PhD Student | HPC | Biomechanics
I’m a PhD Student at the High-Performance Computing Center Stuttgart (HLRS) and my research focuses on a constitutive model of human trabecular bone.
List of Publications (LoP)
Curriculum Vitae (CV)
My work includes the development of algorithms for parallel analyzes of computed tomography scans along with their in silico translation into mechanical models and parameters by direct numerical simulation. Patient specific effective material parameters for the mechanical simulation of bones are the key output of my doctoral research.
I’ve studied automotive and engine engineering and use this knowledge and my experience from academia and the industry for lectures in the field of computer simulations.
Bones withstand the mechanical stresses we are exposed to, while diseases and fractures may require the use of implants. Made out of metal, they often have a high modulus of elasticity, leading to an effect known as stress shielding. Such mechanostatic resorption can lead to aseptic loosening of the implant that does not reproduce the heterogeneity and anisotropy of cancellous bone.
My research accordingly relates to a constitutive model of the spongiosa as this tissue is called too. The development of the model leads to finite element analyses based on direct discretization of microfocus computed tomography images and subsequently to a software stack and a need of HPC resources that are not readily available. Continuing the research of Dr.-Ing. Ralf Schneider, the head of our department, I enhanced, developed and ported the software to our current HPC systems.
Simply put, the goal of my PhD addresses the need of improved mechanical descriptions of human cancellous bone and the use of high-performance computers in context of structural simulations.
Very large compute systems like HLRS Hawk are leading to new advances in science, but also raise questions about how to use their performance. Our research with real-world applications and data helps solve these challenges. About the HLRS systems.
HLRS Hawk from "inside".
Human femoral heads contain a large volume of cancellous tissue, get explanted during surgery for a total hip endoprosthesis and therefore are of special interest. With many thanks to Prof. Dr. Peter Helwig, head of trauma surgery and Dr. Jan-Peter Schenkengel, head of radiology, we have collected more than 30 human femoral heads for use in our research.
The bones are imaged using a clinical computed tomography (CT) scanner at Heidenheim. Later, we scan the samples in a microfocus CT scanner to obtain 3D images of the bone with a spatial resolution of about 0.005 mm. Freezing the bone samples at -80°C preserves their mechanical structure with all features down to 0.02 mm of expected wall thickness.
Delivering human femoral heads, frozen at -20°C to -80°C.
While HLRS does not operate a computed tomography scanner, SimTech does. I would like to express my sincere thanks to Prof. Dr.-Ing. Holger Steeb and Mr. Matthias Ruf from the Institute of Applied Mechanics (Civil Engineering). They supported our research by scanning our specimen.
Human Femoral Head 1 during its first computed tomography scan.
The tissue remains enclosed in the pathology vessel during the scans. Detailed information about the scanner is available online: SimTech computed tomography Scanner.
Part of my doctoral project is the development of the process-chain to read, prepare and process computed tomography images. The stack covers all steps from filtering and finite element analyzes to more diverse data manipulations to investigate specific properties of trabecular bone.
The stack is available on Github. Please be aware that some parts of the source codes are not yet available since they are subject to publications.
I expect the last bits and pieces to be online around August 2023.
Research is done by people to help other people understand the world we were born into. It is crucial to communicate work and results. Otherwise, the whole undertaking, spending of financial resources and time would be pointless. While some of the final theses we supervise are industrial ones with a business critical need for secrecy, my own research is open to public.
In addition to presenting results, I am actively engaged in organizing workshops and volunteer at conferences to support people getting together. Science can be communicated online without a need to meet in person. But since research is done by people, networking and strong personal relationships that are build in-person in my experience are the best way of driving science.
USTUTT | Interview | November 2022 | Stuttgart, Germany
Univ.-Prof. Dr.-Ing. Oliver Riedel invited five PhD students for an interview about doing a doctorate at the University of Stuttgart. The department of mechanical engineering of USTUTT is among the strongest in respect to their research activities and offers a strong curriculum. However many PhD student will move on and may be found in research or industry once they complete their dissertation. We used the opportunity of an interview for the magazine of the department of mechanical engineering at the Fraunhofer IAO for discussing issues relating to the undertaking of a PhD.
