Article
Computer Science, Artificial Intelligence
Jie Pan, Jingwei Huang, Gengdong Cheng, Yong Zeng
Summary: This paper presents, implements, and evaluates a computational framework for automatic mesh generation based on reinforcement learning (RL). Mesh generation is crucial for numerical simulations in computer-aided design and engineering (CAD/E) and is considered a critical issue in the NASA CFD Vision 2030 Study. Existing mesh generation methods face challenges such as high computational complexity, low mesh quality in complex geometries, and limited speed. By formulating mesh generation as a Markov decision process (MDP), the paper applies a state-of-the-art RL algorithm called soft actor-critic to automatically learn the actions for mesh generation. The implementation of this RL algorithm enables the development of a fully automated mesh generation system without human intervention or additional clean-up operations, addressing the gaps in existing mesh generation tools.
Article
Computer Science, Software Engineering
Lorenzo Diazzi, Marco Attene
Summary: The study introduces a new technique for creating a mesh of convex polyhedra representing the interior volume of a triangulated input surface. This approach is highly tolerant to defects in the input and ensures that the input surface is accurately represented as the union of polygonal facets of the output volume mesh. By leveraging the concept of indirect geometric predicate, the algorithm provides a combination of robustness and speed that enables the practical implementation of robust yet flexible solid modeling systems. Extensive testing on the Thingi10k dataset has demonstrated that the method outperforms existing techniques in terms of robustness, precision, and performance.
ACM TRANSACTIONS ON GRAPHICS
(2021)
Article
Computer Science, Software Engineering
Hsueh-Ti Derek Liu, Alec Jacobson
Summary: This paper introduces a new method for stylizing 3D geometry by capturing surface normals to achieve different geometric styles. This method can deform 3D shapes into different styles within a single framework.
COMPUTER GRAPHICS FORUM
(2021)
Article
Computer Science, Software Engineering
Paul Zhang, Judy (Hsin-Hui) Chiang, Xinyi (Cynthia) Fan, Klara Mundilova
Summary: Hexahedral meshing is a well-studied topic in geometry processing, with various associated problems. In this work, we study the topology of singular nodes in hex meshes and show that all eight common singular nodes can be decomposed into singular curves. This decomposition technique reduces distortion and converts singular nodes into singular curves.
COMPUTER-AIDED DESIGN
(2023)
Article
Computer Science, Information Systems
Mingle Xu, Yongchae Jeong, Dong Sun Park, Sook Yoon
Summary: The paper proposes an enhanced class-specific spatial normalization method that generates photorealistic images by learning class-specific distributions and combining spatial-class maps. The method is applied in semantic image generation and object transfiguration, demonstrating competitive performance with fewer parameters.
Article
Computer Science, Software Engineering
Albert Jimenez-Ramos, Abel Gargallo-Peiro, Xevi Roca
Summary: This paper presents a nodal interpolation method for approximating a subdivision model, which is capable of modeling and representing curved geometry without gaps. It maintains sharp features and smooths indicated ones, and can handle unstructured configurations of simulation points, curves, and surfaces.
COMPUTER-AIDED DESIGN
(2022)
Article
Computer Science, Information Systems
Pavlo Yevtushenko, Leonid Goubergrits, Lina Gundelwein, Arnaud Setio, Heiko Ramm, Heiko Lamecker, Tobias Heimann, Alexander Meyer, Titus Kuehne, Marie Schafstedde
Summary: Image-based patient-specific modelling of hemodynamics has the potential to improve diagnostic capabilities and clinical outcomes for cardiovascular diseases, but traditional numerical methods require significant computational resources. Machine learning approaches offer the advantage of quickly calculating patient-specific hemodynamic outcomes with high accuracy, showcasing their ability to perform tasks that previously required resource-intensive simulations.
IEEE JOURNAL OF BIOMEDICAL AND HEALTH INFORMATICS
(2022)
Article
Computer Science, Artificial Intelligence
Chunyan Xu, Li Wei, Zhen Cui, Tong Zhang, Jian Yang
Summary: Meta-VOS is an online adaptive video object segmentation framework that learns to adapt to specific target segmentation, improving model learning and rapidly determining the best patterns for specific targets, updating positive/negative patterns based on learned expertise.
