(German version below)
Data protection is of great importance to Fraunhofer-Gesellschaft and its institutes. Fraunhofer-Gesellschaft is engaged in application-oriented research with the goal of affording its partners system- and technology-oriented innovations as well as ways to strengthen their ability to compete at both a regional and global scale. Our methods include processing personal data within research projects, which in turn requires a lawful handling of personal data by the employees taking part as defined by the processing procedures created for this purpose.
The scientists behind the project COPERIMOplus work closely together with the legal department of Fraunhofer-Gesellschaft in order to ensure that all research data consisting fully or partly of personal data is acquired and processed in full accordance with the General Data Protection Regulation (EU 2016/679) and other applicable data protection laws.
A provision of pseudonymized or anonymized research data provided to Fraunhofer-Gesellschaft through third parties would take place in accordance with our legal requirements, which warrant full legal protection of the data subjects, as well as ensuring the legal compliance of you as a data provider and that of Fraunhofer-Gesellschaft as a data recipient.
Please let us know if you would like us to suggest an appropriate Data Transfer Agreement, including all necessary provisions in order to secure the aforementioned purposes.
We will also be happy to consider a draft Data Transfer Agreement suggested by you and determine whether such a draft is compatible with the purposes and means of the data processing to be carried out within COPERIMOplus.
Please do not hesitate to contact us in case of further questions.
As an alternative to data transfer, we can offer to provide you with a containerized AI model that can easily be re-trained behind your hospital / institute firewall. Contractual obligations in this case will focus on data protection requirements, the re-training itself (you can work with our model if you re-train it with your data) and the conditions for sharing the resulting, re-trained model.
As a result, we offer “federated learning” where the algorithm comes to the data. Please note that in any case we will need a formal, legal agreement for our collaboration.