Is there anything better than fighting alone? Yes. Discussing with other people and taking good advice.
SC22 | Research Poster | November 2022 | Dallas, Texas, USA
The submission of a research poster marks the completion of my research internship at the Innovative Compting Laboratory (ICL) of the University of Tennessee, Knoxville which started in April 2022. We conducted research about computationally informed machine learning algorithms for use in high-performance computing. The results currently are in the process of publication.
Part of the entrance area of the Innovative Compting Laboratory (ICL). Meeting and working with those people exceeded even my optimistic expectations of what can be learned in a limited time frame.
SC22 | Exhibition | November 2022 | Dallas, Texas, USA
The Supercomputing not only is a high-value conference in High-Performance Computing, but it is a fare for Soft- and Hardware in HPC as well. HLRS presented its abilities in computing, visualization and its engagement in national and international collaborations. The Porsche 911 GTS was marked by my colleague Dr.-Ing. Uwe Wössner, head of Visualization, for overlaying the video stream by a camera with the results of a Computational Fluid Dynamics Simulation (CFD). It was a pleasure supporting HLRS on SC22.
Computing and presenting results of scientific applications is one of the key competences of HLRS.
WSSP34 | Presentation | October 2022 | Sendai, Japan
Prof. Kobayashi-sensei and his colleagues of the Cyberscience Center at the Tohoku University in Sendai hosted me and my colleagues from HLRS to join the Workshop on Sustained Simulation Performance. The talk aimed at performance enhancements like we opted for at the Innovative Computing Laboratory, but with a focus on Hawk, the flagship computer of HLRS. On the one hand, energy savings regarding my software-stack are a goal of my studies and on the other hand, other strategies for energy saving in respect to the whole supercomputer were part of the discussion.
The Workshop was held in Sendai, which is connected to Tokyo by the Shinkansen.
ICl/UTK | Research Internship | April-August 2022 | Knoxville, Tennessee, USA
The submission of a research poster marks the completion of my research internship at the Innovative Compting Laboratory (ICL) of the University of Tennessee, Knoxville which started in April 2022. We conducted research about computationally informed machine learning algorithms for use in high-performance computing. The results currently are in the process of publication.
Part of the entrance area of the Innovative Compting Laboratory (ICL). Meeting and working with those people exceeded even my optimistic expectations of what can be learned in a limited time frame.
WCCM-APCOM | Presentation | August 2022 | Online/Yokohama, Japan
Saving computational effort by orders of magnitudes always is a good thing. We achieve that by computing effective stiffness matrices with directly discretized irregular graphs. This proved being much faster than computing them based on directly discretized computed tomography scans. The speed up is a result of a significant reduction of degrees of freedom. This method can improve simulations of structures similar to foams, sponges or more heterogeneous ones. At its gists the structure must correspond to an irregular graph. In addition, investigations on the graph may show histomorphological measures otherwise unknown.
ESMC22 | Presentation | July 2022 | Galway, Ireland
I had the pleasure to resent our new approach to topologically optimize sub volumes of patient specific implants, thus whole implants in the long term. A first plastic based 3D-print visualized the idea for the audience. Discussions showed that by all the way still up front, the approach of optimizing implants is promising. In another step, assembling a structure topologically optimized by the corresponding field of reduced stiffness matrices will show the feasibility of bone-implant-systems.
The Quadrangle at the National University of Ireland, Galway. The first building of NUI Galway, inaugurated 1849 with 68 students.
PASC22 | Volunteering | June 2022 | Basel, Switzerland
Many branches of research profit by the use of high-performance computing. While many people work on computers to make them faster, many others try to exploit the computational power of such machines. PASC22 was an opportunity to learn from both categories; to improve my own understanding of their mindsets and requirements. The organization of conferences like PASC relies on many motivated volunteers and employees to ensure that no one will recognize the organizational efforts. I am glad for having the opportunity to support as such a volunteer and to get in touch with and to learn from those colleagues making PASC an event to enjoy.
