Fraunhofer Institute for Applied Information Technology FITDigitization in healthcare and in the life sciences in general is opening up new opportunities to significantly improve diagnosis, therapy and patient care. In particular, this means that existing processes can be supported, analyzed, automated and improved. Essential basic building blocks for this are the processing and analysis of large amounts of high-quality data and its secure communication between the people and organizations involved in the processes. It is important to us that people are in the focus of of these processes and are effectively supported in their everyday lives and at work.
Digital Health Spaces
Tabbed contents
Data ecosystems in the life sciences
Healthcare and the life sciences need to share data between organizations and individuals. This is a complex task that today is addressed most often by building isolated IT applications or data silos. The research guided by the idea of data ecosystems tackles this problem by proposing general concepts for the sharing of data between different parties in a structured and secure way. In healthcare and the life sciences in general, this is a particularly challenging issue, as many different stakeholders are involved, sensitive data is processed, and many different laws and guidelines have to be considered.
Data serves as a foundation for decision-making and must therefore be generated, processed, and analyzed in high-quality processes. Platforms and applications provide a basis that support and control processes, for example in patient care or in collecting clinical study data. In addition, the quality of data from internal and external sources must be monitored to evaluate the results of analyses and predictions and, if necessary, to reject or cleanse data. At each step, it must be transparent how the current data and thus its quality were created.
The research of the Digital Health Spaces group focuses on:
- Data quality and metadata management
- Patient-centered design and data sovereignty
- Data exchange and data integration
- Data transparency
- Mobile and web applications
Current projects
- AI-NET-PROTECT – Providing Resilient & secure networks [Operating on Trusted Equipment] to CriTical infrastructures
- Medical Data Space
- International Data Space
- Innovation Fund Project SALUS
- Telemedicine Prototype for Diabetes Patients
- Secure Messenger in Emergency Medicine
Data analysis in the life sciences
Medical imaging and microscopy are crucial tools for diagnostics and research. Without them, many diseases would be undetectable, the development of new therapies and drugs impossible. Besides the growing wealth of image data, the life sciences begin to profit from the wider use of wearable devices in the fitness and health sectors. They are starting to generate streams of valuable health data. Machine learning and artificial intelligence methods are used to analyze the data. Hence, they help to deliver precise information and insights faster, speeding up and improving diagnostics and research. Obviously, the quality of the used data and the analysis results play an essential role.
The Digital Health Spaces group studies methods for efficient data, image, and video analysis and for evaluating their quality. We have a strong expertise in:
- Deep learning in image and video analysis
- Segmentation (semantic segmentation and instance segmentation)
- Classification
- Classical image and video analysis
- Segmentation
- Classification
- Registration