the U-Net 3D model in automating seismic fault delineation with unprecedented accuracy. We find: 1) The transition from basic neural networks to sophisticated CNN has enabled remarkable advancements in image recognition, which are directly applicable to analyzing seismic data. The U-Net 3D model, with its innovative architecture, exemplifies this progress by providing a method for detailed and accurate fault detection with reduced manual interpretation bias. 2) The U-Net 3D model has demonstrated its superiority over traditional fault identification methods in several key areas: it has enhanced interpretation accuracy, increased operational efficiency, and reduced the subjectivity of manual methods. 3) Despite these achievements, challenges such as the need for effective data preprocessing, acquisition of high-quality annotated datasets, and achieving model generalization across different geological conditions remain. Future research should therefore focus on developing more complex network architectures and innovative training strategies to refine fault identification performance further. Our findings confirm the transformative potential of deep learning, particularly CNN like the U-Net 3D model, in geosciences, advocating for its broader integration to revolutionize geological exploration and seismic analysis.展开更多
The intracontinental subduction of a>200-km-long section of the Tajik-Tarim lithosphere beneath the Pamir Mountains is proposed to explain nearly 30 km of shortening in the Tajik fold-thrust belt and the Pamir upli...The intracontinental subduction of a>200-km-long section of the Tajik-Tarim lithosphere beneath the Pamir Mountains is proposed to explain nearly 30 km of shortening in the Tajik fold-thrust belt and the Pamir uplift.Seismic imaging revealed that the upper slab was scraped and that the lower slab had subducted to a depth of>150 km.These features constitute the tectonic complexity of the Pamirs,as well as the thermal subduction mechanism involved,which remains poorly understood.Hence,in this study,high-resolution three-dimensional(3D)kinematic modeling is applied to investigate the thermal structure and geometry of the subducting slab beneath the Pamirs.The modeled slab configuration reveals distinct along-strike variations,with a steeply dipping slab beneath the southern Pamirs,a more gently inclined slab beneath the northern Pamirs,and apparent upper slab termination at shallow depths beneath the Pamirs.The thermal field reveals a cold slab core after delamination,with temperatures ranging from 400℃to 800℃,enveloped by a hotter mantle reaching~1400℃.The occurrence of intermediate-depth earthquakes aligns primarily with colder slab regions,particularly near the slab tear-off below the southwestern Pamirs,indicating a strong correlation between slab temperature and seismicity.In contrast,the northern Pamirs exhibit reduced seismicity at depth,which is likely associated with thermal weakening and delamination.The central Pamirs show a significant thermal anomaly caused by a concave slab,where the coldest crust does not descend deeply,further suggesting crustal detachment or mechanical failure.The lateral asymmetry in slab temperature possibly explains the mechanism of lateral tearing and differential slab-mantle coupling.展开更多
Automation and intelligence have become the primary trends in the design of investment casting processes.However,the design of gating and riser systems still lacks precise quantitative evaluation criteria.Numerical si...Automation and intelligence have become the primary trends in the design of investment casting processes.However,the design of gating and riser systems still lacks precise quantitative evaluation criteria.Numerical simulation plays a significant role in quantitatively evaluating current processes and making targeted improvements,but its limitations lie in the inability to dynamically reflect the formation outcomes of castings under varying process conditions,making real-time adjustments to gating and riser designs challenging.In this study,an automated design model for gating and riser systems based on integrated parametric 3D modeling-simulation framework is proposed,which enhances the flexibility and usability of evaluating the casting process by simulation.Firstly,geometric feature extraction technology is employed to obtain the geometric information of the target casting.Based on this information,an automated design framework for gating and riser systems is established,incorporating multiple structural parameters for real-time process control.Subsequently,the simulation results for various structural parameters are analyzed,and the influence of these parameters on casting formation is thoroughly investigated.Finally,the optimal design scheme is generated and validated through experimental verification.Simulation analysis and experimental results show that using a larger gate neck(24 mm in side length) and external risers promotes a more uniform temperature distribution and a more stable flow state,effectively eliminating shrinkage cavities and enhancing process yield by 15%.展开更多
Background:Therapeutic responses of breast cancer vary among patients and lead to drug resistance and recurrence due to the heterogeneity.Current preclinical models,however,are inadequate for predicting individual pat...Background:Therapeutic responses of breast cancer vary among patients and lead to drug resistance and recurrence due to the heterogeneity.Current preclinical models,however,are inadequate for predicting individual patient responses towards different drugs.This study aimed to investigate the patient-derived breast cancer culture models for drug sensitivity evaluations.Methods:Tumor and adjacent tissues from female breast cancer patients were collected during surgery.Patient-derived breast cancer cells were cultured using the conditional reprogramming technique to establish 2D models.The obtained patient-derived conditional reprogramming breast cancer(CRBC)cells were subsequently embedded in alginate-gelatin methacryloyl hydrogel microspheres to form 3D culture models.Comparisons between 2D and 3D models were made using immunohistochemistry(tumor markers),MTS assays(cell viability),flow cytometry(apoptosis),transwell assays(migration),and Western blotting(protein expression).Drug sensitivity tests were conducted to evaluate patient-specific responses to anti-cancer agents.Results:2D and 3D culture models were successfully established using samples from eight patients.The 3D models retained histological and marker characteristics of the original tumors.Compared to 2D cultures,3D models exhibited increased apoptosis,enhanced drug resistance,elevated stem cell marker expression,and greater migration ability—features more reflective of in vivo tumor behavior.Conclusion:Patient-derived 3D CRBC models effectively mimic the in vivo tumor microenvironment and demonstrate stronger resistance to anti-cancer drugs than 2D models.These hydrogel-based models offer a cost-effective and clinically relevant platform for drug screening and personalized breast cancer treatment.展开更多
Brain tumor segmentation from Magnetic Resonance Imaging(MRI)supports neurologists and radiologists in analyzing tumors and developing personalized treatment plans,making it a crucial yet challenging task.Supervised m...Brain tumor segmentation from Magnetic Resonance Imaging(MRI)supports neurologists and radiologists in analyzing tumors and developing personalized treatment plans,making it a crucial yet challenging task.Supervised models such as 3D U-Net perform well in this domain,but their accuracy significantly improves with appropriate preprocessing.This paper demonstrates the effectiveness of preprocessing in brain tumor segmentation by applying a pre-segmentation step based on the Generalized Gaussian Mixture Model(GGMM)to T1 contrastenhanced MRI scans from the BraTS 2020 dataset.The Expectation-Maximization(EM)algorithm is employed to estimate parameters for four tissue classes,generating a new pre-segmented channel that enhances the training and performance of the 3DU-Net model.The proposed GGMM+3D U-Net framework achieved a Dice coefficient of 0.88 for whole tumor segmentation,outperforming both the standard multiscale 3D U-Net(0.84)and MMU-Net(0.85).It also delivered higher Intersection over Union(IoU)scores compared to models trained without preprocessing or with simpler GMM-based segmentation.These results,supported by qualitative visualizations,suggest that GGMM-based preprocessing should be integrated into brain tumor segmentation pipelines to optimize performance.展开更多
