idw – Informationsdienst Wissenschaft
Nachrichten, Termine, Experten
Nachrichten, Termine, Experten
idw - Informationsdienst
Wissenschaft
Not one but two researchers at the University of Bonn and the University Hospital Bonn are to receive much-sought-after Consolidator Grants. Awarded by the European Research Council (ERC), they provide millions of euros in funding for outstanding research projects. Professor Philipp Vollmuth is developing an AI foundation model that is expected to set new benchmarks for the use of AI in radiology, while Privatdozent Dr. med. Michael Sommerauer—who recently swapped the University of Cologne for Bonn—is researching the early detection of Parkinson’s disease.
“Receiving these prestigious grants marks a fantastic success for the Faculty of Medicine and the University Hospital Bonn,” says Professor Bernd Weber, Dean of the Faculty of Medicine and Acting CEO of the University Hospital Bonn. Something else that makes the achievement special, he reveals, is the fact that both his colleagues do clinical work alongside their research, thus demonstrating the feasibility of combining the two. “This is precisely what we are aiming to do in funding our Advanced Clinician Scientists, which enabled us to secure Michael Sommerauer’s services as part of the ACCENT program funded by the Federal Ministry of Education and Research (BMBF),” Weber explains. “In terms of their content, both projects are also a perfect fit for our research focus areas. Congratulations to both researchers.”
Foundation models: a key technology
Radiological imaging helps to diagnose diseases and monitor their course. While more and more of these tests are being done, time pressure in clinical treatment is also increasing. Professor Philipp Vollmuth, Director of the Division for Computational Radiology & Clinical AI (CCIBonn.ai) at the Neuroradiology Clinic and Co-Director of the Center for Medical Data Usability and Translation (ZMDT), wants to use artificial intelligence (AI) to tackle this challenge. He has been awarded an ERC Consolidator Grant worth €2.5 million over the next five years for his “AI-Next” project.
“With AI-Next, we’re striking off in a pioneering new direction,” Vollmuth says. “We’re developing an AI foundation model that’s designed to set new benchmarks for using AI in radiology and, just like an experienced physician, to develop an in-depth ‘understanding’ from a vast amount of image data.” The plan is for the model to learn to identify structures and patterns by itself in an extensive and varied pool of data comprising several million radiological images. “We’re focusing on brain imaging initially, because it’s a highly complex and data-intensive field,” the researcher explains. However, he points out, the methodology used—which is similar to language models like ChatGPT—could be applied to any branch of radiology.
Professor Vollmuth, a member of the Life & Health Transdisciplinary Research Area at the University of Bonn, is keen to explore every ounce of potential that these models offer, from improving image quality and using automation to identify and quantify critical findings—e.g. in the case of a stroke or brain hemorrhage—through to enabling more precise diagnosis of chronic conditions such as multiple sclerosis and Alzheimer’s. AI is also set to help improve predictions of disease progression in cancer patients and automate radiological reports. Within AI-Next, the researcher is building on his existing involvement in research initiatives looking at foundation models such as the Human Radiome Project (THRP) and is working closely with various partners at home and abroad, including the German Center for Neurodegenerative Diseases (DZNE) and the German Cancer Research Center (DKFZ).
Dr. Philipp Vollmuth has held an Else Kröner CS Professorship for Artificial Intelligence in Medical Imaging at the University Hospital Bonn since April. He previously worked at Heidelberg University Hospital, from where he also obtained his Habilitation. He has been a visiting professor at the University of California, San Francisco and the University of Ulsan College of Medicine in Seoul and completed an MBA at IE Business School in Madrid. He has also won numerous prizes endowed with funding to support his career.