IEEE TRANSACTIONS ON IMAGE PROCESSING
(2021)
Article
Physiology
John D. Horn, Michael J. Johnson, Zbigniew Starosolski, Avner Meoded, Dianna M. Milewicz, Ananth Annapragada, Shaolie S. Hossain
Summary: This study aims to assess the stroke risk in pediatric MMD patients by analyzing local hemodynamics. The findings suggest that WSR-based metrics could be predictive of future stroke risk after an initial stroke in pediatric MMD patients. It is recommended to perform patient-specific hemodynamic analysis at multiple timepoints during patient follow-up to better predict and prevent secondary strokes.
FRONTIERS IN PHYSIOLOGY
(2022)
Article
Computer Science, Software Engineering
F. Zoccheddu, E. Gobbetti, M. Livesu, N. Pietroni, G. Cherchi
Summary: HexBox is an intuitive method and tool for creating and editing hexahedral meshes. It allows users to box-model a volumetric mesh by modifying its surface through various operations. It achieves efficiency by maintaining parallel data structures and utilizing recent advancements in grid-based meshing.
COMPUTER GRAPHICS FORUM
(2023)
Article
Biophysics
Simbarashe G. Chidyagwai, Madhurima Vardhan, Michael Kaplan, Reid Chamberlain, Piers Barker, Amanda Randles
Summary: This study used computational fluid dynamics simulations and surgical correction to investigate the impact of AAOCA on patient hemodynamics. The results showed that patients with AAOCA had higher TAWSS values and that coronary unroofing surgery normalized the TAWSS. These findings provide evidence for potential hemodynamic factors underlying coronary ischemia in AAOCA during exercise.
JOURNAL OF BIOMECHANICS
(2022)
Article
Multidisciplinary Sciences
Sheng Zeng, Guohua Geng, Hongjuan Gao, Mingquan Zhou
Summary: This paper proposes a method to parameterise the surface onto a novel geometry image, achieving shape preservation in the reconstructed 3D mesh and high reconstruction accuracy. Storing xyz positions in a 16-bit data format allows for preservation of topological structure without sampling errors.
SCIENTIFIC REPORTS
(2021)
Article
Construction & Building Technology
Wonseok Oh, Ryozo Ooka, Hideki Kikumoto, Mengtao Han
Summary: In this study, the boundary conditions for simulating cough airflow using computational fluid dynamics (CFD) were defined and validated using experimental data. The obtained boundary conditions can be used to investigate the risk of infection and transmission routes of human cough.
BUILDING AND ENVIRONMENT
(2022)
Article
Multidisciplinary Sciences
Moumen El-Melegy, Rasha Kamel, Mohamed Abou El-Ghar, Mohamed Shehata, Fahmi Khalifa, Ayman El-Baz
Summary: This paper proposes an automated and accurate DCE-MRI kidney segmentation method using fuzzy c-means clustering, Markov random field modeling, and a level set formulation. Experimental results demonstrate that the proposed method achieves high and consistent segmentation accuracy regardless of the initialization location of the level set contour.
SCIENTIFIC REPORTS
(2022)
Article
Computer Science, Interdisciplinary Applications
Rafat Damseh, Patrick Delafontaine-Martel, Philippe Pouliot, Farida Cheriet, Frederic Lesage
Summary: A novel graphing scheme is proposed for generating computational anatomical models of cerebrovascular networks, capturing vessel radii information and achieving iterative contraction under a Laplacian optimization framework. The proposed approach produces the lowest geometric and topological error rates on various angiograms, surpassing other techniques in providing representative models that capture all anatomical aspects of vascular structures.
IEEE TRANSACTIONS ON MEDICAL IMAGING
(2021)
Article
Biophysics
Rayanne Pinto Costa, Blaise Simplice Talla Nwotchouang, Junyao Yao, Dipankar Biswas, David Casey, Ruel McKenzie, David A. Steinman, Francis Loth
Summary: In this study, the critical Reynolds numbers for the transition to turbulence were experimentally determined for whole porcine blood and a Newtonian blood analog. The results showed that the transition to turbulence was delayed by approximately 19% for whole blood compared to the Newtonian fluid. This finding suggests that the rheological properties of blood may play a role in suppressing or delaying the onset of turbulence in vivo.