With the intensification of global warming,marine heatwaves(MHWs)have emerged as a significant extreme hazard,garnering widespread attention and creating a pressing need for accurate prediction.The development of arti...With the intensification of global warming,marine heatwaves(MHWs)have emerged as a significant extreme hazard,garnering widespread attention and creating a pressing need for accurate prediction.The development of artificial intelligence,particularly the application of deep learning to sea surface temperature(SST),has significantly improved the feasibility of predictions.This study utilizes SST and Outgoing Longwave Radiation(OLR)data to train a 3D U-Net model for predicting MHWs in the South China Sea(SCS)with lead times ranging from 1 to 7 days,based on the characteristics of intraseasonal weather processes.Analysis of MHWs occurrences from 1982 to 2023 reveals distinct seasonal patterns,with summer MHWs primarily concentrated in the northern and central SCS,and the highest temperature centers located in the Gulf of Tonkin and west of the Philippines.The 2023 MHW forecast results demonstrate that the 3D U-Net model achieves low error rates and high correlation coefficients with observational data.Incorporating OLR data enhances forecast accuracy compared to SST-only inputs,and training the model exclusively with summer data further improves prediction accuracy.These findings indicate that the proposed method can significantly enhance the accuracy of MHW forecasts.展开更多
Background:Successful liver resection in oncologic surgery depends on safety,precision,and efficacy,all of which require a thorough understanding of liver anatomy.Contrast-enhanced computed tomography(CT)-generated th...Background:Successful liver resection in oncologic surgery depends on safety,precision,and efficacy,all of which require a thorough understanding of liver anatomy.Contrast-enhanced computed tomography(CT)-generated three-dimensional(3D)models have been proposed as a valuable tool to enhance this understanding.However,a systematic comparison of different display modalities across professional groups has not yet been performed.Methods:In this prospective,monocentric randomized trial,we compared high-resolution twodimensional(2D)CT images of liver malignancies with their corresponding standardized,non-colored 3D virtual and printed models in facilitating anatomical and spatial understanding as well as surgical decision-making.A total of 91 participants,including 40 surgeons,10 radiologists,and 41 students,evaluated six clinical cases(three centrally and three peripherally located liver malignancies).Each participant assessed one central and one peripheral case per display modality,presented in a random order.Results:Compared to 2D CT images,both 3D virtual and printed models significantly improved the identification of tumor location(P<0.001),enhanced the comprehension of spatial relationships with adjacent liver and portal veins(P<0.001 and P=0.019,respectively),and facilitated clinical decisionmaking(P<0.001).No significant difference was observed between virtual and printed models in terms of effectiveness.Within the different groups,surgeons and students,but not radiologists,more accurately identified tumor location and spatial relationships with adjacent liver and portal veins using 3D models.Subjectively,most surgeons and students preferred 3D printed models over virtual models and 2D CT images.Conclusions:This study demonstrated that standardized,non-colored 3D virtual and printed models equally help preoperative anatomical understanding and decision-making,particularly for surgeons and students.By isolating the influence of display modality,our findings clarify prior inconsistent results and support the integration of cost-effective 3D visualization by applying virtual models into surgical planning and education.Preference for printed models despite comparable efficacy highlights the importance of user-centered implementation strategies.展开更多
To improve the accuracy of rockburst risk evaluation in mining and tunnelling engineering,the influence of intermediate principal stress σ_(2) deserves further consideration,which has been neglected in general predic...To improve the accuracy of rockburst risk evaluation in mining and tunnelling engineering,the influence of intermediate principal stress σ_(2) deserves further consideration,which has been neglected in general prediction frameworks.This study employs an integrated approach that combines true-triaxial unloading experiments with three-dimensional grain-based discrete element modeling(PFC3D-GBM)to examine the effects of σ_(2) on strain systematically burst and elucidate the underlying mechanisms.Through this dual experimental–numerical methodology,the strainburst characteristics under varying σ_(2) are analyzed in detail regarding mechanical responses,failure evolution and patterns,microscope fracture mechanisms,and energy partitioning.The results indicate that elevated σ_(2) can enhance the bearing capacity of rock,thereby necessitating a higher stress condition required for strainburst.However,it also enlarges the potential strainburst intensity,manifesting as deeper rockburst pits and more violent ejection of rock fragments.An increasing σ_(2) facilitates the microscope transgranular fractures,inhibits intergranular tensile fractures,and raises the kinetic energy conversion ratio slightly.It affects the intensity of strainburst through the following mechanisms,including the increase of energy storage limit,the intensification of Poisson effect for lateral expansion,and the enhancement of the transgranular fracturing mechanism.In practical engineering,the depth and range of support needs to be ensured under high σ_(2) conditions,and it is recommended to use prestressing techniques to control the development of significant slabbing.展开更多
Rockfall hazards pose significant risks to both cultural heritage and populated areas,necessitating comprehensive assessment methodologies.Despite extensive research on rockfalls,only a small number of studies have di...Rockfall hazards pose significant risks to both cultural heritage and populated areas,necessitating comprehensive assessment methodologies.Despite extensive research on rockfalls,only a small number of studies have directly compared empirical methods with modelling approaches.This study investigated rockfalls in five settlements within the Cappadocia region of Türkiye,employing both empirical methods and advanced three-dimensional(3D)probabilistic modeling.The energy line angle approach was applied to identify rockfall propagation zones,while high-resolution digital surface models derived from unmanned aerial vehicle(UAV)imagery facilitated detailed 3D rockfall simulations.Cappadocia’s unique geological setting—comprising alternating layers of ignimbrites and weaker fluviolacustrine deposits—renders it highly susceptible to rockfalls intensified by wetting-drying and freeze-thaw cycles.Results indicate that rockfall propagation characteristics vary markedly between settlements:Göre and Tatlarin exhibit shorter runout distances due to basalt-dominated slopes,whereas Akköy,SoğanlıandŞahinefendi display longer trajectories associated with welded ignimbrites.Empirical cone propagation analyses correspond broadly with field observations,but variations in energy line angles(23°-33°)highlight the necessity for site-specific calibration.Comparative evaluations demonstrate that 3D probabilistic modeling better captures local-scale block dynamics and identifies high-risk areas affected by topographic and structural features such as rockfall ditches.These findings emphasize the importance of integrating empirical and 3D approaches to improve hazard zoning,optimize mitigation structures and guide the protection of Cappadocia’s unique cultural heritage landscape.展开更多
Human motion modeling is a core technology in computer animation,game development,and humancomputer interaction.In particular,generating natural and coherent in-between motion using only the initial and terminal frame...Human motion modeling is a core technology in computer animation,game development,and humancomputer interaction.In particular,generating natural and coherent in-between motion using only the initial and terminal frames remains a fundamental yet unresolved challenge.Existing methods typically rely on dense keyframe inputs or complex prior structures,making it difficult to balance motion quality and plausibility under conditions such as sparse constraints,long-term dependencies,and diverse motion styles.To address this,we propose a motion generation framework based on a frequency-domain diffusion model,which aims to better model complex motion distributions and enhance generation stability under sparse conditions.Our method maps motion sequences to the frequency domain via the Discrete Cosine Transform(DCT),enabling more effective modeling of low-frequency motion structures while suppressing high-frequency noise.A denoising network based on self-attention is introduced to capture long-range temporal dependencies and improve global structural awareness.Additionally,a multi-objective loss function is employed to jointly optimize motion smoothness,pose diversity,and anatomical consistency,enhancing the realism and physical plausibility of the generated sequences.Comparative experiments on the Human3.6M and LaFAN1 datasets demonstrate that our method outperforms state-of-the-art approaches across multiple performance metrics,showing stronger capabilities in generating intermediate motion frames.This research offers a new perspective and methodology for human motion generation and holds promise for applications in character animation,game development,and virtual interaction.展开更多
