Department of
Data Science

Student project: Conversational interfaces for digital twins in health

Managing one’s health or a chronic disease in daily life can be a challenge due to burdensome, intrusive and/or infrequent health monitoring. At OnePlanet, we are therefore developing new sensing techniques to continuously and non-obtrusively collect health data. Insights from these multimodal data would allow for timely lifestyle recommendations to maintain or improve health. We derive these insights using our Health Digital Twins, virtual representations of certain physiological, biological or behavioral processes that we can model or simulate in-silico. Large language models (LLMs) can bridge the gap between complex health data and the understanding of insights by patients, caregivers, and healthcare providers, by translating insights into accessible and personalized explanations, recommendations, and alerts. This interface enhances usability, allowing end-users to interact intuitively with their digital twins, understand actionable health insights, and make informed decisions. You will help us assess the feasibility of creating an LLM environment as an interface for our existing data, algorithms and models in the Digital Twin R&D program. Developing an early prototype of such an environment is part of the internship.

In short, the internship involves:

• Conducting a comprehensive literature review on the subject.
• Mapping existing sensor data and algorithm output to ontology standards.
• Prototyping and testing a large language model setup in an iterative development.
• Working in an agile setting to deliver timely, effective results.

The internship work and activities will be organized with a scrum-like methodology: you will maintain the backlog in coordination with your mentors. You will select prioritized tasks from the backlog, and you will tackle and evaluate them on a biweekly basis. At the end of each biweekly iteration, you will showcase the progress made and will reflect on insights and improvements to focus on. Additional stakeholders may take part in the showcases to get better feedback on the study.

To apply and for the full project description, please visit https://www.imec-int.com/en/work-at-imec/job-opportunities/student-project-conversational-interfaces-digital-twins-health.

Contact: Ruud van Stiphout (Imec); Tom Heskes (RU).