Hypertensive Disorder in Pregnancy (HDP) is one of the most common medical complications in pregnancy, which can lead to hospitalization or even maternal death. Therefore, it is of vital importance to get in contact with a health care professional when symptoms occur related to HDP. In addition, pregnancies complicated by HDP cause a big cost burden for health care in the short- and long-term. The implementation of symptom data in the workflow of health care professionals may contribute as decision support, so a professional may be able to diagnose HDP in an earlier stage. Related work shows applications, which try to guide and advise pregnant women, however, these data-driven applications do not involve any health care professionals.
The overall aim of this project was to find out how symptom data could be valuable for different health care professionals regarding HDP and what elements are required to let the symptom data perform at its best. First, contextual research was done to get a better understanding of the design space. Finally, a user-deployment study showed how the symptom data could adapt to the needs of different professionals.
Whereas the contextual research gave insights into the different symptoms regarding HDP and at what specific moments the symptom data is valuable in the workflow of professionals, the user-deployment study collected valuable insights in the sense-making process of professionals when perceiving the symptom data. The findings show that symptom data could support the decision-making process of health care professionals if the data is perceived at the right moment in their workflow and may result in earlier diagnoses regarding HDP.
Collaborating with Pregnancy Risk Team (Philips Research)
Created new research proposal for Pregnancy Risk Team
Development value proposition
User deployment research
Analyzing user experience
Analyzing user data
User interaction design
User experience design for algorithm Pregnancy Risk Team
Implementation of data visualizations
Computer vision to locate the position of the symptom tokens
At the beginning of the project, I followed several courses in the field of research ethics, privacy compliance, etc. Although the courses took some time, it helped me prepare a quality research study and I completely made sure that I did not violate the privacy of my participants.
By combining different methodologies, I was able to create new knowledge about a very complex environment and its stakeholders. The storyboards helped me get a better understanding of the current workflow and the opportunities/risks of the envisioned concept. In the second study, the dashboard helped me provoke the participant, so I gained more knowledge about the sense-making process of participants.
Health care professionals can be often conservative about new technology and I experienced this during the first contextual research. In contrast to the first research, participants were very convinced about the envisioned concepts. It showed me how making the concept visible can help healthcare professionals to empathize more.
Technology & Realisation
Throughout the entire project, I learned how technology could not only support my design, but also myself as a designer. For example, in the symptom exploration I used computer vision to collect the location of the data tokens, so I did not had to do it manually. Later on, I made a script to create a fake patient data set for my dashboard with very little user input.
Whereas Math, Data and Computing anticipate on the future, I believe Technology and Realization builds further on that information. With that being said, I was able to create an interactive dashboard, including a fake patient dataset and a simplified model of the envisioned algorithm, based on the collected data from the Contextual Research. The technology created by me, eventually led to new intellectual property (IP) opportunities for Philips.
Math, Data & Computing
In the first research study, I worked closely together with a data scientist from my project team. The collaboration helped me understand the envisioned algorithm of the project team. This resulted in me supporting the project by collecting the right symptom data, so the data scientist was able to create a first version of the algorithm.
Finally, I was able to create and validate a simplified version of the envisioned algorithm for my second study, based on the collaboration with the data scientist, the collected symptom data from the first research study, and the new coding skills.
Creativity & Aesthetics
In the first stage of my project, I needed to collect the different symptoms regarding hypertensive disorders in pregnancy. Therefore, I designed a symptom exploration based on different techniques in literature. Although some information turned out not to be available during the exploration (e.g. chronological order of symptoms), it did work as a good conversation starter for other important topics. It made me realize how combined research methods could lead to new knowledge for a specific context.
For the user-deployment study, I made a dashboard to provoke and engage with participants. In to process of requesting the new research proposal to a Philips Research committee (ICBE), I learned to focus more on the quality of my work instead of rushing to new results. Therefore, I have made sure the prototypes were from good quality and I used the design guidelines from Philips. During the study, the participants were able to interact with the dashboard to gain a better understanding of the data and led to new interesting situation. This showed me the value of well-considered design.
Business & Entrepreneurship
At the beginning of the project, I looked into literature and found evidence for a new value proposition. The literature gave me a good impression about the context and how the envisioned symptom data could have an impact on the market. When proposing my research to a Philips Research committee (ICBE), I learned how to convince the committee about the value of my research based on the found evidence.
Later on, when the contextual research gave insights in the workflow of the target group, I brought the envisioned idea to several concepts. Each concept for a different stage in health care and with its own value. Hereafter, I discussed each concept based on its potential opportunities and risks. Finally, I was able to discuss the potential values for each stakeholder in a Stakeholders Value Analyses.
Buskermolen, D. O., & Terken, J. (2012, August). Co-constructing stories: a participatory design technique to elicit in-depth user feedback and suggestions about design concepts. In Proceedings of the 12th Participatory Design Conference: Exploratory Papers, Workshop Descriptions, Industry Cases-Volume 2 (pp. 33-36). ACM.