This study presents and verifies a hybrid methodology for reliable determination of parameters in structural rheological models(Zener,Burgers,and Maxwell)describing the viscoelastic behavior of polyurethane specimens ...This study presents and verifies a hybrid methodology for reliable determination of parameters in structural rheological models(Zener,Burgers,and Maxwell)describing the viscoelastic behavior of polyurethane specimens manufactured using extrusion-based 3D printing.Through comprehensive testing,including cyclic compression at strain rates ranging from 0.12 to 120 mm/min(0%-15%strain)and creep/relaxation experiments(10%-30%strain),the lumped parameters were independently determined using both analytical and numerical solutions of the models’differential equations,followed by cross-verification in additional experiments.Numerical solutions for creep and relaxation problems were obtained using finite element analysis,with the three-parameter Mooney-Rivlin model and Prony series employed to simulate elastic and viscous stress components,respectively.Energy dissipation per cycle was quantified during cyclic compression tests.The results demonstrate that all three models adequately describe material behavior within the 0%-15%strain range across various strain rates.Comparative analysis revealed the Burgers model’s superior performance in characterizing creep and stress relaxation at low strain levels.While Zener and Burgers model parameters from uniaxial compression showed limited applicability for energy dissipation calculations,the generalized Maxwell model effectively captured viscoelastic properties across different strain rates.Notably,parameters derived from creep tests provided a more universal assessment of dissipative properties due to optimization based on characteristic curve regions.Both parameter sets described polyurethane’s elastic-hysteretic behavior with approximately 20%error,proving significantly more accurate than the linear strain-time dependence hypothesis.Finite element analysis(FEA)complemented numerical modeling by demonstrating that while the generalized Maxwell model effectively describes initial rapid stress-strain changes,FEA provides superior characterization of steady-state processes.This computational approach yields more physically representative results compared to simplified analytical solutions,despite certain limitations in transient analysis.展开更多
The three-dimensional(3D)geometry of a fault is a critical control on earthquake nucleation,dynamic rupture,stress triggering,and related seismic hazards.Therefore,a 3D model of an active fault can significantly impro...The three-dimensional(3D)geometry of a fault is a critical control on earthquake nucleation,dynamic rupture,stress triggering,and related seismic hazards.Therefore,a 3D model of an active fault can significantly improve our understanding of seismogenesis and our ability to evaluate seismic hazards.Utilising the SKUA GoCAD software,we constructed detailed seismic fault models for the 2021 M_(S)6.4 Yangbi earthquake in Yunnan,China,using two sets of relocated earthquake catalogs and focal mechanism solutions following a convenient 3D fault modeling workflow.Our analysis revealed a NW-striking main fault with a high-angle SW dip,accompanied by two branch faults.Interpretation of one dataset revealed a single NNW-striking branch fault SW of the main fault,whereas the other dataset indicated four steep NNE-striking segments with a left-echelon pattern.Additionally,a third ENE-striking short fault was identified NE of the main fault.In combination with the spatial distribution of pre-existing faults,our 3D fault models indicate that the Yangbi earthquake reactivated pre-existing NW-and NE-striking fault directions rather than the surface-exposed Weixi-Qiaohou-Weishan Fault zone.The occurrence of the Yangbi earthquake demonstrates that the reactivation of pre-existing faults away from active fault zones,through either cascade or conjugate rupture modes,can cause unexpected moderate-large earthquakes and severe disasters,necessitating attention in regions like southeast Xizang,which have complex fault systems.展开更多
The spatial distribution of discontinuities and the size of rock blocks are the key indicators for rock mass quality evaluation and rockfall risk assessment.Traditional manual measurement is often dangerous or unreach...The spatial distribution of discontinuities and the size of rock blocks are the key indicators for rock mass quality evaluation and rockfall risk assessment.Traditional manual measurement is often dangerous or unreachable at some high-steep rock slopes.In contrast,unmanned aerial vehicle(UAV)photogrammetry is not limited by terrain conditions,and can efficiently collect high-precision three-dimensional(3D)point clouds of rock masses through all-round and multiangle photography for rock mass characterization.In this paper,a new method based on a 3D point cloud is proposed for discontinuity identification and refined rock block modeling.The method is based on four steps:(1)Establish a point cloud spatial topology,and calculate the point cloud normal vector and average point spacing based on several machine learning algorithms;(2)Extract discontinuities using the density-based spatial clustering of applications with noise(DBSCAN)algorithm and fit the discontinuity plane by combining principal component analysis(PCA)with the natural breaks(NB)method;(3)Propose a method of inserting points in the line segment to generate an embedded discontinuity point cloud;and(4)Adopt a Poisson reconstruction method for refined rock block modeling.The proposed method was applied to an outcrop of an ultrahigh steep rock slope and compared with the results of previous studies and manual surveys.The results show that the method can eliminate the influence of discontinuity undulations on the orientation measurement and describe the local concave-convex characteristics on the modeling of rock blocks.The calculation results are accurate and reliable,which can meet the practical requirements of engineering.展开更多
This research pioneers the integration of geographic information systems(GIS)and 3D modeling within a virtual reality(VR)framework to assess the viability and planning of a 20 MW hybrid wind-solarphotovoltaic(PV)syste...This research pioneers the integration of geographic information systems(GIS)and 3D modeling within a virtual reality(VR)framework to assess the viability and planning of a 20 MW hybrid wind-solarphotovoltaic(PV)system connected to the local grid.The study focuses on Dakhla,Morocco,a region with vast untapped renewable energy potential.By leveraging GIS,we are innovatively analyzing geographical and environmental factors that influence optimal site selection and system design.The incorporation of VR technologies offers an unprecedented level of realism and immersion,allowing stakeholders to virtually experience the project's impact and design in a dynamic,interactive environment.This novel methodology includes extensive data collection,advanced modeling,and simulations,ensuring that the hybrid system is precisely tailored to the unique climatic and environmental conditions of Dakhla.Our analysis reveals that the region possesses a photovoltaic solar potential of approximately2400 k Wh/m^(2) per year,with an average annual wind power density of about 434 W/m^(2) at an 80-meter hub height.Productivity simulations indicate that the 20 MW hybrid system could generate approximately 60 GWh of energy per year and 1369 GWh over its 25-year lifespan.To validate these findings,we employed the System Advisor Model(SAM)software and the Global Solar Photovoltaic Atlas platform.This comprehensive and interdisciplinary approach not only provides a robust assessment of the system's feasibility but also offers valuable insights into its potential socio-economic and environmental impact.展开更多
