Dr Haiou Zhu
Academic and research departments
Centre for Vision, Speech and Signal Processing (CVSSP), Surrey Institute for People-Centred Artificial Intelligence (PAI).About
Biography
Dr Haiou Zhu is a Lecturer in Inclusive Design at the UKRI AI Centre for Doctoral Training in AI for Digital Media Inclusion. Her research combines inclusive design and behavioural science to inform the design of digital interventions and AI systems. She holds a PhD in Design from Loughborough University and was previously a postdoctoral researcher in the Department of Psychiatry at the University of Oxford, where she worked on co-designing with patients and public for mental health. She now leads projects extending co-design into environmental health and AI-enabled science, including studies on indoor greening, respiratory health, and generative AI for environmental governance. Her work is published in The Lancet Planetary Health, Scientific Reports, Applied Ergonomics and JMIR journals.
Areas of specialism
ResearchResearch interests
- AI for digital media inclusion and human-centred AI — integrating inclusive design and participatory research with AI development so creative and generative AI tools are built with, rather than just for, the people who use them.
- Behavioural science and decision-making — applying behavioural science theories and methods to design for behaviour change.
- Digital health and serious games for health and wellbeing — co-designing and evaluating digital interventions, including serious games, with patients and users.
Research projects
CHILD-AIR (Children's Home Intervention to Lower Domestic AIR pollution)Behavioural insights and lifestyle adaptations to reduce indoor pollution levels and potentially improve health outcomes in children with respiratory disease
GREENIN Micro Network Plus Feasibility StudyUnderstanding and Measuring Pathways from Engagement with Indoor Greening Infrastructure and Wellbeing
Research interests
- AI for digital media inclusion and human-centred AI — integrating inclusive design and participatory research with AI development so creative and generative AI tools are built with, rather than just for, the people who use them.
- Behavioural science and decision-making — applying behavioural science theories and methods to design for behaviour change.
- Digital health and serious games for health and wellbeing — co-designing and evaluating digital interventions, including serious games, with patients and users.
Research projects
Behavioural insights and lifestyle adaptations to reduce indoor pollution levels and potentially improve health outcomes in children with respiratory disease
Understanding and Measuring Pathways from Engagement with Indoor Greening Infrastructure and Wellbeing
Teaching
Inclusive Research - UKRI Centre for Doctoral Training in AI for Digital Media Inclusion
Publications
The assessment of local muscle fatigue during dynamic exercises is viewed as essential, since people tend to maximise effective muscle training and minimise harmful muscle injuries. This study aims to investigate how surface electromyography (sEMG) and subjective metrics together can play a role in dynamic muscle fatigue prediction during exercises to promote the design of health and fitness technology. 20 healthy male participants were recruited in the experiment, and sEMG and self-reported data were collected. Features in temporal and spatial domain were extracted from sEMG data, such as RMS (root mean square) and FInsm5 (spectral parameter proposed by Dimitrov). Our results showed that some sEMG features indicated directional changes with the increase of dynamic muscle fatigue. Spearman correlation analysis indicated that Borg ratings had strong correlations with RMS and FInsm5 (spectral parameter proposed by Dimitrov) slopes. Then this paper discusses how to use sEMG and Borg data to evaluate muscle fatigue during exercises. A framework is proposed based on the joint analysis of spectra and amplitudes across RMS and FInsm5 slopes. This paper further discusses how to design health and fitness technology for the benefits and the limitations of the study.
Ecological-collective-flourishing (e-co-flourishing) describes the interconnected flourishing of human and non-human natural entities. Despite existing conceptual frameworks, translation into practice remains inadequate owing to insufficient synthesis of empirically tested mechanisms driving change across human and ecological systems. We conducted an umbrella review of systematic reviews examining mechanisms at the e-co-flourishing interface. Searches of seven databases and four grey literature sources identified nine eligible reviews. Evidence supported several cross-domain mechanisms, including psychological (eg, attention restoration and reduced stress), social (eg, cohesion and contact), physical (eg, activity), and environmental pathways (eg, reduced stressors and enhanced environmental quality). Psychological mechanisms showed the most consistent support, whereas evidence for other mechanism domains was mixed. Evidence for causality was weak and largely derived from cross-sectional and qualitative studies, and no reviews examined reciprocal mechanisms influencing both human and non-human flourishing. Future research should prioritise rigorous designs, theory-driven testing, and replication to strengthen evidence for equitable and sustainable e-co-flourishing pathways.