Day 1 at the reception desk, welcoming the attendees.
WSSP33 | Organization | May 2022 | Stuttgart, Germany
The workshop was held at HLRS. With Prof. Jack Dongarra as our special guest and keynote speaker, we covered the range from doctoral students to the winner of the Turing Award 2021. It was a pleasure to organize this workshop and to coordinate our colleagues of HLRS who did a great job of hosting the scientific community.
We are in the process of publishing the contributions to the proceedings of the Workshops on Sustained Simulation Performance 31 and 32. The previous book is available at Springer.
WSSP32 | Presentation | December 2021 | Online/Sendai, Japan
High-performance computers are not build to serve themselves. Instead, workflows for research or product development are to be supported. My presentation at WSSP32 covered the components of my workflow for outlining the use of different hardware architectures and their specific benefits. Most parts of my workflow are uploaded for use in this repository.
University life combines teaching and research in a rewarding way. Students can learn about their field of interest. Research can pass on the latest findings and benefit from the students' ideas.
Exploiting the power of High-Performance Computers is a non trivial task, but also part of the infrastructure of many research disciplines. There is a need for trained students entering research at the interface of engineering, natural science and others for connecting those with HPC. To address this issue, I am responsible for teaching the lectures Parallelrechner - Architektur und Anwendung (Parallel Computers, Architecture and Application) at the University of Stuttgart.
Public lectures are part of our Project IKILeUS (“Integrated artificial intelligence in teaching at the University of Stuttgart”). With about 12 different speakers, we will shed a light on various aspects of artificial intelligence. IKILeUS was awarded to the University of Stuttgart in November 2021. Within three years, the joint venture will use artificial intelligence (AI) from half a dozen institutes. On the one hand, AI will improve teaching in various lectures. On the other hand, a series of lectures will provide the know-how needed to use the tool for data analysis in later professional life.
During two terms, we cover the basic aspects and mindset of modeling computer simulations. Contemporary research and development heavily rely on them. Money may be saved, the understanding of the subject gets deeper and sometimes, a real system could not even be investigated experimentally.
The lectures are called Modellierung, Simulation und Optimierungsverfahren 1/2 (MSO) and are organized at our e-learning solution Ilias.
Since I will be responsible for designing the new public lectures on artificial intelligence starting in 2022, the MSO has been passed on to my colleagues and I will only give a few of those lectures in the upcoming semesters.
HLRS and the Institute of High-Performance Computing are closely connected to the University of Stuttgart. Teaching and supervising students and student research assistants are an appreciated part of my work.
Feel free to contact me (gebert@hlrs.de) if you …
Some of the final theses we supervise are industrial ones with a business critical need for secrecy. We often can account for that with a non-disclosure agreement (NDA). Theses @ List of Publications .
Programs and tools for programming on HPC systems and improving workflows. Intended to support students and without claiming completeness.
This website was created in my personal capacity and contains links to the websites of third parties („external links“). As the content of these websites is not under my control, I cannot assume any liability for such external content. In all cases, the provider of information of the linked websites is liable for the content and accuracy of the information provided. At the point in time when the links were placed, no infringements of the law were recognizable to me. As soon as an infringement of the law becomes known to me, I will immediately remove the link in question.
The following sections consider german legal information. However, if you have any issue about anything on these webpages, feel free to contact me.