This study introduces a novel method for reconstructing the 3D model of aluminum foam using cross-sectional sequence images.Combining precision milling and image acquisition,high-qual-ity cross-sectional images are ob...This study introduces a novel method for reconstructing the 3D model of aluminum foam using cross-sectional sequence images.Combining precision milling and image acquisition,high-qual-ity cross-sectional images are obtained.Pore structures are segmented by the U-shaped network(U-Net)neural network integrated with the Canny edge detection operator,ensuring accurate pore delineation and edge extraction.The trained U-Net achieves 98.55%accuracy.The 2D data are superimposed and processed into 3D point clouds,enabling reconstruction of the pore structure and aluminum skeleton.Analysis of pore 01 shows the cross-sectional area initially increases,and then decreases with milling depth,with a uniform point distribution of 40 per layer.The reconstructed model exhibits a porosity of 77.5%,with section overlap rates between the 2D pore segmentation and the reconstructed model exceeding 96%,confirming high fidelity.Equivalent sphere diameters decrease with size,averaging 1.95 mm.Compression simulations reveal that the stress-strain curve of the 3D reconstruction model of aluminum foam exhibits fluctuations,and the stresses in the reconstruction model concentrate on thin cell walls,leading to localized deformations.This method accurately restores the aluminum foam’s complex internal structure,improving reconstruction preci-sion and simulation reliability.The approach offers a cost-efficient,high-precision technique for optimizing material performance in engineering applications.展开更多
The deep convolutional neural network U-net has been introduced into adaptive subtraction, which is a critical step in effectively suppressing seismic multiples. The U-net approach has higher precision than the tradit...The deep convolutional neural network U-net has been introduced into adaptive subtraction, which is a critical step in effectively suppressing seismic multiples. The U-net approach has higher precision than the traditional linear regression approach. However, the existing 2D U-net approach with 2D data windows can not deal with elaborate discrepancies between the actual and simulated multiples along the gather direction. It may lead to erroneous preservation of primaries or generate obvious vestigial multiples, especially in complex media. To further enhance the multiple suppression accuracy, we present an adaptive subtraction approach utilizing 3D U-net architecture, which can adaptively separate primaries and multiples utilizing 3D windows. The utilization of 3D windows allows for enhanced depiction of spatial continuity and anisotropy of seismic events along the gather direction in comparison to 2D windows. The 3D U-net approach with 3D windows can more effectively preserve the continuity of primaries and manage the complex disparities between the actual and simulated multiples. The proposed 3D U-net approach exhibits 1 dB improvement in the signal-to-noise ratio compared to the 2D U-net approach, as observed in the synthesis data section, and exhibits more outstanding performance in the preservation of primaries and removal of residual multiples in both synthesis and reality data sections. Moreover, to expedite network training in our proposed 3D U-net approach we employ the transfer learning (TL) strategy by utilizing the network parameters of 3D U-net estimated in the preceding data segment as the initial network parameters of 3D U-net for the subsequent data segment. In the reality data section, the 3D U-net approach incorporating TL reduces the computational expense by 70% compared to the one without TL.展开更多
Accurate brain tumour segmentation is critical for diagnosis and treatment planning, yet challenging due to tumour complexity. Manual segmentation is time-consuming and variable, necessitating automated methods. Deep ...Accurate brain tumour segmentation is critical for diagnosis and treatment planning, yet challenging due to tumour complexity. Manual segmentation is time-consuming and variable, necessitating automated methods. Deep learning, particularly 3D U-Net architectures, has revolutionised medical image analysis by leveraging volumetric data to capture spatial context, enhancing segmentation accuracy. This paper reviews brain tumour segmentation methods, emphasising 3D U-Net advancements. We analyse contributions from the Brain Tumour Segmentation (BraTS) challenges (2014-2023), highlighting key improvements and persistent challenges, including tumour heterogeneity, limited annotated data, varied imaging protocols, computational constraints, and model generalisation. Unlike previous reviews, we synthesise these challenges, proposing targeted research directions: enhancing model robustness through domain adaptation and multi-institutional data sharing, developing lightweight architectures for clinical deployment, integrating multi-modal and clinical data, and incorporating explainability techniques to build clinician trust. By addressing these challenges, we aim to guide future research toward developing more robust, generalisable, and clinically applicable segmentation models, ultimately improving patient outcomes in neuro-oncology.展开更多
Highway planning requires geological surveys and stability analysis of the surrounding area.In the early stage of the survey,the modeling and stability analysis of the survey area can be carried out by using GIS softw...Highway planning requires geological surveys and stability analysis of the surrounding area.In the early stage of the survey,the modeling and stability analysis of the survey area can be carried out by using GIS software to intuitively understand the topography of the study area.The use of DEM to extract terrain factors can be used for simple stability analysis and the source data is easy to obtain,simple to operate,fast to analyze,and reliable analysis results.In this paper,taking the X104 road section in Ganxian County as an example,the ArcGIS platform is used to carry out 3D modeling visualization and stability analysis,and the stability evaluation map of the study area is obtained.展开更多
With the widespread application of 3D visualization in digital exhibition halls and virtual reality,achieving efficient rendering and high-fidelity presentation has become a key challenge.This study proposes a hybrid ...With the widespread application of 3D visualization in digital exhibition halls and virtual reality,achieving efficient rendering and high-fidelity presentation has become a key challenge.This study proposes a hybrid point cloud generation method that combines traditional sampling with 3D Gaussian splatting,aiming to address the issues of rendering delay and missing details in existing 3D displays.By improving the OBJ model parsing process and incorporating an adaptive area-weighted sampling algorithm,we achieve adaptive point cloud generation based on triangle density.Innovatively,we advance the ellipsoidal parameter estimation process of 3D Gaussian splatting to the point cloud generation stage.By establishing a mathematical relationship between the covariance matrix and local curvature,the generated point cloud naturally exhibits Gaussian distribution characteristics.Experimental results show that,compared to traditional methods,our approach reduces point cloud data by 38% while maintaining equivalent visual quality at a 4096×4096 texture resolution.By introducing mipmap texture optimization strategies and a GPU-accelerated rasterization pipeline,stable rendering at 60 frames per second is achieved in a WebGL environment.Additionally,we quantize and compress the spherical harmonic function parameters specific to 3D Gaussian splatting,reducing network transmission bandwidth to 52% of the original data.This study provides a new technical pathway for fields requiring high-precision display,such as the digitization of cultural heritage.展开更多
文摘Deep learning, especially through convolutional neural networks (CNN) such as the U-Net 3D model, has revolutionized fault identification from seismic data, representing a significant leap over traditional methods. Our review traces the evolution of CNN, emphasizing the adaptation and capabilities of the U-Net 3D model in automating seismic fault delineation with unprecedented accuracy. We find: 1) The transition from basic neural networks to sophisticated CNN has enabled remarkable advancements in image recognition, which are directly applicable to analyzing seismic data. The U-Net 3D model, with its innovative architecture, exemplifies this progress by providing a method for detailed and accurate fault detection with reduced manual interpretation bias. 2) The U-Net 3D model has demonstrated its superiority over traditional fault identification methods in several key areas: it has enhanced interpretation accuracy, increased operational efficiency, and reduced the subjectivity of manual methods. 3) Despite these achievements, challenges such as the need for effective data preprocessing, acquisition of high-quality annotated datasets, and achieving model generalization across different geological conditions remain. Future research should therefore focus on developing more complex network architectures and innovative training strategies to refine fault identification performance further. Our findings confirm the transformative potential of deep learning, particularly CNN like the U-Net 3D model, in geosciences, advocating for its broader integration to revolutionize geological exploration and seismic analysis.
