Research output
Publications.
Geometric learning, neural surrogates, and medical imaging. Each work has its own page, with links to the paper and available research materials.
Co-authorship
Publications by country
18 papers · 16 countries
Countries represented in the papers’ author affiliations. Each paper counts once per country.
Hover or tap for institutions. Select a country to filter the publications.
Institution Papers
A paper may involve several institutions.
All countries
View publicationsAffiliations, sources & counting
Counts cover the 18 papers and book chapters with verified affiliations, including preprints and credited consortium contributions. All authors’ affiliations are included. A paper counts once per country and once per institution, regardless of how many authors list it. Departments and campus locations are grouped under their parent institution; separately named research centres are retained. The country totals stay the same when searching or filtering the publication list.
Affiliations come from the linked paper versions. Where a country is omitted, the named institution’s location is used. Datasets and the single-author thesis are excluded. The ICML Topological Deep Learning Challenge report is also excluded because it lists contributor roles without institutional affiliations.
- Geometric deep learning for local growth prediction on abdominal aortic aneurysm surfacesarXiv v2, author affiliations · Author block before the abstract (PDF page 1)
Netherlands: Amsterdam Cardiovascular Sciences; Amsterdam UMC; University of Twente.
- When Does Equivariance Help? Canonical Alignment in Neural Fluid SurrogatesarXiv v2, author affiliations · Author block before abstract (PDF page 1)
Netherlands: Amsterdam Cardiovascular Sciences; Amsterdam UMC; University of Twente.
Germany: Munich Center for Machine Learning; Technical University of Munich.
- Active Learning for Deep Learning-Based Hemodynamic Parameter EstimationPublisher author affiliations · Author information / Authors and Affiliations
Netherlands: Amsterdam Cardiovascular Sciences; Amsterdam UMC; University of Twente.
- Enhancing clinical decision making in AAA diagnosis and treatment with AI: how can it help with prediction of AAA growth, endoleaks and spinal cord ischemiaSICVE 2024 book, contributor affiliations · Contributor list on PDF pages 8–11, cross-checked against chapter 4.3 author list on PDF page 131 (printed page 116)
Netherlands: Amsterdam UMC; University of Twente.
- Beyond Pixels: Medical Image Quality Assessment with Implicit Neural RepresentationsPublisher author affiliations · Author information / Authors and Affiliations
Netherlands: University of Twente.
Türkiye: Istanbul Technical University.
- CenterlinePointNet++: A New Point Cloud Based Architecture for Coronary Artery Pressure Drop and vFFR EstimationPublisher author affiliations · Author information / Authors and Affiliations
Poland: Hemolens Diagnostics; Wrocław University of Science and Technology.
- A Core Outcome Set for Intact Abdominal Aortic Aneurysms Under SurveillancePublished article, University of Helsinki repository · Printed page 275 (PDF page 2), author affiliations 1–16; collaborator affiliations on printed pages 286–288 (PDF pages 13–15)
Netherlands: ALLAI; Amsterdam Cardiovascular Sciences; Amsterdam Gastroenterology and Metabolism; Amsterdam Public Health; Amsterdam UMC; Radboud University Medical Centre; University of Amsterdam; University of Twente.
Serbia: University Clinical Center of Serbia; University of Belgrade; Vinča Institute of Nuclear Sciences.
Portugal: São João University Hospital; University of Porto.
United Kingdom: University of Oxford.
Finland: Helsinki University Hospital; University of Helsinki.
Germany: Asklepios Clinic Wandsbek; Asklepios Medical School; CCB Cardioangiologic Center Bethanien; University of Mainz.
France: CNRS; Hospital of Antibes Juan-les-Pins; Nice University Hospital; Université Côte d’Azur.
Norway: Haukeland University Hospital; University of Bergen.
- Coronary Ostia Localization Using Residual U-Net with Heatmap Matching and 3D DSNTPublisher author affiliations · Author information / Authors and Affiliations
Poland: Hemolens Diagnostics; Wrocław Medical University; Wrocław University of Science and Technology.
- Deep vectorised operators for pulsatile hemodynamics estimation in coronary arteries from a steady-state priorarXiv v2, author affiliations · Author block before the abstract
Netherlands: University of Twente.
Italy: Centro Cardiologico Monzino IRCCS; Politecnico di Milano; University of Milan.
- Eigenvector Grouping for Point Cloud Vessel LabelingPublished paper (PDF) · Page 1, author affiliations
Poland: Hemolens Diagnostics; Tooploox; Wrocław University of Science and Technology.
- GReAT: Leveraging Geometric Artery Data to Improve Wall Shear Stress AssessmentPublisher author affiliations · Author information / Authors and Affiliations
Germany: Munich Center for Machine Learning; Technical University of Munich.
United Kingdom: Imperial College London.
Netherlands: Erasmus Medical Center; University of Twente.
- Hemodynamic Neural Field: Continuous Blood Flow Estimation in Vessel Geometries via Global Shape ConditioningCamera-ready paper (PDF) · Page 1, author affiliations 1–3
France: CNRS; Nantes Université; École Centrale Nantes.
Netherlands: University of Twente.
- Learning hemodynamic scalar fields on coronary artery meshes: A benchmark of geometric deep learning modelsarXiv v2, author affiliations · Author block before abstract; same country set in indexed final published PDF, page 1
Italy: Centro Cardiologico Monzino IRCCS; Politecnico di Milano; University of Milan.
Netherlands: Amsterdam UMC; University of Amsterdam; University of Twente.
