Realizing the Human Digital Twin for Stroke (Aphasia) Rehabilitation
This project aims to develop a human digital twin system for GazeTalk — an augmentative and alternative communication (AAC) and aphasia rehabilitation app in pilot development at DTU. Current AI conversation partners in aphasia tools are generic: they respond to the patient, but do not model who the patient is. A digital twin approach would capture each patient's pre-aphasia linguistic identity — their vocabulary, syntax, topic preferences, and communication style — and use this model to personalize rehabilitation goals, conversation difficulty, and progress tracking over time.
The goal of this project is to design and prototype a personalization architecture for GazeTalk based on individual patient language profiles. More specifically, the project should:
- Conduct a deep literature review spanning NLP personalization, aphasia linguistics, and digital twin concepts in healthcare
- Co-design the linguistic identity model with clinicians — defining what a clinically useful patient profile looks like
- Define data sources (pre-aphasia writing samples, caregiver-provided text, recordings) and design appropriate consent and data handling protocols
- Build a prototype pipeline: ingest pre-aphasia data → construct language profile → adapt AI conversation behavior accordingly
- Evaluate the prototype using simulated patient profiles before any real patient involvement
This project directly continues the development of GazeTalk and lays the groundwork for truly personalized aphasia rehabilitation. The project will be supervised by postdoctoral researcher Ekky Tammarar Alfian and co-supervised by Prof. John Paulin Hansen. This project can be formulated into a Master thesis/Bachelor thesis/student project. The project is expected to start in the Fall semester 2026 (negotiable).
Thesis type: Design and development of human digital twin prototype for aphasia rehabilitation
Technical skills: NLP and language modelling; data pipeline design; prompt engineering
Research skills: Natural Language Processing; Clinical Informatics; Human-Computer Interaction
Educational background: Computer Science (NLP/AI track); Cognitive Science; Human-Centered Computing;
ECTS credits: MSc thesis / BSc thesis / Special course (5 or 10 ECTS)
Contact: <correo electrónico eliminado por motivos de seguridade>
By working on the project, you will be part of a team aiming to improve lives for stroke patients!