基金the Chinese Academy of Sciences Pioneer Hundred Talents Program and the Second Tibetan Plateau Scientific Expedition and Research Program(Grant No.2019QZKK0708)supported by a MEXT(Ministry of Education,Culture,Sports,Science and Technology)KAKENHI(Grants-in-Aid for Scientific Research)grant(Grant No.21H05203)Kobe University Strategic International Collaborative Research Grant(Type B Fostering Joint Research).
文摘The intracontinental subduction of a>200-km-long section of the Tajik-Tarim lithosphere beneath the Pamir Mountains is proposed to explain nearly 30 km of shortening in the Tajik fold-thrust belt and the Pamir uplift.Seismic imaging revealed that the upper slab was scraped and that the lower slab had subducted to a depth of>150 km.These features constitute the tectonic complexity of the Pamirs,as well as the thermal subduction mechanism involved,which remains poorly understood.Hence,in this study,high-resolution three-dimensional(3D)kinematic modeling is applied to investigate the thermal structure and geometry of the subducting slab beneath the Pamirs.The modeled slab configuration reveals distinct along-strike variations,with a steeply dipping slab beneath the southern Pamirs,a more gently inclined slab beneath the northern Pamirs,and apparent upper slab termination at shallow depths beneath the Pamirs.The thermal field reveals a cold slab core after delamination,with temperatures ranging from 400℃to 800℃,enveloped by a hotter mantle reaching~1400℃.The occurrence of intermediate-depth earthquakes aligns primarily with colder slab regions,particularly near the slab tear-off below the southwestern Pamirs,indicating a strong correlation between slab temperature and seismicity.In contrast,the northern Pamirs exhibit reduced seismicity at depth,which is likely associated with thermal weakening and delamination.The central Pamirs show a significant thermal anomaly caused by a concave slab,where the coldest crust does not descend deeply,further suggesting crustal detachment or mechanical failure.The lateral asymmetry in slab temperature possibly explains the mechanism of lateral tearing and differential slab-mantle coupling.
基金financially supported by the National Key Research and Development Program of China (2022YFB3706802)。
文摘Automation and intelligence have become the primary trends in the design of investment casting processes.However,the design of gating and riser systems still lacks precise quantitative evaluation criteria.Numerical simulation plays a significant role in quantitatively evaluating current processes and making targeted improvements,but its limitations lie in the inability to dynamically reflect the formation outcomes of castings under varying process conditions,making real-time adjustments to gating and riser designs challenging.In this study,an automated design model for gating and riser systems based on integrated parametric 3D modeling-simulation framework is proposed,which enhances the flexibility and usability of evaluating the casting process by simulation.Firstly,geometric feature extraction technology is employed to obtain the geometric information of the target casting.Based on this information,an automated design framework for gating and riser systems is established,incorporating multiple structural parameters for real-time process control.Subsequently,the simulation results for various structural parameters are analyzed,and the influence of these parameters on casting formation is thoroughly investigated.Finally,the optimal design scheme is generated and validated through experimental verification.Simulation analysis and experimental results show that using a larger gate neck(24 mm in side length) and external risers promotes a more uniform temperature distribution and a more stable flow state,effectively eliminating shrinkage cavities and enhancing process yield by 15%.
基金supported by the Natural Science Foundation of Guangdong Province(No.2021B1515120053)Guangdong Basic and Applied Basic Research Foundation(Grant No.2024A1515140166).
文摘Background:Therapeutic responses of breast cancer vary among patients and lead to drug resistance and recurrence due to the heterogeneity.Current preclinical models,however,are inadequate for predicting individual patient responses towards different drugs.This study aimed to investigate the patient-derived breast cancer culture models for drug sensitivity evaluations.Methods:Tumor and adjacent tissues from female breast cancer patients were collected during surgery.Patient-derived breast cancer cells were cultured using the conditional reprogramming technique to establish 2D models.The obtained patient-derived conditional reprogramming breast cancer(CRBC)cells were subsequently embedded in alginate-gelatin methacryloyl hydrogel microspheres to form 3D culture models.Comparisons between 2D and 3D models were made using immunohistochemistry(tumor markers),MTS assays(cell viability),flow cytometry(apoptosis),transwell assays(migration),and Western blotting(protein expression).Drug sensitivity tests were conducted to evaluate patient-specific responses to anti-cancer agents.Results:2D and 3D culture models were successfully established using samples from eight patients.The 3D models retained histological and marker characteristics of the original tumors.Compared to 2D cultures,3D models exhibited increased apoptosis,enhanced drug resistance,elevated stem cell marker expression,and greater migration ability—features more reflective of in vivo tumor behavior.Conclusion:Patient-derived 3D CRBC models effectively mimic the in vivo tumor microenvironment and demonstrate stronger resistance to anti-cancer drugs than 2D models.These hydrogel-based models offer a cost-effective and clinically relevant platform for drug screening and personalized breast cancer treatment.
基金Princess Nourah Bint Abdulrahman University Researchers Supporting Project number(PNURSP2025R826),Princess Nourah Bint Abdulrahman University,Riyadh,Saudi ArabiaNorthern Border University,Saudi Arabia,for supporting this work through project number(NBU-CRP-2025-2933).
