HDR UK Applied Analytics Webinar: A method for machine learning generation of realistic synthetic datasets – George Despotou (52:41)

Digital health applications can improve quality and effectiveness of healthcare, by offering a number of new tools to users, which are often considered a medical device. Assuring their safe operation requires, amongst others, clinical validation, needing large datasets to test them in realistic clinical scenarios. Access to datasets is challenging, due to patient privacy concerns. This may result in overheads such as delays and additional cost, that may ultimately deprive patients from potential benefits from innovations.  Use of synthetic datasets is increasingly seen as a way for early validation of applications, overcoming the privacy issues. However, the synthetic datasets will need to be demonstrably equivalent to the real datasets. The presentation will give an overview of a method for the generation of realistic synthetic datasets, statistically equivalent to real clinical datasets, using Generative Adversarial Networks (GAN). (George Despotou) 07/09/22 Paper mentioned during the seminar: Assessing, visualizing and improving the utility of synthetic data, Gillian M Raab, Beata Nowok, Chris Dibben (2021) available from https://arxiv.org/abs/2109.12717