The digital patient record consists of a large number of isolated data entries that are provided by various health information systems and medical devices. This data is often very complex and can be considered completely only in the decision making process by a physician, when it is available in an understandable, clear manner before, during and after a surgery. The research group “Digital Patient- and Process Model” provides concepts for the formal description and modeling of patient data related to a disease or an organ. These concepts form the basis for developing patient models that link and analyse all data relevant for a surgery. Patient models are supposed to provide a complete view on a patient, their disease and disease occurrences.
In order to bring the digital patient and decision process model into real clinical applications, the group collaborates closely with surgical departments to identify real use cases and also to verify the models. For this purpose, the developed concepts will be realized in prototypes and relevance for their practical application will be assessed.
The work of the research group is centered on the development of digital patient models. The group will establish the underlying information structure for such models. Further, methods for integrating patient data will be developed. They allow to bring together various data types including text, images, attribute-value pairs in one model. In particular, free-textual descriptions need to be processed and analyzed to come to structured, machine-processable information that can be considered in other processes. To describe such information in a standardized way, ontologies will be exploited.
Information processing and knowledge management
In order to get an additional value from the data integrated in the patient model and also to identify links between information entities, methods for semantic analysis are necessary. One working area of the research group will thus deal with preparing and extending such methods from the research areas of textmining, semantic analysis, information retrieval and knowledge management. Those technologies connect information entities automatically and make them available among others for automatic interpretation and reasoning purposes.
Before and after a surgery decisions are made that require awareness of all relevant information on a patient and their disease. To allow for this awareness by means of clinical decision support systems, the research group will follow again a model-based approach that bases upon models of surgical decision processes. The model-based approach will help to develop tools for automatically providing prognosis for possible treatments or making suggestions for possible interactions and treatments.
The variety of information entities that are stored in the patient model and that are considered in a decision process are often complex. In order to enable a physician to quickly get an overview on the relevant aspects or relationships and to point him to relevant pieces of information, the data needs to be prepared and visualized in an easy understandable manner. Beyond, methods for managing the patient and decision model are required that also allow verification and adaptation of the models.