文摘Brain tumor segmentation from Magnetic Resonance Imaging(MRI)supports neurologists and radiologists in analyzing tumors and developing personalized treatment plans,making it a crucial yet challenging task.Supervised models such as 3D U-Net perform well in this domain,but their accuracy significantly improves with appropriate preprocessing.This paper demonstrates the effectiveness of preprocessing in brain tumor segmentation by applying a pre-segmentation step based on the Generalized Gaussian Mixture Model(GGMM)to T1 contrastenhanced MRI scans from the BraTS 2020 dataset.The Expectation-Maximization(EM)algorithm is employed to estimate parameters for four tissue classes,generating a new pre-segmented channel that enhances the training and performance of the 3DU-Net model.The proposed GGMM+3D U-Net framework achieved a Dice coefficient of 0.88 for whole tumor segmentation,outperforming both the standard multiscale 3D U-Net(0.84)and MMU-Net(0.85).It also delivered higher Intersection over Union(IoU)scores compared to models trained without preprocessing or with simpler GMM-based segmentation.These results,supported by qualitative visualizations,suggest that GGMM-based preprocessing should be integrated into brain tumor segmentation pipelines to optimize performance.
基金Guangdong Major Project of Basic and Applied Basic Research(2020B0301030004)。
文摘With the intensification of global warming,marine heatwaves(MHWs)have emerged as a significant extreme hazard,garnering widespread attention and creating a pressing need for accurate prediction.The development of artificial intelligence,particularly the application of deep learning to sea surface temperature(SST),has significantly improved the feasibility of predictions.This study utilizes SST and Outgoing Longwave Radiation(OLR)data to train a 3D U-Net model for predicting MHWs in the South China Sea(SCS)with lead times ranging from 1 to 7 days,based on the characteristics of intraseasonal weather processes.Analysis of MHWs occurrences from 1982 to 2023 reveals distinct seasonal patterns,with summer MHWs primarily concentrated in the northern and central SCS,and the highest temperature centers located in the Gulf of Tonkin and west of the Philippines.The 2023 MHW forecast results demonstrate that the 3D U-Net model achieves low error rates and high correlation coefficients with observational data.Incorporating OLR data enhances forecast accuracy compared to SST-only inputs,and training the model exclusively with summer data further improves prediction accuracy.These findings indicate that the proposed method can significantly enhance the accuracy of MHW forecasts.
文摘Background:Successful liver resection in oncologic surgery depends on safety,precision,and efficacy,all of which require a thorough understanding of liver anatomy.Contrast-enhanced computed tomography(CT)-generated three-dimensional(3D)models have been proposed as a valuable tool to enhance this understanding.However,a systematic comparison of different display modalities across professional groups has not yet been performed.Methods:In this prospective,monocentric randomized trial,we compared high-resolution twodimensional(2D)CT images of liver malignancies with their corresponding standardized,non-colored 3D virtual and printed models in facilitating anatomical and spatial understanding as well as surgical decision-making.A total of 91 participants,including 40 surgeons,10 radiologists,and 41 students,evaluated six clinical cases(three centrally and three peripherally located liver malignancies).Each participant assessed one central and one peripheral case per display modality,presented in a random order.Results:Compared to 2D CT images,both 3D virtual and printed models significantly improved the identification of tumor location(P<0.001),enhanced the comprehension of spatial relationships with adjacent liver and portal veins(P<0.001 and P=0.019,respectively),and facilitated clinical decisionmaking(P<0.001).No significant difference was observed between virtual and printed models in terms of effectiveness.Within the different groups,surgeons and students,but not radiologists,more accurately identified tumor location and spatial relationships with adjacent liver and portal veins using 3D models.Subjectively,most surgeons and students preferred 3D printed models over virtual models and 2D CT images.Conclusions:This study demonstrated that standardized,non-colored 3D virtual and printed models equally help preoperative anatomical understanding and decision-making,particularly for surgeons and students.By isolating the influence of display modality,our findings clarify prior inconsistent results and support the integration of cost-effective 3D visualization by applying virtual models into surgical planning and education.Preference for printed models despite comparable efficacy highlights the importance of user-centered implementation strategies.
基金supported by the National Natural Science Foundation of China(No.42507210)the Fundamental Research Funds for the Central Universities(No.2025XJSB01)+1 种基金the State Key Laboratory for Tunnel Engineering(No.SKLTEK202421)the Foundation of Key Laboratory of Deep Coal Resource Mining(China University of Mining and Technology),Ministry of Education(No.KLDCRMMOE24KF11).
文摘To improve the accuracy of rockburst risk evaluation in mining and tunnelling engineering,the influence of intermediate principal stress σ_(2) deserves further consideration,which has been neglected in general prediction frameworks.This study employs an integrated approach that combines true-triaxial unloading experiments with three-dimensional grain-based discrete element modeling(PFC3D-GBM)to examine the effects of σ_(2) on strain systematically burst and elucidate the underlying mechanisms.Through this dual experimental–numerical methodology,the strainburst characteristics under varying σ_(2) are analyzed in detail regarding mechanical responses,failure evolution and patterns,microscope fracture mechanisms,and energy partitioning.The results indicate that elevated σ_(2) can enhance the bearing capacity of rock,thereby necessitating a higher stress condition required for strainburst.However,it also enlarges the potential strainburst intensity,manifesting as deeper rockburst pits and more violent ejection of rock fragments.An increasing σ_(2) facilitates the microscope transgranular fractures,inhibits intergranular tensile fractures,and raises the kinetic energy conversion ratio slightly.It affects the intensity of strainburst through the following mechanisms,including the increase of energy storage limit,the intensification of Poisson effect for lateral expansion,and the enhancement of the transgranular fracturing mechanism.In practical engineering,the depth and range of support needs to be ensured under high σ_(2) conditions,and it is recommended to use prestressing techniques to control the development of significant slabbing.
基金financially supported by The Scientific and Technological Research Council of Türkiye(T??B1TAK)with the project number 121C420。
文摘Rockfall hazards pose significant risks to both cultural heritage and populated areas,necessitating comprehensive assessment methodologies.Despite extensive research on rockfalls,only a small number of studies have directly compared empirical methods with modelling approaches.This study investigated rockfalls in five settlements within the Cappadocia region of Türkiye,employing both empirical methods and advanced three-dimensional(3D)probabilistic modeling.The energy line angle approach was applied to identify rockfall propagation zones,while high-resolution digital surface models derived from unmanned aerial vehicle(UAV)imagery facilitated detailed 3D rockfall simulations.Cappadocia’s unique geological setting—comprising alternating layers of ignimbrites and weaker fluviolacustrine deposits—renders it highly susceptible to rockfalls intensified by wetting-drying and freeze-thaw cycles.Results indicate that rockfall propagation characteristics vary markedly between settlements:Göre and Tatlarin exhibit shorter runout distances due to basalt-dominated slopes,whereas Akköy,SoğanlıandŞahinefendi display longer trajectories associated with welded ignimbrites.Empirical cone propagation analyses correspond broadly with field observations,but variations in energy line angles(23°-33°)highlight the necessity for site-specific calibration.Comparative evaluations demonstrate that 3D probabilistic modeling better captures local-scale block dynamics and identifies high-risk areas affected by topographic and structural features such as rockfall ditches.These findings emphasize the importance of integrating empirical and 3D approaches to improve hazard zoning,optimize mitigation structures and guide the protection of Cappadocia’s unique cultural heritage landscape.
