AI ecosystem in health care: Health system-based analysis of disease patterns using lung diseases as an example.

The project "Health system-based analysis of disease patterns using lung diseases as an example" is part of the joint project "AI ecosystem in health care" funded by the state of Schleswig-Holstein. At the Institute of Medical Informatics, we are developing new AI-based methods for evaluating image data in combination with additional clinical information in cooperation with various clinics from the University Medical Center Schleswig-Holstein, Campus Lübeck, and the Institute of Information Systems.

The treatment of many lung diseases is currently evolving from established standard therapies towards personalised medicine. In the case of lung diseases, this development is particularly important as they have an extremely broad spectrum from malignant tumours, pneumonias including e.g. Covid 19 to chronic diseases such as COPD, asthma, pulmonary hypertension or pulmonary fibrosis and have a high socio-economic importance. Medical images and radiological biomarkers play an essential role in the differentiation of individual clinical pictures, but a large amount of additional information about the patient, such as age, gender, laboratory data, physiological measurements, etc., is also relevant for the diagnosis and therapy recommendation.

Participating clinical departments provide a large amount of annotated images, structured data collections, and a high level of medical expertise in the field of lung diseases. At the Institute of Medical Informatics frameworks for the combined AI-based analysis of lung image data and clinical patient information of different disease patterns such as covid-19, lung tumors, etc. will be developed. The possibilities for improving diagnostics and increasing the safety of therapy recommendations will be demonstrated using specific use cases. In addition, largely standardized processes for data acquisition, annotation and integration of AI solutions in clinics will be developed in the application field of lung diseases, which should enable faster development of AI solutions in the clinical environment in the future.

Cooperation Partners

Prof. Dr. rer. nat. habil. Ralf Möller
Institute of Information Systems, Universität zu Lübeck

Prof. Dr. med. Jörg Barkhausen
Department of Radiology and Nuclear Medicine, UKSH Campus Lübeck

Prof. Dr. med. Michael Hans Frank Schweigert
Department of Surgery, UKSH Campus Lübeck

PD Dr. med. Daniel Drömann
Medicine - Pulmonology, UKSH Campus Lübeck