The research group Artificial Intelligence in Medical Applications (AIM) focuses on developing innovative AI technologies to enhance diagnostic accuracy and efficiency in medical image analysis. We specialize in applying deep learning and self-supervised learning methods to medical imaging domains such as gastroenterology and radiology, addressing critical clinical challenges through interdisciplinary collaboration between medicine and computer science. Our research emphasizes the integration of AI in real-time clinical environments, fostering improved patient outcomes and streamlined workflows.
AIM's current projects include the development of transformer-based models for endoscopic and radiologic image analysis, as well as AI systems that reduce manual annotation requirements by leveraging existing medical reports. In partnership with the Universitätsklinikum Würzburg (UKW) and the Charles University Prague, we aim to advance diagnostic support tools that enable immediate decision-making during clinical procedures. Our work in multimodal AI, combining medical imaging with natural language processing, represents a key area of innovation in disease detection and classification. We are dedicated to fostering transparency and trust in AI systems by incorporating explainable AI techniques into clinical practice, enhancing model interpretability for medical professionals.
