Image Processing and Data Management

“Big Data” means amounts of data (image or alphanumerical data), which are too huge or complex or which may quickly change, so that they cannot be analyzed with classic data processing methods. Classic visual image processing software, relational databases as well as statistics and visualization software are often incapable to process such large amounts of data. Thus Big Data handling requires new platforms, data storage and machine learning methods which can run in parallel on up to hundreds or thousands of processors or servers. Companies hope that the analysis of Big Data will open possibilities to obtain competitive advantages, to find cost savings and to create new business fields.

Data management and the responsible handling of data are essential topics in medicine, Life Science and production processes. The management of data (e.g. test results) take up a quarter of researcher’s working time. Additionally different laboratory devices provide different data types and an overview is quickly lost. Even companies have long been applying the digital revolution. Cyber-physical systems connect processes, products, media, and operators via internet. These IT-systems are tested at laboratory scale at the Fraunhofer IWS.

The competence center "Data Management" is a cooperation with the TU Dresden and the University Hospital Dresden. Methods for big data and interactive assistance systems are jointly developed. The focus of the research is on digital data used for early recognition, temperature measurements, diagnostics, therapy schedule, therapy support and therapy success monitoring. Research and development range from image acquisition to image processing, modeling and visualization to user-interface and application design.