- Neural Fields for Continuous Periodic Motion Estimation in 4D Cardiovascular ImagingPublisher author affiliations · Author information / Authors and Affiliations
Italy: Fondazione Monasterio; University of Pisa.
Netherlands: University of Twente.
- Application of Fully Convolutional Neural Networks in the Assessment of Cerebral White Matter Involvement in Primary Sjögren’s SyndromePublished article, Authors and Affiliations · Author information → Authors and Affiliations (11 affiliation entries)
Poland: 4th Military Clinical Hospital, Wrocław; Hetalox; Pomeranian Medical University in Szczecin; Wrocław Medical University; Wrocław University of Science and Technology.
Netherlands: University of Twente.
- TopoBench: A Framework for Benchmarking Topological Deep LearningPublished paper (PDF) · Page 1, all 28 numbered author affiliations
Italy: Istituto Italiano di Tecnologia; Sapienza University of Rome.
United States: Atalaya Capital Management; Columbia University; Harvard University; Northeastern University; University of California, Santa Barbara; University of Oregon; University of Puget Sound; University of Rochester; University of South Florida; University of Texas Rio Grande Valley.
Netherlands: Delft University of Technology; Tilburg University; University of Twente; Vrije Universiteit Amsterdam.
Switzerland: University of Fribourg.
Portugal: Instituto Superior Técnico.
Germany: Perelyn GmbH; RWTH Aachen University.
Denmark: University of Copenhagen.
Spain: Barcelona Supercomputing Center; Catalan Institution for Research and Advanced Studies; Nostrum Biodiscovery; Universitat Politècnica de Catalunya.
Finland: Aalto University.
Poland: Adam Mickiewicz University.
Greece: PolyShape.
- Global Control for Local SO(3)-Equivariant Scale-Invariant Vessel SegmentationPublisher author affiliations · Author information / Authors and Affiliations
Netherlands: Amsterdam Cardiovascular Sciences; Amsterdam UMC; University of Twente.
- Wall Shear Stress Estimation in Abdominal Aortic Aneurysms: Towards Generalisable Neural Surrogate ModelsarXiv v1, author affiliations · Author block before the abstract
Netherlands: Amsterdam Cardiovascular Sciences; Amsterdam UMC; University of Twente.
Germany: Munich Center for Machine Learning; Technical University of Munich.
Affiliations checked October 2026. Map: Natural Earth, public domain.
21 research outputs
No publications match these filters. Try a different search or reset the filters.
2026
A Core Outcome Set for Intact Abdominal Aortic Aneurysms Under Surveillance
Patryk Rygiel · VASCUL-AID and Expert Consensus Meeting collaborator
Active Learning for Deep Learning-Based Hemodynamic Parameter Estimation
2025 conference · 2026 proceedings
Geometric deep learning for local growth prediction on abdominal aortic aneurysm surfaces
Beyond Pixels: Medical Image Quality Assessment with Implicit Neural Representations
2025 conference · 2026 proceedings
Application of Fully Convolutional Neural Networks in the Assessment of Cerebral White Matter Involvement in Primary Sjögren’s Syndrome
Hemodynamic Neural Field: Continuous Blood Flow Estimation in Vessel Geometries via Global Shape Conditioning
GReAT: Leveraging Geometric Artery Data to Improve Wall Shear Stress Assessment
2025 conference · 2026 proceedings
2025
Deep vectorised operators for pulsatile hemodynamics estimation in coronary arteries from a steady-state prior
Learning hemodynamic scalar fields on coronary artery meshes: A benchmark of geometric deep learning models
Wall Shear Stress Estimation in Abdominal Aortic Aneurysms: Towards Generalisable Neural Surrogate Models
Global Control for Local SO(3)-Equivariant Scale-Invariant Vessel Segmentation
2024 conference · 2025 proceedings
Neural Fields for Continuous Periodic Motion Estimation in 4D Cardiovascular Imaging
2024 conference · 2025 proceedings
TopoBench: A Framework for Benchmarking Topological Deep Learning
* Equal contribution
2024
Enhancing clinical decision making in AAA diagnosis and treatment with AI: how can it help with prediction of AAA growth, endoleaks and spinal cord ischemia
ICML Topological Deep Learning Challenge 2024: Beyond the Graph Domain
2023
CenterlinePointNet++: A New Point Cloud Based Architecture for Coronary Artery Pressure Drop and vFFR Estimation
2022
Coronary Ostia Localization Using Residual U-Net with Heatmap Matching and 3D DSNT
* Equal contribution
Eigenvector Grouping for Point Cloud Vessel Labeling
Datasets
AAA-100: A Curated Dataset of 3D Watertight Abdominal Aortic Aneurysm Models
A curated collection of watertight three-dimensional aneurysm models provides reusable vascular geometry for computational research.
Theses
Estimation of FFR in coronary arteries with deep learning
My MSc thesis studies point-cloud neural networks as surrogates for coronary fractional flow reserve estimation.
Applied research
Patent applications
WO2024184690A1 · 2024
A method of training an artificial deep neural network for estimation of hemodynamic parameter, a method of estimation of hemodynamic parameter, computer program products and computer systems
Tomasz Konopczyński, Maciej Zięba, Patryk Rygiel, Paweł Płuszka
Patent application
WO2023166330A1 · 2023
A method of detection of a landmark in a volume of medical images
Tomasz Konopczyński, Patryk Rygiel, Miłosz Gajowczyk, Piotr Gródek, Łukasz Moroz, Bartłomiej Ciupał
Patent application