Towards Inclusive AI System Development for Disease Risk Prediction: Collecting, Prioritising and Incorporating User Stories from Heterogeneous Stakeholders
: Shopland, Nicholas; Burton, Andrew; Brown, David J.; Mahmud, Mufti; Rahman, Muhammad Arifur; Baldwin, David R.; Airola, Antti; Pahikkala, Tapio; Kontio, Elina; Salmi, Jussi; Ferreira, Carlos Alexandre; Pereira, Tânia; Oliveira, Hélder Filipe; Maceda Garcia, Almudena; Chatzichristos, Christos
: Duffy, Vincent G.; Gao, Qin; Zhou, Jia
: International Conference on Human-Computer Interaction
Publisher: Springer Nature Switzerland
: 2026
Lecture Notes in Computer Science
: HCI International 2025 – Late Breaking Papers : 27th International Conference on Human-Computer Interaction, HCII 2025, Gothenburg, Sweden, June 22–27, 2025, Proceedings, Part X
: 16340
: 386
: 403
: 978-3-032-13024-2
: 978-3-032-13022-8
: 0302-9743
: 1611-3349
DOI: https://doi.org/10.1007/978-3-032-13022-8_27
: https://doi.org/10.1007/978-3-032-13022-8_27
Using artificial intelligence (AI) to advance data-driven innovations in preventive health care and clinical decision-making is an expanding field of development. A core feature of such innovation is to ensure that trustworthy and privacy-preserving methods are used. Europe is leading the way in this regard, with agreement on the European Health Data Space and the AI Act coming into force in 2024. PHASE IV AI seeks to advance the current state-of-the-art data synthesis methods, giving AI developers access to larger pools of decentralised, de-identified data through multiparty computing. It will also develop metrics for testing and validation, and protocols that enable synthetic data generation (through multi-party computation). Access to this data market and the data service ecosystem will be through a Health Data Hub in the European Health Data Space. Defining the requirements for the Health Data Hub and the wider system is essential for the success of PHASE IV AI, but it can be challenging when stakeholders have demanding professional vocations. Various methods could be adopted to gather input from the many stakeholders, but where time is valuable, the generation of user stories from hybrid focus group interviews is anticipated to be an effective and efficient method for capturing the range of interests expressed by multiple groups. The aim was to describe the process and outputs from consultations with medical professionals, software developers and small and medium-sized enterprise decision makers through the process of online and hybrid group interviews. The engagement of these professionals in the interview sessions, the interview analysis and extraction of user stories, their refinement and prioritisation and finally their use by the project developers were described. This process looked to provide constructive user stories that provide meaningful recommendations to the developers, and result in an effective product for use in the trans-European context, which will have meaningful impact beyond the end of the PHASE IV AI project.
:
This work is supported by UKRI through the Horizon Europe Guarantee Scheme (project number: 10078953) for the European Commission funded PHASE IV AI project (grant agreement number: 101095384) under the HORIZON Research and Innovation Actions (call: HORIZON-HLTH-2022-IND-13-02) for the European Health and Digital Executive Agency of the European Union.