Verantwortlich für alle Belange im Rahmen dieser Webseiten, auch im Sinne der Datenschutzgesetze, insbesondere der EU-Datenschutzgrundverordnung (DSGVO), ist: Johannes Gebert
Johannes Gebert Nobelstraße 19 70569 Stuttgart Germany
gebert@hlrs.de
Die Inhalte unserer Seiten wurden mit größter Sorgfalt erstellt. Für die Richtigkeit, Vollständigkeit und Aktualität der Inhalte können wir jedoch keine Gewähr übernehmen. Als Diensteanbieter sind wir gemäß § 7 Abs.1 TMG für eigene Inhalte auf diesen Seiten nach den allgemeinen Gesetzen verantwortlich. Nach §§ 8 bis 10 TMG sind wir als Diensteanbieter jedoch nicht verpflichtet, übermittelte oder gespeicherte fremde Informationen zu überwachen oder nach Umständen zu forschen, die auf eine rechtswidrige Tätigkeit hinweisen. Verpflichtungen zur Entfernung oder Sperrung der Nutzung von Informationen nach den allgemeinen Gesetzen bleiben hiervon unberührt. Eine diesbezügliche Haftung ist jedoch erst ab dem Zeitpunkt der Kenntnis einer konkreten Rechtsverletzung möglich. Bei Bekanntwerden von entsprechenden Rechtsverletzungen werden wir diese Inhalte umgehend entfernen.
Unser Angebot enthält Links zu externen Webseiten Dritter, auf deren Inhalte wir keinen Einfluss haben. Deshalb können wir für diese fremden Inhalte auch keine Gewähr übernehmen. Für die Inhalte der verlinkten Seiten ist stets der jeweilige Anbieter oder Betreiber der Seiten verantwortlich. Die verlinkten Seiten wurden zum Zeitpunkt der Verlinkung auf mögliche Rechtsverstöße überprüft. Rechtswidrige Inhalte waren zum Zeitpunkt der Verlinkung nicht erkennbar. Eine permanente inhaltliche Kontrolle der verlinkten Seiten ist jedoch ohne konkrete Anhaltspunkte einer Rechtsverletzung nicht zumutbar. Bei Bekanntwerden von Rechtsverletzungen werden wir derartige Links umgehend entfernen.
Die durch die Seitenbetreiber erstellten Inhalte und Werke auf diesen Seiten unterliegen dem deutschen Urheberrecht. Die Vervielfältigung, Bearbeitung, Verbreitung und jede Art der Verwertung außerhalb der Grenzen des Urheberrechtes bedürfen der schriftlichen Zustimmung des jeweiligen Autors bzw. Erstellers. Downloads und Kopien dieser Seite sind nur für den privaten, nicht kommerziellen Gebrauch gestattet. Soweit die Inhalte auf dieser Seite nicht vom Betreiber erstellt wurden, werden die Urheberrechte Dritter beachtet. Insbesondere werden Inhalte Dritter als solche gekennzeichnet. Sollten Sie trotzdem auf eine Urheberrechtsverletzung aufmerksam werden, bitten wir um einen entsprechenden Hinweis. Bei Bekanntwerden von Rechtsverletzungen werden wir derartige Inhalte umgehend entfernen.
Die Nutzung unserer Webseite ist in der Regel ohne Angabe personenbezogener Daten möglich. Soweit auf unseren Seiten personenbezogene Daten (beispielsweise Name, Anschrift oder eMail-Adressen) erhoben werden, erfolgt dies, soweit möglich, stets auf freiwilliger Basis. Diese Daten werden ohne Ihre ausdrückliche Zustimmung nicht an Dritte weitergegeben. Wir weisen darauf hin, dass die Datenübertragung im Internet (z.B. bei der Kommunikation per E-Mail) Sicherheitslücken aufweisen kann. Ein lückenloser Schutz der Daten vor dem Zugriff durch Dritte ist nicht möglich. Der Nutzung von im Rahmen der Impressumspflicht veröffentlichten Kontaktdaten durch Dritte zur Übersendung von nicht ausdrücklich angeforderter Werbung und Informationsmaterialien wird hiermit ausdrücklich widersprochen. Die Betreiber der Seiten behalten sich ausdrücklich rechtliche Schritte im Falle der unverlangten Zusendung von Werbeinformationen, etwa durch Spam-Mails, vor.
Website Impressum von impressum-generator.de
Introduction theme for Hugo. Made by open source contributors.