Dirk-Bartz-Prize

Permanent URI for this community

https://diglib.eg.org/handle/10.2312/2633056

Browse

Browsing Dirk-Bartz-Prize by Subject "Machine learning"

Now showing 1 - 2 of 2
  • No Thumbnail Available
    Item
    Transdisciplinary Visualization of Aortic Dissections
    (The Eurographics Association, 2023) Mistelbauer, Gabriel; Bäumler, Kathrin; Mastrodicasa, Domenico; Hahn, Lewis D.; Pepe, Antonio; Sandfort, Veit; Hinostroza, Virginia; Ostendorf, Kai; Schroeder, Aaron; Sailer, Anna M.; Willemink, Martin J.; Walters, Shannon; Preim, Bernhard; Fleischmann, Dominik; Raidou, Renata; Kuhlen, Torsten
    Aortic dissection is a life-threatening condition caused by the abrupt formation of a secondary blood flow channel within the vessel wall. Patients surviving the acute phase remain at high risk for late complications, such as aneurysm formation and aortic rupture. The timing of these complications is variable, making long-term imaging surveillance crucial for aortic growth monitoring. Morphological characteristics of the aorta, its hemodynamics, and, ultimately, risk models impact treatment strategies. Providing such a wealth of information demands expertise across a broad spectrum to understand the complex interplay of these influencing factors. We present results of our longstanding transdisciplinary efforts to confront this challenge. Our team has identified four key disciplines, each requiring specific expertise overseen by radiology: lumen segmentation and landmark detection, risk predictors and inter-observer analysis, computational fluid dynamics simulations, and visualization and modeling. In each of these disciplines, visualization supports analysis and serves as communication medium between stakeholders, including patients. For each discipline, we summarize the work performed, the related work, and the results.
  • Thumbnail Image
    Item
    X-ray simulations with gVirtualXray in medicine and life sciences
    (The Eurographics Association, 2025) Vidal, Franck P.; Afshari, Shaghayegh; Albiol, Alberto; Albiol, Francisco; Bellot, Alberto Corbí; Brun, Anna Louise; Chou, Chengy-Ying; Desbarats, Pascal; García, Marcos; Giovannelli, Jean-Francois; Hatton, Clémentine; Henry, Audrey; Kelly, Graham; Michelet, Claire; Mihail, Radu P.; Racy, Malek; Rouwane, Ali; Seznec, Herve; Sújar, Aarón; Tugwell-Allsup, Jenna; Villard, Pierre-Frédéric; Meuschke, Monique; Kuhlen, Torsten W.
    gVirtualXray (gVXR) is a programming interface framework to simulate realistic X-ray projections in realtime on graphics processing units (GPUs). It solves the Beer-Lambert law (attenuation law) using a deterministic X-ray simulation algorithm based on 3D computer graphics, namely rasterisation. Implemented as multi-pass rendering makes it more computationally optimal than the ray-tracing technique, which is a brute-force and straightforward approach to simulate X-ray images. Although written in C++ using OpenGL and its shading language (GLSL) to leverage the GPU, gVXR is available for other programming languages such as Python. Extensive validation studies, including comparisons with Monte Carlo simulations and real experimental data, have confirmed the accuracy of gVXR's simulations. gVXR was initially used in medical virtual reality (VR) for training purposes. It was then used in medical physics, and high-throughput data applications including mathematical optimisation and machine learning (ML). Micro-imaging studies on the C. elegans biological model are also reported.