 Already during medical school I was fascinated by neurodegenerative diseases, how they can be diagnosed and treated. My name is Stefan Klöppel, I am head of the Department of Old Age Psychiatry at the UPD in Bern. My research focuses on early diagnosis and prognosis of neurodegenerative diseases based on MR images of the brain. More recently we started to develop cognitive interventions and predict their response at the level of the individual. When I started using machine learning over 15 years ago, our support vector machines learned from training examples to distinguish between new cases into either healthy or those affected by a neurodegenerative disorder. As a clinician I liked these methods because you can easily visualize the basis for the decision of the AI system. This type of faith validity is more difficult to achieve with many of the current methods such as deep learning. What we had envisioned back then was the development of a larger data set including cases with a high level of diagnostic certainty. Those would be used to train AI systems which would then be applied outside specialized centers. While that worked nicely on clean study data, the application to clinic proved so difficult that these methods still haven't made it into the routine outside a handful of centers. I am really excited about the founding of CHIME with its sections on medical imaging. But probably at least as interesting will be to take a good look into related fields of AI research such as sensor technologies for example. Algorithms analyzing sensor data have become much more powerful producing meaningful and clinical behavioral data in real time. We are just starting a new project with the RTOC team of Tobias Nief with the aim to improve the care of our inpatients. Sensors could for example help us to identify patients developing anxiety also in the middle of the night or warn us to take extra care of those at risk of falling. I am sure AI systems will dramatically change the field of psychiatry. At the UPD you can already observe the transition with apps advising patients on the optimal time in bed as well as a range of studies on e-mental health currently conducted at the Department of Child and Adolescent Psychiatry. The ubiquity of smartphones combined with AI help clinicians to detect changes in patients' social behavior which may indicate a deterioration or onset of a psychiatric condition. 15 years ago clinicians and computer scientists were living in completely separate worlds each with a complex but non-overlapping language. Initiatives such as KN are perfectly geared to bridge these gaps.