JOURNAL OF BIOMECHANICAL ENGINEERING-TRANSACTIONS OF THE ASME
(2022)
Article
Biology
Kristen M. Meiburger, Francesco Marzola, Guillaume Zahnd, Francesco Faita, Christos P. Loizou, Nolann Laine, Catarina Carvalho, David A. Steinman, Lorenzo Gibello, Rosa Maria Bruno, Ricarda Clarenbach, Martina Francesconi, Andrew N. Nicolaides, Herve Liebgott, Aurelio Campilho, Reza Ghotbi, Efthyvoulos Kyriacou, Nassir Navab, Maura Griffin, Andrie G. Panayiotou, Rachele Gherardini, Gianfranco Varetto, Elisabetta Bianchini, Constantinos S. Pattichis, Lorenzo Ghiadoni, Jose Rouco, Maciej Orkisz, Filippo Molinari
Summary: This study presents technical outlooks on computerized carotid intima-media thickness (CIMT) measurements and provides considerations for the development and comparison of these methods. A multi-center database of 500 images was collected and three manual segmentations and seven computerized methods were employed to measure the CIMT. The computerized methods showed comparable results with previous studies and demonstrated promising segmentation outcomes. The entire database has been made publicly available for further research and comparison.
COMPUTERS IN BIOLOGY AND MEDICINE
(2022)
Article
Biophysics
Daniel E. MacDonald, Mehdi Najafi, Lucas Temor, David A. Steinman
Summary: Recent studies have found high-frequency flow instabilities in aneurysms consistent with clinical reports. This study introduces a novel method for quantifying and visualizing these instabilities in cardiovascular computational fluid dynamics (CFD) data. The results show that a spectral bandedness index (SBI) is significantly associated with rupture status and could potentially be used to identify aneurysms at lower risk of rupture.
JOURNAL OF BIOMECHANICAL ENGINEERING-TRANSACTIONS OF THE ASME
(2022)
Article
Computer Science, Interdisciplinary Applications
Lorenzo Botti, Luca Verzeroli
Summary: In this work, we present a DG framework for nonlinear elasticity based on the Bassi-Rebay (BR2) formulation. The framework is applicable to both compressible and incompressible hyperelastic materials and can handle large deformations. To ensure stability, we combine higher-order lifting operators for BR2 stabilization with an adaptive stabilization strategy that relies on BR2 Laplace operator stabilization and a penalty parameter based on the spectrum of the fourth-order elasticity tensor. Dirichlet boundary conditions for displacement are enforced using Lagrange multipliers and the Nitsche method. Efficiency in the solution strategy is achieved using state-of-the-art agglomeration-based h-multigrid preconditioners, and the code implementation supports distributed memory execution on modern parallel architectures. Several benchmark test cases are proposed to investigate important computational aspects, such as the performance of the h-multigrid iterative solver with varying stabilization parameters and the influence of Dirichlet boundary conditions on Newton's method globalisation strategy.
JOURNAL OF COMPUTATIONAL PHYSICS
(2022)
Article
Neuroimaging
Daniel E. MacDonald, Nicole M. Cancelliere, Arianna Rustici, Vitor M. Pereira, David A. Steinman
Summary: This study presents a novel method called SURGE for extracting high curvature features from 3DRA images, which can reduce neck width estimation errors. Compared with traditional methods, SURGE provides a more accurate, faster, and more objective approach.
JOURNAL OF NEUROINTERVENTIONAL SURGERY
(2023)
Article
Cell Biology
Clara Alice Musi, Luca Colnaghi, Arianna Giani, Erica Cecilia Priori, Giacomo Marchini, Matteo Tironi, Claudio Conci, Giulio Cerullo, Roberto Osellame, Manuela Teresa Raimondi, Andrea Remuzzi, Tiziana Borsello
Summary: The complexity of the human brain has so far hindered the development of in vitro models. However, using the Nichoid microscaffold device to culture brain cells and generate brain micro-tissues provides a new tool to study brain cell interactions and morphology.
Article
Computer Science, Interdisciplinary Applications
Valentina Mazzi, Giuseppe De Nisco, Karol Calo, Claudio Chiastra, Joost Daemen, David A. Steinman, Jolanda J. Wentzel, Umberto Morbiducci, Diego Gallo
Summary: This study aimed to infer the near-wall transport of low-density lipoproteins (LDL) in arteries by computing wall shear stress (WSS). The results indicated that WSS topological skeleton features were more effective in identifying LDL concentration polarization profiles and showed promising correspondence with in vivo measured wall thickness growth in coronary arteries.
COMPUTER METHODS AND PROGRAMS IN BIOMEDICINE
(2022)
Article
Biophysics
Rayanne Pinto Costa, Blaise Simplice Talla Nwotchouang, Junyao Yao, Dipankar Biswas, David Casey, Ruel McKenzie, Frederick Sebastian, Rouzbeh Amini, David A. Steinman, Francis Loth
Summary: This study investigated the impact of wall compliance on the transition to turbulence in blood compared to a Newtonian fluid. The results showed that there were no significant differences in the wall vibration between whole blood and the Newtonian fluid, indicating that rheology had minimal impact on the vibration downstream of a stenosis for transition to turbulence.
JOURNAL OF BIOMECHANICAL ENGINEERING-TRANSACTIONS OF THE ASME
(2023)
Article
Cardiac & Cardiovascular Systems
Daniel E. MacDonald, Nicole M. Cancelliere, Vitor M. Pereira, David A. Steinman
Summary: The overestimation of intracranial aneurysm neck width has minimal impact on hemodynamics derived from computational fluid dynamics (CFD), with the differences being at most equal to or less than those caused by other sources of error/uncertainty in CFD analysis.
CARDIOVASCULAR ENGINEERING AND TECHNOLOGY
(2023)
Article
Biophysics
David A. Bruneau, Kristian Valen-Sendstad, David A. Steinman
Summary: Clinical, experimental, and computational studies have found that vibrations in cerebral aneurysms induced by blood flow instability could lead to wall deformation and cell behavior disruption. This study imposed increasing flow rates on realistic aneurysm models and observed narrow-band vibrations in two out of three geometries tested, with the strongest vibrations occurring when the fluid frequency band matched one of the aneurysm sac's natural frequencies. The study provides a potential mechanistic explanation for the high-frequency sounds in cerebral aneurysms and suggests that narrow-band flow might stimulate the aneurysm wall more effectively than broad-band, turbulent-like flow.
BIOMECHANICS AND MODELING IN MECHANOBIOLOGY
(2023)
Article
Mathematics, Applied
Luca Verzeroli, Francesco Carlo Massa, Lorenzo Botti
Summary: In this work, a Hybridizable Discontinuous Galerkin (HDG) formulation is introduced for Computational Contact Mechanics (CCM). The HDG formulation for nonlinear elasticity is extended to enforce boundary conditions and non-penetration frictionless contact constraints. The convergence of Newton's method is globalized by means of an incremental load method coupled with an active set strategy. The HDG formulation is numerically validated and applied to benchmark computations and blow molding simulations, demonstrating its effectiveness in contact mechanics problems.
FINITE ELEMENTS IN ANALYSIS AND DESIGN
(2023)
Article
Peripheral Vascular Disease
Martina Doneda, Sofia Poloni, Michela Bozzetto, Andrea Remuzzi, Ettore Lanzarone
Summary: This study used machine learning to predict blood flow volumes and vessel diameters of AVF, providing physicians with fast and accurate patient-specific predictions for AVF surgical planning decisions, aiming to improve the success rate of vascular access.
JOURNAL OF VASCULAR ACCESS
(2023)
Article
Radiology, Nuclear Medicine & Medical Imaging
Anna Caroli, Giulia Villa, Paolo Brambilla, Matias Trillini, Kanishka Sharma, Sandro Sironi, Giuseppe Remuzzi, Norberto Perico, Andrea Remuzzi
Summary: This study presents and validates a diffusion MRI (DWI)-based method for quantifying total kidney and cyst volume (TCV) and characterizing non-cystic tissue microstructure in autosomal dominant polycystic kidney disease (ADPKD). DWI shows potential in ADPKD to assess and predict disease progression, evaluate the impact of novel therapies, and identify microcysts and peritubular interstitial fibrosis in non-cystic kidney tissue.
EUROPEAN RADIOLOGY
(2023)
Article
Mathematics, Applied
Lorenzo Botti, Daniele A. Di Pietro
Summary: This study introduces a p-multilevel preconditioner for hybrid high-order discretizations of the Stokes equation and evaluates its performance on two variants of the method, comparing it with a classical discontinuous Galerkin scheme. The effects of h- and k-dependency, as well as different static condensation procedures on the efficiency of the solution strategy, are investigated. It is shown that the choice of static condensation options can impact the efficiency of the solution strategy, especially in the case of graded meshes.
COMMUNICATIONS ON APPLIED MATHEMATICS AND COMPUTATION
(2022)
Article
Engineering, Biomedical
Lucas Temor, Nicole M. Cancelliere, Daniel E. MacDonald, Peter W. Coppin, Vitor M. Pereira, David A. Steinman
Summary: Translating CFD data to DICOM series enables interactive visualization in standard radiological software. Our approach allows real-time modifications of transfer functions for representations of scalar isosurfaces, volumetric rendering, and tubular pathlines.
INTERNATIONAL JOURNAL OF COMPUTER ASSISTED RADIOLOGY AND SURGERY
(2022)