This workshop will explore the role of HCI in enabling the implementation of AI in real-world healthcare. Drawing on fieldwork, prototypes, and case studies, we will reflect on key concepts in AI, including 'explainability', 'trust', and 'accuracy', to support researchers and developers in taking a more sociotechnical and human-centered perspective when using AI systems to change care delivery.

AI research has been growing steadily and promising to improve results for many areas of healthcare. HCI has contributed to these discussions, mainly with studies on explainability of advanced algorithms. However, only few machine learning systems make it to everyday care. This challenging move has been named the 'last mile' of AI in healthcare, emphasizing the sociotechnical uncertainties and unforeseen learnings from involving users in the design or use of AI-based systems. With this workshop we aim to set the stage for a new wave of HCI research that accounts for and begins to develop new insights, concepts, and methods, for going the 'last mile' with AI in healthcare.

See the complete workshop proposal.

Contributing

We expect position papers describing or discussing artifacts, empirical results, theoretical reflections, and design explorations, that reflect on insights, concepts, or methods related to the implementation and use of AI in healthcare. We encourage submissions from researchers, engineers, designers, data scientists, social scientists, medical professionals, and patients, who are interested in broadening reflections around human-centred AI in healthcare and the sociotechnical challenges appearing in the wild. Possible topics include, but are not limited to:

• Empirical studies of healthcare AI in the wild
• Methodological challenges for studying AI systems in the wild
• Human-centred AI in the HCI design process
• Theories and framings for broadening HCI approaches to AI and health
• Sociotechnical impacts and critical perspectives
• Expanding the conception of what AI-tools in healthcare should offer

Prospective participants are invited to submit a 4-5 pages position paper using the CHI 2021 Format (templates: MS Word and Latex) - references are excluded from page count. Non-academics can apply to the workshop with a motivation letter. Submissions should be sent to: chi2021wsaihealth@gmail.com by February 28th 2021 (extended).

Authors of accepted position papers will prepare a 2-3 minute video that explains their contribution to others. Videos will be shared on the website before the workshop to help participants starting to reflect about the topic before the workshop takes place. The workshop itself will take place via Zoom or MS Teams.

Accepted position papers will be made available on the Workshop's website. Moreover, all authors will be invited to expand their research and submit a full paper to a special issue that is being prepared on the same topic for a key HCI journal.

Important dates

February 28th 2021: Submission deadline (extended)
March 19th 2021: Notification of acceptance
April 23rd 2021: Camera-ready version of papers
April 30th 2021: Short videos from authors
May 8th and 9th 2021: Workshop at CHI 2021

Organizers

Tariq Osman Andersen, University of Copenhagen and Vital Beats ApS
Francisco Nunes, Fraunhofer Portugal AICOS
Lauren Wilcox, Google and Georgia Tech
Elizabeth Kaziunas, AI Now, New York University
Stina Matthiesen, University of Copenhagen
Farah Magrabi, Macquarie University