Spotting Parkinson’s early
Parkinson’s disease usually escapes detection until a person is already showing signs of significant motor disorders and has lost a great many nerve cells. “By this point, it’s too late to pinpoint any treatments that could slow or stop the disease’s progression,” says Privatdozent Dr. med. Professor Michael Sommerauer. “This is why it’s so important to spot the condition earlier, i.e. before the patient develops symptoms that make it harder for them to do everyday things.” The “Re-Start PD” project is designed to help identify Parkinson’s early, gain a better understanding of its early stages and thus come up with new treatments. It is to receive nearly €2.5 million in funding from the ERC over the next five years.
Michael Sommerauer has come to Bonn on an Advanced Clinician Scientist scholarship from the University’s Faculty of Medicine. At the Clinic for Parkinson’s Disease, Sleep and Movement Disorders in the University Hospital Bonn’s Center for Neurology, he is in charge of the sleep laboratory, where his work focuses on sleep disorders as a warning sign for Parkinson’s. “REM sleep behavior disorder is a precursor to Parkinson’s in certain people,” Sommerauer says. To this end, he is also offering dedicated office hours for patients and, together with fellow physicians from the Netherlands and Austria, has secured over €1 million in EU funding to prevent Parkinson’s through increased movement and exercise. “So the ERC grant fits in really well with what I’ve been doing up to now,” the researcher adds. He is aiming to create an early-warning system for Parkinson’s complete with tablet app for large sections of the population, on which he is collaborating closely with colleagues from Bonn, Munich, Marburg, Oxford and Boston.
Michael Sommerauer spent the past six years working at the Neurology Clinic at the University Hospital Cologne, where he wrote his Habilitation thesis on sleep-wake disorders in cases of Parkinson’s disease and applied for the ERC Consolidator Grant that he has now been awarded in Bonn. After obtaining his medical degree from RWTH Aachen University, he went on to work at Aarhus University Hospital and the University Hospital Zurich.
ERC Consolidator Grants
The ERC awards its Consolidator Grants annually to fund excellent established researchers whose work demonstrates a high level of scientific quality on a par with their international peers. The grant is designed to bolster independent research teams that are already in place and consolidate their scientific work. See https://erc.europa.eu/apply-grant/consolidator-grant for more information.
Prof. Dr. Philipp Vollmuth
Division for Computational Radiology & Clinical AI (CCIBonn.ai)
University Hospital Bonn and
Center for Medical Data Usability and Translation (ZMDT)
University of Bonn
Phone +49 228 287-16507
Email: philipp.vollmuth@ukbonn.de
Internet: www.CCIBonn.ai
Privatdozent Dr. med. Michael Sommerauer
Center for Neurology
Clinic for Parkinson’s Disease, Sleep and Movement Disorders
University Hospital Bonn
University of Bonn
Phone +49 228 287-13091
Email: Michael.Sommerauer@ukbonn.de
Professor Philipp Vollmuth from the Division for Computational Radiology & Clinical AI (CCIBonn.ai) ...
Photo: Tobias Schwerdt
Privatdozent Dr. med. Michael Sommerauer from the Center for Neurology.
Photo: Private
Merkmale dieser Pressemitteilung:
Journalisten, jedermann
Medizin
überregional
Forschungsprojekte, Wettbewerbe / Auszeichnungen
Englisch
Sie können Suchbegriffe mit und, oder und / oder nicht verknüpfen, z. B. Philo nicht logie.
Verknüpfungen können Sie mit Klammern voneinander trennen, z. B. (Philo nicht logie) oder (Psycho und logie).
Zusammenhängende Worte werden als Wortgruppe gesucht, wenn Sie sie in Anführungsstriche setzen, z. B. „Bundesrepublik Deutschland“.
Die Erweiterte Suche können Sie auch nutzen, ohne Suchbegriffe einzugeben. Sie orientiert sich dann an den Kriterien, die Sie ausgewählt haben (z. B. nach dem Land oder dem Sachgebiet).
Haben Sie in einer Kategorie kein Kriterium ausgewählt, wird die gesamte Kategorie durchsucht (z.B. alle Sachgebiete oder alle Länder).
