Due to the large amounts of data involved (e.g., ECG data, radiological imaging data, data from medical laboratories, etc.), the medical field is also increasingly relying on artificial intelligence (AI) in the form of machine learning. For everything pertaining to human health, it is extremely important that such systems make the correct decisions. The quality of the underlying data is therefore crucial. In this context, the data has to be interpreted at the systems level, as this is the only way to grasp and understand the relationships between the data. PTB is working to create the quality criteria and the procedures needed to metrologically assess these systemic dimensions of data in order to establish a harmonized quality standard for reliable AI in medical devices.