AI in pathology is very promising, e.g. deep learning even exceeds human performance in specific cases. However, in the context of medicine it is important for a human expert to validate the outcome and/or to interact with the AI. Current AI models lack an explicit explanation component that allows a human to understand the results. There is a need for transparency and thus traceability of such solutions to make them usable for medicine. The combined use of human intelligence and AI for context understanding should bring important insights and new methodological solutions.
Machine learning requires big training data sets that well cover the spectrum of a variety of human diseases in different organ systems. Data sets have to meet quality- and regulatory criteria and must be well annotated for machine learning at patient-, sample- and image-level. Here biobanks play a central role providing large collections of high-quality well-annotated samples and data. The main challenges are finding biobanks containing ‘‘fit-for-purpose’’ samples, providing quality related meta-data, gaining access to standardized medical data and annotations, and mass scanning of slides including efficient data management solutions.
Participation is free, pre-registration via eMail to: firstname.lastname@example.org