As an ethical design approach embedding with the human value of inclusiveness, inclusive design could contribute to economic value creation. However, research on the relationship between economic value and human values in inclusive design has seldom been explored. This preliminary literature review focuses on how value and values have been discussed in inclusive design research. The findings first present the evolving conceptions of inclusive design that formulate and transform the understanding of value and values. Then, existing literature on the economic value of inclusive design, and inclusive design for human values at both individual and social levels are reviewed respectively. We categorize these disparate discussions into 'value creation' and 'value distribution' and propose opportunities for an integrated approach that would bridge discussions on the economic value and human values in future research.
Background: Digital health (DH) brings considerable benefits, but it comes with potential risks. Human Factors (HF) play a critical role in providing high-quality and acceptable DH solutions. Consultation with designers is crucial for reflecting on and improving current DH design practices. Objectives: We investigated the general DH design processes, challenges, and corresponding strategies that can improve the digital patient experience (PEx). Methods: A semi-structured interview study with 24 design professionals. All audio recordings were transcribed, deidentified, grammatically corrected, and imported into ATLAS.ti for data analysis. Three coders participated in data coding following the thematic analysis approach. Results: We identified eight DH design stages and grouped them into four phases: preparation, problem-thinking, problem-solving, and implementation. The analysis presented twelve design challenges associated with contextual, practical, managerial, and commercial aspects that can hinder the design process. We identified eight common strategies used by respondents to tackle these challenges. Conclusions: We propose a Digital Health Design (DHD) framework to improve the digital PEx. It provides an overview of design deliverables, activities, stakeholders, challenges, and corresponding strategies for each design stage.
Converging ecological and social crises have exposed the limits of prevailing models of human progress, highlighting the need for more inclusive and interdisciplinary approaches to advancing wellbeing and flourishing across human and non-human life. In this Personal View, we propose an ecological-collective-flourishing (e-co-flourishing) approach, whereby one’s own human flourishing is pursued along with that of natural ecosystems and their collective and individual constituents. We offer a heuristic translational framework that includes a methodological approach; guiding principles; and proposed programme theory in terms of hypothesised outcomes, pathways of change (mechanisms), key contextual factors (moderators), and intervention characteristics. This framework aims to bridge theory to real-world applications of e-co-flourishing interventions that are effective, scalable, and promote change through targeted mechanisms of action. We present Mindfulness-Based Cognitive Therapy-Balanced Living for Us and the Earth’s Ocean (MBCT-BLUE) as an illustrative application of this framework to outline how the framework can be used to identify, refine, develop, and evaluate interventions with e-co-flourishing potential. We also provide definitions of key conceptual terms and proxy measurements. Practitioners, researchers, and decision makers working at the interface of mental health, public health, sustainability, and planetary health can use this framework and its resources to guide emerging research and innovation.
Background: The COVID-19 lockdowns led to significant resource constraints, potentially impacting mental health and decision-making behaviors. Understanding the psychological and behavioral consequences could inform designing interventions to mitigate the negative impacts of episodic scarcity during crises like pandemics. Objective: To investigate the effects of perceived scarcity on mental health (stress and fear), cognitive functioning, time and risk preferences (present bias and risk aversion), and trade-offs between groceries, health, and temptation goods during and after the COVID-19 lockdown in Shanghai. Methods: A quasi-natural experiment was conducted in Shanghai during and after the COVID-19 lockdown. Web-based surveys were administered in May 2022 (during lockdown) and September 2022 (post-lockdown). Propensity score matching was used to balance demographic factors between the groups (During: n=332; After: n=339). Data were analyzed using regression analyses, controlling for potential confounders and applying propensity score matching weights. Results: Perceived scarcity was significantly higher during the lockdown (mean 7.97 (SD 2.1)) than after (mean 4.35 (SD 2.27); P