AI supports targeted therapy recommendations for tumor diseases – BMBF funds joint research project at University Medical Center Mainz with 2.53 million Euros
The treatment and support of patients with advanced tumor diseases is a complex task involving numerous disciplines. For the interdisciplinary medical exchange, there are so-called tumor boards, in which the physicians discuss the treatment strategy. For this purpose, a large amount of medical data is available for each individual patient, which must be evaluated individually – a mammoth task. Digital methods based on artificial intelligence are ideal for analyzing the comprehensive and complex patient data and recommending a targeted therapy. Individualized tumor therapy increases treatment quality and reduces risks and side effects.
The Department of Urology and Pediatric Urology at University Medical Center Mainz has launched the new collaborative project "AI-supported Clinical Decision in Urologic Oncology (german: KI-unterstützte Therapiebegleitung am Beispiel der Urologie”, or KITTU for short). Together with the German Research Center for Artificial Intelligence (DFKI) in Kaiserslautern and Innoplexus AG in Frankfurt am Main, the experts want to develop an AI assistance system for the treatment of urological tumor diseases. The goal is to identify the optimal treatment using artificial intelligence and thus support physicians and patients in making the best treatment decisions. The overall objective is to optimize oncological treatment by recommending an individualized and evidence-based therapy for each patient. The German Federal Ministry of Education and Research (BMBF) is funding the project with 2.53 million Euros over a period of three years.
"We are very pleased that KITTU will receive research funding of 2.53 million euros from the German Federal Ministry of Education and Research (BMBF) for the first three years. This gives us the chance to do pioneering scientific work in the field of digital and AI-based tumor therapy and to further advance the digitalization of healthcare internationally. We are confident that, together with leading partners in the field of artificial intelligence, we will be able to achieve translational successes and lead tumor therapy into a digital future," says Univ.-Prof. Dr. Ulrich Förstermann, Scientific Director and Dean of the University Medical Center Mainz, about the KITTU project.
"KITTU is designed in such a way that the intended results, if successfully developed, can later be directly implemented in routine patient care. The benefits of such a software platform would be immense, both for patients and their physicians: because through an envisaged central digital hub, we hope to be able to further improve interdisciplinary cooperation between the treating physicians in the long term, as well as to reduce the necessary effort for administrative activities," explains Dr. Gregor Duwe, project coordinator and physician at the Department of Urology and Pediatric Urology at the University Medical Center Mainz.
In order to develop an AI software that can derive independent recommendations from patient data, a specially defined and specifically trained AI algorithm is required. Decisions of past tumor boards as well as relevant clinical studies are available as a primary source of information for the establishment of such an AI algorithm.
"In addition, our oncology experts weigh the relevance of the clinical decisions in order to equip the AI system, also known as a 'knowledge graph', with the ability to justify therapy recommendations. These then serve as a kind of second opinion, so to speak, to support those involved in selecting the appropriate therapy," explains Prof. Dr. Dr. Jürgen Scheele, Chief Medical Officer of Innoplexus AG. And adds, "We are very pleased to be entering innovative new territory in patient-centered medical research together with the Department of Urology and Pediatric Urology at University Medical Center Mainz and the German Research Center for Artificial Intelligence."
Prof. Dr. Andreas Dengel, Executive Director of DFKI in Kaiserslautern, explains the long-term benefits of KITTU as follows: "KITTU creates the basis for further improving the developed AI learning methods and their explainability within the framework of prospective, clinical studies and subsequently transferring them to other tumor diseases. To this end, the collaborative project also lays the foundation for an international and interdisciplinary network between clinics
Related websites on the project: