Cohort 1 student Marceli publishes in JMIR Human Factors entitled ‘Design and Evaluation of an Intensive Care Unit Dashboard Built in Response to the COVID-19 Pandemic: Semistructured Interview Study’.

Marceli had the following to say about the experience: ‘Dashboards and interactive displays are becoming increasingly prevalent in most health care settings and have the potential to streamline access to information, consolidate disparate data sources and deliver new insights. Our research focuses on intensive care units (ICUs) which are heavily instrumented, critical care environments that generate vast amounts of data and frequently require individualized support for each patient. Consequently, clinicians experience a high cognitive load, which can translate to suboptimal performance. The global COVID-19 pandemic exacerbated this problem by generating a large number of additional hospitalizations, which necessitated a new tool that would help manage ICUs’ census. In a previous study, we interviewed clinicians at the University Hospitals Bristol and Weston National Health Service Foundation Trust to capture the requirements for bespoke dashboards that would alleviate this problem.

This study aims to design, implement, and evaluate an ICU dashboard to allow for monitoring of the high volume of patients in need of critical care, particularly tailored to high-demand situations, such as those seen during the COVID-19 pandemic.

Building upon the previously gathered requirements, we developed a dashboard, integrated it within the ICU of a National Health Service trust, and allowed all staff to access our tool. For evaluation purposes, participants were recruited and interviewed following a 25-day period during which they were able to use the dashboard clinically. The semistructured interviews followed a topic guide aimed at capturing the usability of the dashboard, supplemented with additional questions asked post hoc to probe themes established during the interview. Interview transcripts were analyzed using a thematic analysis framework that combined inductive and deductive approaches and integrated the Technology Acceptance Model.

A total of 10 participants with 4 different roles in the ICU (6 consultants, 2 junior doctors, 1 nurse, and 1 advanced clinical practitioner) participated in the interviews. Our analysis generated 4 key topics that prevailed across the data: our dashboard met the usability requirements of the participants and was found useful and intuitive; participants perceived that it impacted their delivery of patient care by improving the access to the information and better equipping them to do their job; the tool was used in a variety of ways and for different reasons and tasks; and there were barriers to integration of our dashboard into practice, including familiarity with existing systems, which stifled the adoption of our tool.

Our findings show that the perceived utility of the dashboard had a positive impact on the clinicians’ workflows in the ICU. Improving access to information translated into more efficient patient care and transformed some of the existing processes. The introduction of our tool was met with positive reception, but its integration during the COVID-19 pandemic limited its adoption into practice.

This project informs the future developments pertaining to the use of dashboards and interactive displays within intensive care unit settings.’

Link to Paper: https://humanfactors.jmir.org/2023/1/e49438

 

Cohort 2 student Harry Emerson publishes in the Journal of Biomedical Informatics entitled ‘Offline reinforcement learning for safer blood glucose control in people with type 1 diabetes’.

Harry had the following to say about his publication: ‘The publication explores how machine learning algorithms can be used to improve insulin dosing decisions for people with type 1 diabetes. Artificial pancreas devices have shown great success in reducing the burden of diabetes management, but rely on simplistic and reactive control algorithms. This work applies offline reinforcement learning as a method for learning sophisticated and safe strategies from pre-collected patient data. The method is verified in a simulation of 30 people and explores practical challenges, such as human error, device malfunction and data quality. The presented method significantly improved blood glucose control compared to current state-of-the-art control algorithms and was shown to be more robust than previous reinforcement learning approaches to the constraints of real-world data. The algorithm demonstrated the greatest benefit in children, which represent a particularly important group as they are often unable to manage their diabetes without assistance.

I implemented a selection of reinforcement learning algorithms in the type 1 diabetes simulator and created a full pipeline to perform data collection, training and evaluation. I modified the established UVA/Padova simulator to incorporate common blood glucose scenarios in which to evaluate the approach.

This represents the first demonstration of the benefits of offline reinforcement learning in blood glucose control. This work provides a basis for continued reinforcement learning research, demonstrating the potential of the approach to improve the health outcomes of people with type 1 diabetes, while highlighting the method’s shortcomings and areas of necessary future development. The publication was picked up by Wired Science, who write an article about myself and my research https://www.wired.com/story/managing-type-1-diabetes-is-tricky-can-ai-help/.

