AI in medicine: COVID-19 and beyond

During an acute crisis such as the COVID-19 pandemic, many people need to make rapid decisions based on the latest evidence from medical science: medical researchers, pharmaceutical companies, public health authorities, health care staff and governments. Yet the amount of data that needs to be reviewed is too large and complex to be analysed by humans at this point: millions of scientific articles which may harbour important clues for new treatments, millions of patient journals which may identify disease patterns and risk groups and mountains of data from laboratory and clinical studies that increase at an incredible pace. The same need exists in relation to other global health issues such as heart disease, cancer, neurodegeneration. AI-based methods can help address this challenge in part by extracting, connecting and analysing information at a scale that far surpasses the abilities of humans. In addition, artificial intelligence can help improve many aspects of clinical care, e.g. by increasing the speed, accuracy and reproducibility of diagnostics or automating routine tasks. At this stage, most medical AI tools are still at the research and development stage. However, it is clear that over the next years AI will transform medical research and clinical care. This also comes with societal, legal and ethical challenges that need to be addressed. This lecture will give an overview over the uses of AI for COVID-19 and other medical areas and discuss the possibilities and challenges ahead.