Gyuchan Thomas Jun, Satheesh Gangadharan, Georgina Cosma, Panos Balatsoukas, Cecilia Landa-Avila, Francesco Zaccardi, Michelle O’Reilly, Ashley Akbari, Vasa Curcin, Rohit Shankar, Reza Kiani, Neil Sinclaire, and Chris Knifton
The aim of this presentation is to present a systemic design framework a research team developed for their UK’s NIHR (National Institute for Health Research)-funded research project, DECODE (Data-driven machine-learning aided stratification and management of multiple long-term conditions in adults with learning disabilities). DECODE will analyse healthcare data on people with learning disabilities from England and Wales to find out what multiple long-term conditions (MLTCs) are more likely to occur together and what happens to some of these MLTCs over time. The end goal of DECODE is to utilise the AI-enabled new knowledge and develop actionable insights for effective joined-up social and health care for people with learning disabilities. The framework we are proposing consists of four steps: i) context analysis to understand the context of AI application; ii) AI output visualisation to develop user-friendly visualisations to display the outputs of AI analysis in a meaningful and accessible way; iii) actionable insight exploration to explore leverage points to improve join-up care coordination; iv) change process planning to evaluate the feasibility and ethical/legal risk of the usage scenarios. This framework will be of interest to many systemic designers who aim to develop a safe, ethical and cost-effective AI in healthcare.
Keywords: artificial intelligence, health care, social care, people with learning disabilities, systemic design