Link to Paper: https://www.sciencedirect.com/science/article/pii/S1532046423000977

Cohort 2 student Sam James published a journal article via Proceedings of the 2023 CHI Conference on Human Factors in Computing Systems entitled ‘Chronic Care in a Life Transition: Challenges and Opportunities for Artificial Intelligence to Support Young Adults With Type 1 Diabetes Moving to University’.

Sam had the following to say about his paper: ‘The paper uses data collected in one-to-one interviews with young adults in the UK who had recently experienced the move to university and were living with type 1 diabetes (T1D). From a thematic analysis of the interviews, findings were made about this life transition and its impact on T1D management. These focused on the changes in lifestyle and the changes in support network. The changes in lifestyle included changes to drinking habits, eating habits, sleeping habits, physical
activity habits and overall schedule. The changes to support network highlighted; the increased independence, parents’ role in T1D management, explaining T1D to people and the assistive roles
people at university fill. From these findings, several opportunities and challenges for technology during the transition to university are discussed, with a focus on artificial intelligence and the closedloop system. These include automated personalisation, customisation, data limitations, the limitations of artificial intelligence in unusual scenarios and the potential of human-centred based design solutions. The paper then considers the wider implication of these findings for other chronic conditions and suggests the need for further research to allow personalised solutions to develop that consider the problems caused by life transitions.

I was first author on the paper and performed the majority of work across the process with input from my supervisors, who made up the remaining authors. The work I did included the project setup (exploring the research space and gaining ethical approval for the study), data collection (participant recruitment, online interviews and transcribing them), data analysis (thematic analysis)
and write-up (selecting the novel parts of the analysis and creating a paper to explain them).

The work aims to highlight the difficulties that life transitions cause in chronic condition management and to trigger more research into technology during these periods. This hopefully will
lead to the design of management systems that can cope with the challenges of life transitions or increase awareness of times they may be less effective and why.’

Link to Paper: https://dl.acm.org/doi/10.1145/3544548.3580901

Cohort 1 student Romana Burgess publishes a journal article in the Journal of Non-verbal Behaviour entitled: ‘A Quantitative Evaluation of Thin Slice Sampling for Parent–Infant Interactions.’

Romana had this to say about her paper: ‘Broadly, the paper looks at whether we can use brief observations (“thin slices”) of behaviours to approximate those same behaviours over a longer period of time. The purpose of this work is to find an approach to alleviate the “coding burden”, i.e., the amount of time that researchers spend coding behaviours from observational data. In essence, behavioural coding is extremely time-intensive and laborious, and this paper both explores and quantifies the value of thin slice sampling as an alternative approach.

The analysis is based on video data of interactions between parents and their infants. These data come from two cohort studies: the Avon Longitudinal Study of Parents and Children (ALSPAC) – based in Bristol – and Grown in Wales (GiW) – based in Cardiff. Some videos were recorded in a research clinic, but most were recorded in the participants own homes.

The videos were coded in 5-minute segments for a large range of behaviours, for example, vocalisations, facial expressions, and body orientation. Then, I used Markov modelling to quantify long-term patterns and transitions between behaviours for 15 distinct thin slices of the full 5-minute interactions, and I compared measures drawn from the full sessions to those from shorter slices.

The paper identified many instances where thin slice sampling was an appropriate coding approach, although there was significant variation across behaviours. From here, I was able to quantify how long is appropriate to code for each behaviour, depending on video context and individual research objectives.

This work constitutes the first project of three that comprise my PhD thesis, and is crucial to its completion. The second project considers a different approach to alleviating the coding burden, and these two works together contribute to the third project, which looks at linking coded facial expressions during parent-infant interactions to parental mental health.’

Link to Paper: https://link.springer.com/article/10.1007/s10919-022-00420-7

Cohort 4 student Veronica Blanco Gutierrez publishes a journal article entitled ‘Culture and breastfeeding support’ in the British Journal of Midwifery.

Veronica had this to say about the paper: ‘This article discusses the importance of taking into consideration different cultural aspects of every women when health professionals, particularly midwives, provide breastfeeding support. It is vital for the provision of breastfeeding care to have social determinants of health, such as culture, at the heart of care. Adequate and tailored breastfeeding support is key in the provision of care. I am very passionate about breastfeeding support and how to improve the breastfeeding experience for women and their babies. I am hoping to undertake research on improving breastfeeding support to improve health outcomes.’

Link To the Paper: https://www.magonlinelibrary.com/doi/abs/10.12968/bjom.2022.30.12.713