DBIS

This thesis focuses on designing an Agentic Graph Retrieval-Augmented Generation (RAG) system specifically for question answering in oral maxillofacial surgery (OMS) guidelines. By leveraging graph-based knowledge representation and advanced language models, the system aims to improve accuracy and efficiency in accessing and interpreting surgical guidelines. Key research areas include the integration of graph databases, ontology-based retrieval, and natural language processing to facilitate precise, contextually relevant responses to complex clinical queries.

This thesis explores the application of diffusion models for the segmentation of brain tumors in 3D MRI images. By leveraging the robust generative capabilities of diffusion models, the research investigates how these models can accurately identify and segment tumor regions in volumetric MRI data. The study focuses on enhancing tumor detection accuracy and addressing challenges associated with 3D medical image segmentation, such as complex tumor shapes and varying intensities.

This thesis aims to develop a single-cell-centric biomedical foundation model that leverages the capabilities of generative pre-trained transformers to enhance the analysis of single-cell RNA data. The model will address critical tasks in single-cell biology, such as cell-type annotation, perturbation prediction, identification of pathogenic cells, and gene network inference.

This thesis is co-supervised by Sikander Hayat and Rafael Kramann, Department of Medicine II, University Hospital Aachen.

Please send your application to Yongli Mou, M.Sc. (mou@dbis.rwth-aachen.de) and CC. Dr. Sikander Hayat (shayat@ukaachen.de)

An AI-based skill assessment tool

The goal of this thesis is to evaluate the applicability of Virtual Graphs in a Distributed Analytics environment.

Mixed reality agents are simulated humans who are displayed in a mixed reality environment, e.g., in augmented reality or virtual reality. They provide the opportunity to support teaching activities with automation. Possible use cases include general presentations in one place, e.g., of lecture content and station-based routes as seen in museums or with tourist guides. If mixed reality agents can take over these repetitive tasks, a consistent quality of the conveyed content is ensured as each run is the same. The agent has the potential to be superior to recorded video presentations as it could possess interactive capabilities, e.g., in answering common questions. Moreover, with the right data structure to encode the presented content, changing and updating the presentation of an agent can be done by quickly adjusting the underlying configuration file whereas videos require a re-recording of content.