>> Digitalization makes our everyday lives easier. From scheduling appointments to finding your way around a city, more and more processes are being handled digitally. This also applies to healthcare. Here, digitalization can support members of the healthcare profession and at the same time even increase the quality of the work: documentation in the shock room via automated speech recognition, CT images that can be visualized as a hologram over the patient during operations, or electronic neural bypasses that improve the mobility of paralyzed people restore. All examples that show that with digitalization, interventions can be tailored more individually to patients. From this, new insights and methods can be developed that no one would have dared to dream of just a few years ago. There are also significant opportunities to make everyday medical processes more efficient and thus save time and human resources. At the same time, hardly any other area generates as much data and gains as many new insights as in medicine and the life sciences. It is important to use this data, bundle it, conduct research with it and use the findings to benefit healthcare. Before new data is collected, it is more efficient and controllable to identify existing data treasures and make them available for research in a legally secure manner. Such a treasure trove of data is represented by the data stored by health insurance companies and collected for billing purposes. In the last two legislative periods, the Federal Ministry of Health has already laid the foundation for facilitating healthcare research: the data processing center was set up at DIMDI and research with social
data in Section 75 SGB Tenth Book of the Social Code made easier. Now it is time to make social data from health insurance companies (so-called billing data) accessible to healthcare research on a broader scale and much faster than before. The data processing center at DIMDI is therefore to be further developed into a research data center. For this purpose, the billing data should be collected, processed and stored in a pseudonymized form at the research data center. Upon request, anonymized and aggregated data sets can then be used for research and planning purposes.
A conceivable further step would be to enable insured people in Germany to make their health data available for scientific research projects. A regulation on so-called data donation is not about deciding what is more important: data protection or health protection, but rather how data protection and health protection can be brought into balance in accordance with interests. As an outgrowth of the fundamental right to informational self-determination, data protection is a valuable asset that must be protected. It is therefore important to carefully consider both legally and technically what a legally secure data donation process could look like. The focus should always be on the patient, because it is their data that is collected with regard to their person and should now be further processed while maintaining data protection. On the other hand, society also has a legitimate interest in innovative products and methods that can improve healthcare. However, we will make the greatest progress when we open up about data use and find a way that allows researchers to evaluate health data while maintaining the highest possible data protection standards. The use of data offers us previously unused opportunities to discover completely new connections in diseases, to research new approaches to diagnostic and therapeutic options and to review current care and, for example, to standardize procedures in order to exclude drugs that are not effective with certain genetic predispositions or simply to reduce side effects. We should not let this opportunity go to waste.
The future of medicine will be even more data-driven medicine. The best example of this is the use of artificial intelligence. By using and evaluating large amounts of data, artificial intelligence can support doctors in making decisions about diagnosis and choice of therapy, thereby tailoring therapies to the patient in a more individual and therefore targeted manner. AI systems can already detect tumors on CT images much more quickly and accurately and, for example, mark anatomical structures such as spinal cord, blood vessels and tumor tissue. Even brightness distributions within the tumor can be recorded that are hardly noticeable to the naked eye. In this way, tumor cells can be irradiated more specifically. If there is sufficient image data, it will even be possible in the future to make statements about tissue characteristics without having to do a biopsy. Artificial intelligence should not and will not replace the doctor, but rather give them time and contribute to more precise and personal care. In order to be able to use artificial intelligence in healthcare as usefully as possible, it is crucial to find out what specific benefits which AI applications have for whom and what challenges there are for a broad and, above all, sensible use of artificial intelligence in healthcare.
A total of 500 million euros has been made available precisely for these purposes as part of the Federal Government's AI strategy. Of this, the Federal Ministry of Health received 16.5 million euros, which will be used to support projects that examine the specific benefits of artificial intelligence for patient care. The projects are intended to identify legal, technical, economic, systemic and ethical challenges so that the necessary measures can then be taken to enable the broad and sensible use of AI applications in the German healthcare system.
When we talk about the use of artificial intelligence and the use of data, we cannot stop at our national borders. Networking data storage within Europe makes sense, especially in areas such as cancer research, personalized medicine and the fight against rare diseases.
This way we can expand the data basis and from each other
benefit. That's why Germany wants to set the course for a common European health data space with the German Council Presidency in the second half of 2020. In this way, registers and databases can be networked across Europe. This provides a more reliable basis for research projects and helps, for example, to discover comparable cases of rare cancers, to optimize care routes and to develop new therapies. Together with our partners in the other member states, we want to work on a common understanding of data use in healthcare. In doing so, we want to remain competitive in the development of AI applications and establish a value-based and patient-well-oriented European way of using data. Here, too, we need common quality criteria for checking AI applications. In this way, Europe remains competitive with the USA and China. We are convinced that stronger networking, both internationally and nationally, can produce new knowledge, innovations and processes from which we can all benefit. <<

Citation note: