‘In recent years, we have seen the challenges that AI projects pose at the interface between research, clinics and hospital operations. They often fail to achieve their goals – not because of poor algorithms, but because of a lack of structures, evidence and trust,’ says Prof. Alexander Meyer. "The IKIM is the answer to this: we bring data scientists directly into the clinics, research and develop AI agents for routine operations, and build a systematic evidence base for medical AI not only in the laboratory, but also in the real clinical environment. Our goal is measurably better care – this also means production-ready systems that relieve the burden on hospital staff and help patients," emphasises the institute director.
Prof. Heyo K. Kroemer, Chairman of the Executive Board of Charité, adds: "The transformative potential of artificial intelligence for the medicine of the future is enormous: processes will be fundamentally redesigned, diagnoses accelerated and therapies made more precise. With the new institute, we are sending a clear signal that technological innovations can be responsibly transferred into patient-oriented care. This will enable us to continue to ensure the highest level of treatment quality and a sustainable healthcare infrastructure in the future."
From basic research to patient care
The newly founded institute systematically links computer science with medicine and basic sciences, building on its well-established collaboration with the Berlin Institute for the Foundations of Learning and Data (BIFOLD). Together, they develop new AI methods, adapt them for medical applications, and then bring them into clinical practice. Data scientists are becoming an integral part of clinical teams – just as much as their colleagues from biochemistry and biology are today.
The main areas of focus at IKIM are:
- Application-oriented development: The institute researches and develops AI systems for patient care, hospital management and medical research. The focus is on agent-based solutions for automating knowledge-based processes and on developing and testing large AI models for intensive care and preventive medicine.
- Evidence-based implementation: The introduction of AI applications in hospitals is scientifically supported by clinical studies. IKIM tests effectiveness and safety, thereby laying the foundation for responsible use in patient care.
- Explainable AI: The institute researches methods that make AI decisions comprehensible to physicians. This transparency is crucial for trust in AI systems and also meets regulatory requirements.
- Transfer and teaching: A service unit advises researchers and clinicians on AI projects. In addition, the institute provides impetus for new curricula and further training courses in order to firmly anchor AI skills in medical education and training as well as in continuing professional development.
New projects for the medicine of the future
In addition to heading the institute, Alexander Meyer will also take on the professorship for Artificial Intelligence in Medicine. He brings with him many years of experience in AI projects and has initiated important digitisation projects at Charité and the German Heart Centre Charité (DHZC) in recent years: from systems for real-time prediction of postoperative complications during intensive care monitoring to current projects on AI-supported medical documentation and the preventive use of health data while driving using AI. As Chief Medical Information Officer of the DHZC, he continues to drive forward the digital transformation, real-time analytics and operational excellence of the DHZC – including in the new building project.
Building on these experiences, Alexander Meyer is planning new projects designed to make Charité fit for the future of medicine. Together with his team, he has set himself the goal of developing an AI strategy for Charité and establishing the necessary structures for the practical application of AI. The digital model clinic, which is currently being developed at the DHZC and serves as a testing ground for innovations, plays an important role in this.
BIFOLD Charité Professorship for Machine Learning
Alexander Meyer is supported by Prof. Grégoire Montavon, who was recently appointed to the first BIFOLD-Charité professorship for machine learning in medicine. The research group he leads, ‘Explainable Machine Learning in Medicine,’ will be based at IKIM. A particular focus of his research is on developing new approaches in the field of explainable AI that can be integrated into modern machine learning (ML) models for medical diagnostics and research.
The Institute for Artificial Intelligence in Medicine is supported by the Berlin Institute for the Foundations of Learning and Data at the Technical University of Berlin (BIFOLD), where Alexander Meyer has been Principal Investigator since 2018. BIFOLD has been cooperating successfully with Charité for many years – for example, on projects such as the use of AI in patient monitoring, pathology, data processing and also in the context of projects on so-called explainable AI (XAI).
Other major sponsors include the German Heart Centre Foundation and the Charité Foundation, which are providing financial support for the newly created W3 professorship for AI in medicine. With its funding, the German Heart Centre Foundation aims to help pave the way for intelligent, data-based patient care in the new DHZC building. The Charité Foundation is supporting the appointment as part of a recruiting grant.
Source: Press release Charité University Medicine
