Infographics
Infographics that provide information about Parkinson's disease.


Development of AI
Artificial intelligence is not new. Since the Dartmouth conference in 1956, researchers have alternated between big promises and disappointment. The first chess and logic programs led people to expect human-level intelligence within a decade. That expectation fell apart in the “AI winter” of the 1970s. Expert systems in the 1980s worked for simple cases but not for complex ones. The real shift came when AI stopped following hand-written rules and started learning patterns from data. Mobile phones and sensors then produced huge amounts of that data. Today AI is part of everyday life, and generative AI has reached millions of people in a matter of weeks.
AI-PROGNOSIS draws on several of these steps. It uses machine learning, and it uses data from everyday devices such as smartwatches and smartphones. The goal is AI tools that can help detect Parkinson’s disease earlier and predict how it may develop in each person. The history is also a warning. Parkinson’s is complex and varies a great deal from person to person, which is exactly the kind of problem that earlier AI waves failed on. That is why AI-PROGNOSIS works with people living with Parkinson’s, clinicians and other stakeholders. The aim is AI that is trustworthy and transparent, and useful in the lives of the people it is meant to serve.