Reconstruction of 4D spatio-temporal phenomena  Page description

Help  Print 
Back »

 

Details of project

 
Identifier
124124
Type K
Principal investigator Szirmay Kalos, László
Title in Hungarian 4D rendszerek rekonstrukciója
Title in English Reconstruction of 4D spatio-temporal phenomena
Keywords in Hungarian szimuláció, rekonstrukció, Monte Carlo módszer, tomográfia
Keywords in English simulation, reconstrunction, Monte Carlo method, tomography
Discipline
Information Technology (Council of Physical Sciences)100 %
Ortelius classification: Applied informatics
Panel Informatics and Electrical Engineering
Department or equivalent Department of Control Engineering and Information Technology (Budapest University of Technology and Economics)
Participants Csébfalvi, Balázs
Magdics, Milán
Szécsi, László
Tóth, Balázs
Tóth, Márton József
Umenhoffer, Tamás
Starting date 2017-09-01
Closing date 2023-06-30
Funding (in million HUF) 28.452
FTE (full time equivalent) 19.33
state running project





 

Final report

 
Results in Hungarian
A kutatási eredmények felsorolása: - A kinetikus modellillesztési feladat robosztus megoldása iteratív PET rekonstrukció inicializálása során. - Új, a bemeneti véraktivitást nem igénylő módszer, amely kompartment modellezésen és a Feng bemeneti véraktivitás modellen alapul. - PET rekonstrukció javítása utólagos szűréssel. - Detektorkristályokon belüli szóródás modellezése a DOI (mélységi elnyelődés) alapú PET rekonstrukcióhoz. - Mozgáskompenzációs módszerek PET vizsgálatokhoz. - Mélytanuló eljárások alkalmazása orvosi képek szegmentálásához és automatikus hólyagrák diagnózishoz. - Nem-analóg módszerek kifejlesztése többszörös szóródás szimulációjához. - Adaptív többszörös fontosság szerinti mintavételezés általánosítása és összekapcsolás a statisztika divergenciákkal, végül pedig hatékony optimalizálás a kritérium linearizálásával. - Fontosság szerinti mintavételi eljárások a fotorealisztikus és illusztratív képszintézisben. - Skalármező interpoláció és gradiens számítás a grafikus hardver lehetőségeivel. - Játék és szimulációs motorok általánosítása az euklideszi geometrián túl elliptikus és hiperbolikus geometriára. - A duális számok általánosítása magasabb rendű automatikus deriváláshoz. - Módszerek nem konvex optimalizáláshoz.
Results in English
Our contributions are the following: - Robust solution of the kinetic model fitting problem both in the initial phase and during the ML–EM iteration. - Method based on compartmental modeling and the Feng blood input function that does not require the blood input function and the fraction of blood parameters explicitly. - Post-processing filters for PET. - Modeling of inter-crystal scattering into DOI-based PET reconstructions. - Compensation of patient movements during PET scans. - Application of deep learning techniques for medical image segmentation and diagnosis of bladder tumors. - Establishing a non-analog approach for multiple scattering participating media rendering. - Generalizations of adaptive multiple importance sampling, explanation with statistical divergences, and solution via linearization of the equations. - Application of importance sampling methods in both physically based and non-photorealistic rendering. - Scalar field interpolation and gradient calculation with the sophisticated exploitation of the graphics hardware. - Generalization of game engine pipelines for elliptic and hyperbolic geometries. - Generalization of dual numbers for higher order automatic derivative computations. - Forced convexication for numerical optimization.
Full text https://www.otka-palyazat.hu/download.php?type=zarobeszamolo&projektid=124124
Decision
Yes





 

