Skip to main navigationSkip to main content
The University of Southampton
Health Work

Keynote talk at the NExT++ Workshop in AI in Health & Finance at Nanjing University of Science and Technology

Published: 29 November 2018
Reham Al Tamine, Peter West and Dr
Reham Al Tamine, Peter West and Dr Richard Giordano in front of a drawing of their presentation

Dr Richard Giordano and two of his PhD students, Reham Al Tamine and Peter West, gave an invited keynote talk at the NExT++ Workshop in AI in Health & Finance at Nanjing University of Science and Technology in China. Reham and Peter are doctoral candidates in the Web Science Doctoral Training Centre in ECS. 

NExT++ is a joint research centre established by the National University of Singapore, Tsinghua University of China and the University of Southampton. Prof Dame Wendy Hall of  ECS is the Co-Director of NExT++.  Funded by the National Research Foundation of Singapore, NExT++ carry out research intended to help users and corporations cope with both big data and paradigm challenges related to artificial intelligence. 

The talk, “Enabling Trust, Accountability, and Routine Use of AI-Enabled Healthcare”, focused on three areas of the research that pertain to accountability and trust: Peter’s research looks how doctors use and interpret large-scale, patient-generated data, and the barriers and enablers, including decision-support, of making such data useful in clinical settings; Reham, in collaboration with the Oxford Internet Institute, has been researching patients’ privacy concerns and their willingness to share personal information in various healthcare contexts; Richard’s work builds on Nicolson Price’s concept of “Blackbox Medicine”—the use of large-scale high-quality data streams and complex models, algorithms, and heuristics to support machine learning and ultimately diagnostic and/or prognostic inferences for personalized medicine. 

Such algorithms are opaque, even to the designers, not because they are deliberately hidden but as a consequence of their complexity. To what extent do clinicians and patients trust a judgement that cannot be explained? Both from technical and social points of view, how can such algorithms be explained, and what is the nature of the explanation?  We expect further work on this question with colleagues from Tsinghua University, who will be visiting Southampton in a couple of weeks.

Privacy Settings