Over 120 participants and 15 speakers explored a wide range of perspectives on the opportunities and challenges around using artificial intelligence for improving health and care. The programme also included poster presentations, demos of (prototype) AI-based healthcare tools and apps, and an ‘Intro to AI’ crash course for those who wanted to brush up on their AI and machine learning knowledge.
Prof. Sandjai Bhulai (Mathematics) kicked off the day with examples of how the use of data and algorithms can have a large positive impact in daily health care practice, such as improving ambulance response times to emergency calls or reducing waiting times for people to get admitted to a care home. Dr. Jenia Kim and Dr. Marike van der Leeden (Amsterdam UMC) showed how Natural Language Processing on electronic health records can be used for automated assessments of Covid-19 patients, while Lukas de Clercq (AmsterdamUMC) applies machine learning methods on routine care data for timely detection of heart failure.
Pitfalls and challenges
Whilst many promising applications were demonstrated throughout the day, various pitfalls and challenges were also addressed. Dr. Emma Beauxis-Aussalet (Computer Science) explained the danger of bias and discrimination inadvertently arising when machine learning algorithms have been trained on unbalanced data. Prof. Henk Marquering (Amsterdam UMC) cautioned about unrealistic expectations of AI, specifically in the domain of Radiology, whilst concluding speakers Prof. Marleen Huysman and Dr. Mohammad Rezazade Mehrizi (KIN Center for Digital Innovation) provided insights into the challenges of implementing AI technologies in practice in the (clinical) workplace.
The challenges presented today, and the discussions that followed, highlighted the need for interdisciplinary collaborations in order to fully utilise the potential of AI in the health and care domain. This day has hopefully contributed to fostering such collaborations and thus increasing the impact of our research.
VU Campus Center for AI and Health
The VU Campus Center for AI and Health is a virtual research network which has been built up over the past 18 months in order to bring together researchers who work on AI for improving health and health care. It is a collaboration between VU Amsterdam, Amsterdam UMC and ACTA, with over 60 permanent staff members from 6 faculties as well as a large number of postdocs and PhD students.
After various online and small in-person events, this conference was the first opportunity for the entire network to get together on the VU Campus for a full day of knowledge exchange and networking. As Prof. Sandjai Bhulai put it in the title of his keynote presentation: Artificial Intelligence in healthcare ... it matters!