List of publications

 
Szirmay-Kalos L., Magdics M., Varnyú D.: Direct dynamic tomographic reconstruction without explicit blood input function, BIOMEDICAL SIGNAL PROCESSING AND CONTROL 80: 104313, 2023
Csébfalvi Balázs: One Step Further Beyond Trilinear Interpolation and Central Differences: Triquadratic Reconstruction and its Analytic Derivatives at the Cost of One Additional Texture Fetch, COMPUTER GRAPHICS FORUM 42 : 2 pp. 191-200., 2023
Sbert Mateu, Szirmay-Kalos László: A linear heuristic for multiple importance sampling, EURASIP JOURNAL ON ADVANCES IN SIGNAL PROCESSING 2023: (1) 31, 2023
László Szirmay-Kalos, Dóra Varnyú, András Fridvalszky, László Szécsi: Virtual Reality Simulation of Surgical Procedures, Grafgeo, 2022
Milán Magdics, Márton Tóth and Balázs Tóth: CT-based training data generation for DL-based X-ray image processing, Grafgeo, 2022
Balázs Csébfalvi: A Practical Application of Optimal Regular Volume Sampling, Grafgeo, 2022
András Fridvalszky, László Szécsi: Efficient Surface Remeshing on the GPU, Grafgeo, 2022
Viktória Burkus, Attila Kárpáti, László Szécsi: Ray Marching Fluids near Porous Surfaces, Grafgeo, 2022
Dóra Varnyú, László Szirmay-Kalos: Deep learning-based regularization of PET reconstruction, Grafgeo, 2022
Balázs Csébfalvi: A Practical Application of Optimal Regular Volume Sampling, Grafgeo, 2022
László Szirmay-Kalos, Milán Magdics, Mateu Sbert: Multiple Scattering in Inhomogeneous Participating Media Using Rao-Blackwellization and Control Variates, COMPUTER GRAPHICS FORUM, Vol 37, No 2, 2018
Tamás Umenhoffer, László Szirmay-Kalos, László Szécsi, Zoltán Lengyel, Gábor Marinov: An image-based method for animated stroke rendering, The Visual Computer, Volume 34, Issue 6–8, pp 817–827, 2018
Mateu Sbert, Vlastimil Havran, László Szirmay-Kalos, Víctor Elvira: Multiple importance sampling characterization by weighted mean invariance, The Visual Computer, June 2018, Volume 34, Issue 6–8, pp 843–852, 2018
Mateu Sbert, Vlastimil Havran, Laszlo Szirmay-Kalos: Multiple importance sampling revisited: breaking the bounds, EURASIP Journal on Advances in Signal Processing, December 2018, 2018
Tóth Péter: On the LOR-driven approach of back-projection during PET reconstruction, Ninth Hungarian Conference on Computer Graphics and Geometry, 2018
László Szirmay-Kalos, László Szécsi: Improved stratification for Metropolis light transport, Computers & Graphics, Volume 68, November 2017, Pages 11-20, 2017
Ágota Kacsó, Balázs Tóth, László Szirmay-Kalos: Improving the Robustness of Compartmental Model Fitting for Emission Tomography, IX. MAGYAR SZÁMÍTÓGÉPES GRAFIKA ÉS GEOMETRIA KONFERENCIA. Konferencia helye, ideje: Budapest, Magyarország, 2018.03.21-2018.03.22. Budapest: Neumann János Számítógép-tudo, 2018
B Csébfalvi: Fast Catmull‐Rom Spline Interpolation for High‐Quality Texture Sampling, Computer Graphics Forum 37 (2), 455-462, 2018
GF Rácz, B Csébfalvi: Cosine‐Weighted B‐Spline Interpolation on the Face‐Centered Cubic Lattice, Computer Graphics Forum 37 (3), 503-511, 2018
Péter Tóth: The noisy nature of PET-reconstruction at the edge of the voxel volume, KÉPAF, 2019
László Szirmay-Kalos, Dóra Varnyú: Model-based motion compensation for dynamic PET, KÉPAF, 2019
Dóra Varnyú, László Szirmay-Kalos: Thick line integration with filtered sampling, KÉPAF, 2019
Balázs Csébfalvi: Beyond Trilinear Interpolation: Higher Quality for Free., ACM Transactions on Graphics, 2019
László Szirmay-Kalos, Ágota Kacsó, Milán Magdics, Balázs Tóth: Dynamic PET Reconstruction on the GPU, Periodica Polytechnica Electrical Engineering and Computer Science, 62(4), pp. 134-143. doi: https://doi.org/10.3311/PPee.11739., 2018