基金supported by the National Natural Science Foundation of China(Grant No.72161034).
文摘Human motion modeling is a core technology in computer animation,game development,and humancomputer interaction.In particular,generating natural and coherent in-between motion using only the initial and terminal frames remains a fundamental yet unresolved challenge.Existing methods typically rely on dense keyframe inputs or complex prior structures,making it difficult to balance motion quality and plausibility under conditions such as sparse constraints,long-term dependencies,and diverse motion styles.To address this,we propose a motion generation framework based on a frequency-domain diffusion model,which aims to better model complex motion distributions and enhance generation stability under sparse conditions.Our method maps motion sequences to the frequency domain via the Discrete Cosine Transform(DCT),enabling more effective modeling of low-frequency motion structures while suppressing high-frequency noise.A denoising network based on self-attention is introduced to capture long-range temporal dependencies and improve global structural awareness.Additionally,a multi-objective loss function is employed to jointly optimize motion smoothness,pose diversity,and anatomical consistency,enhancing the realism and physical plausibility of the generated sequences.Comparative experiments on the Human3.6M and LaFAN1 datasets demonstrate that our method outperforms state-of-the-art approaches across multiple performance metrics,showing stronger capabilities in generating intermediate motion frames.This research offers a new perspective and methodology for human motion generation and holds promise for applications in character animation,game development,and virtual interaction.
文摘This study presents and verifies a hybrid methodology for reliable determination of parameters in structural rheological models(Zener,Burgers,and Maxwell)describing the viscoelastic behavior of polyurethane specimens manufactured using extrusion-based 3D printing.Through comprehensive testing,including cyclic compression at strain rates ranging from 0.12 to 120 mm/min(0%-15%strain)and creep/relaxation experiments(10%-30%strain),the lumped parameters were independently determined using both analytical and numerical solutions of the models’differential equations,followed by cross-verification in additional experiments.Numerical solutions for creep and relaxation problems were obtained using finite element analysis,with the three-parameter Mooney-Rivlin model and Prony series employed to simulate elastic and viscous stress components,respectively.Energy dissipation per cycle was quantified during cyclic compression tests.The results demonstrate that all three models adequately describe material behavior within the 0%-15%strain range across various strain rates.Comparative analysis revealed the Burgers model’s superior performance in characterizing creep and stress relaxation at low strain levels.While Zener and Burgers model parameters from uniaxial compression showed limited applicability for energy dissipation calculations,the generalized Maxwell model effectively captured viscoelastic properties across different strain rates.Notably,parameters derived from creep tests provided a more universal assessment of dissipative properties due to optimization based on characteristic curve regions.Both parameter sets described polyurethane’s elastic-hysteretic behavior with approximately 20%error,proving significantly more accurate than the linear strain-time dependence hypothesis.Finite element analysis(FEA)complemented numerical modeling by demonstrating that while the generalized Maxwell model effectively describes initial rapid stress-strain changes,FEA provides superior characterization of steady-state processes.This computational approach yields more physically representative results compared to simplified analytical solutions,despite certain limitations in transient analysis.
基金financial support from the National Key R&D Program of China (No. 2021YFC3000600)National Natural Science Foundation of China (No. 41872206)National Nonprofit Fundamental Research Grant of China, Institute of Geology, China, Earthquake Administration (No. IGCEA2010)
文摘The three-dimensional(3D)geometry of a fault is a critical control on earthquake nucleation,dynamic rupture,stress triggering,and related seismic hazards.Therefore,a 3D model of an active fault can significantly improve our understanding of seismogenesis and our ability to evaluate seismic hazards.Utilising the SKUA GoCAD software,we constructed detailed seismic fault models for the 2021 M_(S)6.4 Yangbi earthquake in Yunnan,China,using two sets of relocated earthquake catalogs and focal mechanism solutions following a convenient 3D fault modeling workflow.Our analysis revealed a NW-striking main fault with a high-angle SW dip,accompanied by two branch faults.Interpretation of one dataset revealed a single NNW-striking branch fault SW of the main fault,whereas the other dataset indicated four steep NNE-striking segments with a left-echelon pattern.Additionally,a third ENE-striking short fault was identified NE of the main fault.In combination with the spatial distribution of pre-existing faults,our 3D fault models indicate that the Yangbi earthquake reactivated pre-existing NW-and NE-striking fault directions rather than the surface-exposed Weixi-Qiaohou-Weishan Fault zone.The occurrence of the Yangbi earthquake demonstrates that the reactivation of pre-existing faults away from active fault zones,through either cascade or conjugate rupture modes,can cause unexpected moderate-large earthquakes and severe disasters,necessitating attention in regions like southeast Xizang,which have complex fault systems.
基金supported by the National Natural Science Foundation of China(Grant Nos.41941017 and 42177139)Graduate Innovation Fund of Jilin University(Grant No.2024CX099)。
文摘The spatial distribution of discontinuities and the size of rock blocks are the key indicators for rock mass quality evaluation and rockfall risk assessment.Traditional manual measurement is often dangerous or unreachable at some high-steep rock slopes.In contrast,unmanned aerial vehicle(UAV)photogrammetry is not limited by terrain conditions,and can efficiently collect high-precision three-dimensional(3D)point clouds of rock masses through all-round and multiangle photography for rock mass characterization.In this paper,a new method based on a 3D point cloud is proposed for discontinuity identification and refined rock block modeling.The method is based on four steps:(1)Establish a point cloud spatial topology,and calculate the point cloud normal vector and average point spacing based on several machine learning algorithms;(2)Extract discontinuities using the density-based spatial clustering of applications with noise(DBSCAN)algorithm and fit the discontinuity plane by combining principal component analysis(PCA)with the natural breaks(NB)method;(3)Propose a method of inserting points in the line segment to generate an embedded discontinuity point cloud;and(4)Adopt a Poisson reconstruction method for refined rock block modeling.The proposed method was applied to an outcrop of an ultrahigh steep rock slope and compared with the results of previous studies and manual surveys.The results show that the method can eliminate the influence of discontinuity undulations on the orientation measurement and describe the local concave-convex characteristics on the modeling of rock blocks.The calculation results are accurate and reliable,which can meet the practical requirements of engineering.