M Sbert, V Havran, L Szirmay-Kalos: Optimal Deterministic Mixture Sampling, Eurographics Conference, Short papers, 2019
Dóra Varnyú and László Szirmay-Kalos: Comparison of Non-Linear Filtering Methods for Positron Emission Tomography, Infocommunications Journal, June 2020 Volume XII Number 2 ISSN 2061-2079, 2020
Szirmay-Kalos, Laszlo: Kinetic Model Fitting with Forced Convexification, In: 2019 IEEE Nuclear Science Symposium and Medical Imaging Conference (NSS/MIC) Piscataway (NJ), Amerikai Egyesült Államok : IEEE Press, (2019), 2019
Szirmay-Kalos László: On Higher Order Automatic Differentiation, Proceedings of the Workshop on the Advances of Information Technology (WAIT), 2020, 2020
Magdics Milán, Szirmay-Kalos László, Tóth Balázs, Szécsi László: Depth of Interaction Modeling for PET, Proceedings of the Workshop on the Advances of Information Technology (WAIT), 2020, pp 79-88, 2020
László Szirmay-Kalos, Milán Magdics, Dóra Varnyú, and Balázs Tóth: Modeling of Depth of Interaction with Inter-crystal Scattering for PET Reconstruction, 2020 IEEE Nuclear Science Symposium and Medical Imaging Conference, 2020
Fridvalszky, András and Tóth, Balázs: Multisample Anti-aliasing in Deferred Rendering, Eurographics 2020 - Short Papers, 2020
László Szirmay-Kalos; Ágota Kacsó; Milán Magdics; Balázs Tóth: Robust Compartmental Model Fitting in Direct Emission Tomography Reconstruction, VISUAL COMPUTER (0178-2789 1432-2315), 2021
Szirmay-Kalos László: De-raining and de-snowing, Proceedings of the Workshop on the Advances of Information Technology (WAIT) 2021, 2021
László Szirmay-Kalos, Milán Magdics: Euclidean to non-Euclidean Transformation of Games, Proceedings of the Workshop on the Advances of Information Technology (WAIT) 2021, 2021
László Szirmay-Kalos ; Milán Magdics: Gaming in Elliptic Geometry, Eurographics Short Papers, 2021
László Szirmay-Kalos; Márton Tóth: Bregman Approach to Single Image De-Raining, Eurographics Short Papers, 2021
László Szirmay-Kalos: Higher Order Automatic Differentiation with Dual Numbers, Periodica Polytechnica Electrical Engineering and Computer Science, 65(1), pp. 1–10, 2021, 2021
Dóra Varnyú, László Szirmay-Kalos: Determination of the Blood Input Function for Dynamic Emission Tomography with Deep Learning, Proceedings of the Workshop on the Advances of Information Technology (WAIT) 2021, 2021
László Szirmay-Kalos, Milán Magdics: Adapting Game Engines to Curved Spaces, VISUAL COMPUTER (0178-2789 1432-2315): (2022), 2022
Magdics Milán;  Szirmay-Kalos László; Varnyú Dóra: Direct Tomographic Reconstruction without Explicit Blood Input Function, WAIT 22, 2022
L. Szirmay-Kalos, M. Sbert: Robust Sample Budget Allocation for MIS., Eurographics, Short papers,, 2022
Mateu Sbert; László Szirmay-Kalos: Robust Multiple Importance Sampling with Tsallis φ-Divergences, Entropy, 2022
V. Burkus, A. Kárpáti, G. Klár, L. Szécsi: Real-time Sponge and Fluid Simulation, Eurographics 2022 - Short Papers, 2022
V. Burkus, A. Kárpáti, L. Szécsi: A Simple BRDF Model with Easy Importance Sampling, Proceedings of the Workshop on the Advances in Information Technology, 2022
V. Burkus, A. Kárpáti, L. Szécsi: Vector Coupled Map Lattice for Generating Directional Samples, Proceedings of the Workshop on the Advances in Information Technology, 2022
Fridvalszky, András; Tóth, Balázs: Accounting for Non-Uniform Ambient Lighting in Ray Traced Ambient Occlusion, PERIODICA POLYTECHNICA-ELECTRICAL ENGINEERING AND COMPUTER SCIENCE 66 : 3 pp. 248-252., 2022




Back »