文摘This research pioneers the integration of geographic information systems(GIS)and 3D modeling within a virtual reality(VR)framework to assess the viability and planning of a 20 MW hybrid wind-solarphotovoltaic(PV)system connected to the local grid.The study focuses on Dakhla,Morocco,a region with vast untapped renewable energy potential.By leveraging GIS,we are innovatively analyzing geographical and environmental factors that influence optimal site selection and system design.The incorporation of VR technologies offers an unprecedented level of realism and immersion,allowing stakeholders to virtually experience the project's impact and design in a dynamic,interactive environment.This novel methodology includes extensive data collection,advanced modeling,and simulations,ensuring that the hybrid system is precisely tailored to the unique climatic and environmental conditions of Dakhla.Our analysis reveals that the region possesses a photovoltaic solar potential of approximately2400 k Wh/m^(2) per year,with an average annual wind power density of about 434 W/m^(2) at an 80-meter hub height.Productivity simulations indicate that the 20 MW hybrid system could generate approximately 60 GWh of energy per year and 1369 GWh over its 25-year lifespan.To validate these findings,we employed the System Advisor Model(SAM)software and the Global Solar Photovoltaic Atlas platform.This comprehensive and interdisciplinary approach not only provides a robust assessment of the system's feasibility but also offers valuable insights into its potential socio-economic and environmental impact.
基金supported by the Key Research and DevelopmentPlan in Shanxi Province of China(No.201803D421045)the Natural Science Foundation of Shanxi Province(No.2021-0302-123104)。
文摘This study introduces a novel method for reconstructing the 3D model of aluminum foam using cross-sectional sequence images.Combining precision milling and image acquisition,high-qual-ity cross-sectional images are obtained.Pore structures are segmented by the U-shaped network(U-Net)neural network integrated with the Canny edge detection operator,ensuring accurate pore delineation and edge extraction.The trained U-Net achieves 98.55%accuracy.The 2D data are superimposed and processed into 3D point clouds,enabling reconstruction of the pore structure and aluminum skeleton.Analysis of pore 01 shows the cross-sectional area initially increases,and then decreases with milling depth,with a uniform point distribution of 40 per layer.The reconstructed model exhibits a porosity of 77.5%,with section overlap rates between the 2D pore segmentation and the reconstructed model exceeding 96%,confirming high fidelity.Equivalent sphere diameters decrease with size,averaging 1.95 mm.Compression simulations reveal that the stress-strain curve of the 3D reconstruction model of aluminum foam exhibits fluctuations,and the stresses in the reconstruction model concentrate on thin cell walls,leading to localized deformations.This method accurately restores the aluminum foam’s complex internal structure,improving reconstruction preci-sion and simulation reliability.The approach offers a cost-efficient,high-precision technique for optimizing material performance in engineering applications.
基金supported by National Natural Science Foundation of China(42364008,41804110)in part by Guizhou Provincial Basic Research Program(Natural Science)(ZK[2022]060)+1 种基金in part by China Postdoctoral Science Foundation(2022M723127)in part by Youth Innovation Team Project of Shandong Provincial Education Department(2022KJ141).
文摘The deep convolutional neural network U-net has been introduced into adaptive subtraction, which is a critical step in effectively suppressing seismic multiples. The U-net approach has higher precision than the traditional linear regression approach. However, the existing 2D U-net approach with 2D data windows can not deal with elaborate discrepancies between the actual and simulated multiples along the gather direction. It may lead to erroneous preservation of primaries or generate obvious vestigial multiples, especially in complex media. To further enhance the multiple suppression accuracy, we present an adaptive subtraction approach utilizing 3D U-net architecture, which can adaptively separate primaries and multiples utilizing 3D windows. The utilization of 3D windows allows for enhanced depiction of spatial continuity and anisotropy of seismic events along the gather direction in comparison to 2D windows. The 3D U-net approach with 3D windows can more effectively preserve the continuity of primaries and manage the complex disparities between the actual and simulated multiples. The proposed 3D U-net approach exhibits 1 dB improvement in the signal-to-noise ratio compared to the 2D U-net approach, as observed in the synthesis data section, and exhibits more outstanding performance in the preservation of primaries and removal of residual multiples in both synthesis and reality data sections. Moreover, to expedite network training in our proposed 3D U-net approach we employ the transfer learning (TL) strategy by utilizing the network parameters of 3D U-net estimated in the preceding data segment as the initial network parameters of 3D U-net for the subsequent data segment. In the reality data section, the 3D U-net approach incorporating TL reduces the computational expense by 70% compared to the one without TL.
文摘Accurate brain tumour segmentation is critical for diagnosis and treatment planning, yet challenging due to tumour complexity. Manual segmentation is time-consuming and variable, necessitating automated methods. Deep learning, particularly 3D U-Net architectures, has revolutionised medical image analysis by leveraging volumetric data to capture spatial context, enhancing segmentation accuracy. This paper reviews brain tumour segmentation methods, emphasising 3D U-Net advancements. We analyse contributions from the Brain Tumour Segmentation (BraTS) challenges (2014-2023), highlighting key improvements and persistent challenges, including tumour heterogeneity, limited annotated data, varied imaging protocols, computational constraints, and model generalisation. Unlike previous reviews, we synthesise these challenges, proposing targeted research directions: enhancing model robustness through domain adaptation and multi-institutional data sharing, developing lightweight architectures for clinical deployment, integrating multi-modal and clinical data, and incorporating explainability techniques to build clinician trust. By addressing these challenges, we aim to guide future research toward developing more robust, generalisable, and clinically applicable segmentation models, ultimately improving patient outcomes in neuro-oncology.
基金National Undergraduate Training Program for Innovation and Entrepreneurship(Project No.:202310407006)。
文摘Highway planning requires geological surveys and stability analysis of the surrounding area.In the early stage of the survey,the modeling and stability analysis of the survey area can be carried out by using GIS software to intuitively understand the topography of the study area.The use of DEM to extract terrain factors can be used for simple stability analysis and the source data is easy to obtain,simple to operate,fast to analyze,and reliable analysis results.In this paper,taking the X104 road section in Ganxian County as an example,the ArcGIS platform is used to carry out 3D modeling visualization and stability analysis,and the stability evaluation map of the study area is obtained.
文摘With the widespread application of 3D visualization in digital exhibition halls and virtual reality,achieving efficient rendering and high-fidelity presentation has become a key challenge.This study proposes a hybrid point cloud generation method that combines traditional sampling with 3D Gaussian splatting,aiming to address the issues of rendering delay and missing details in existing 3D displays.By improving the OBJ model parsing process and incorporating an adaptive area-weighted sampling algorithm,we achieve adaptive point cloud generation based on triangle density.Innovatively,we advance the ellipsoidal parameter estimation process of 3D Gaussian splatting to the point cloud generation stage.By establishing a mathematical relationship between the covariance matrix and local curvature,the generated point cloud naturally exhibits Gaussian distribution characteristics.Experimental results show that,compared to traditional methods,our approach reduces point cloud data by 38% while maintaining equivalent visual quality at a 4096×4096 texture resolution.By introducing mipmap texture optimization strategies and a GPU-accelerated rasterization pipeline,stable rendering at 60 frames per second is achieved in a WebGL environment.Additionally,we quantize and compress the spherical harmonic function parameters specific to 3D Gaussian splatting,reducing network transmission bandwidth to 52% of the original data.This study provides a new technical pathway for fields requiring high-precision display,such as the digitization of cultural heritage.
