Student (f/m/d) for master thesis in data-driven health research
Topic: Agentic AI for Circadian Metabolic Research as a Multimodal Analysis of Nutrition, Glucose, and Light Exposure Data in Healthy Adults
Starting Date: October 1st 2025
Supervising Institutions:
- Institute of Occupational Medicine, Healthy Living Spaces lab, RWTH Aachen University (Dr. Jan-Frieder Harmsen)
- Department of Gastroenterology, Metabolic Diseases and Intensive Care, University Hospital RWTH Aachen (Prof. Dr. Carolin Schneider)
Background:
Circadian misalignment is increasingly recognized as a key pathophysiological driver of metabolic diseases, especially in populations exposed to night shift work. Recent advances in wearable sensor technology, such as continuous glucose monitoring (CGM), light exposure tracking, and smartphone-based food photography, now allow for highly granular 24-hour data acquisition in real-world settings.
Conventional data analysis approaches are often insufficient for capturing the complexity of these multimodal, asynchronous datasets. Here, agentic artificial intelligence (AI) offers novel opportunities for both methodological development and scientific discovery.
Aims:
This thesis project aims to develop agentic AI pipelines for the integrated analysis of CGM time series, meal photographs, and light exposure data collected over 10 days in everyday life of healthy young and older adults.
The dataset comprises ten consecutive days of data from 60 healthy participants, including 30 younger and 30 older adults. Each participant underwent continuous glucose monitoring (CGM), capturing interstitial glucose values at 15-minute intervals. Light exposure was continuously recorded using state-of-the-art wearable sensors, while food intake was logged via timestamped photographs taken with smartphones. All data have undergone initial quality control and harmonization.
The goal is to explore how modern agent-based AI frameworks can autonomously generate annotations, perform time-aware analyses, and support hypothesis generation in the context of circadian-metabolic research. The thesis is both methodologically innovative and biomedically relevant, with future applications in shift worker populations and digital precision prevention.
Methodological Components:
- Autonomous Workflows:
- Design and deployment of multi-tool AI agents
- Autonomous execution of analysis chains on CGM time series
- LLM-driven reflection and optimization of analytical decisions
- Computer Vision:
- Automated recognition of food types and portion sizes from meal images using pretrained vision-language models
- Calorie and macronutrient estimation based on image context and metadata
- Time-Series Analysis & Machine Learning:
- Detection of circadian glucose patterns (e.g., Cosinor models, Fourier transforms, LSTM networks)
- Clustering and classification of metabolic response profiles across age groups and time-of-day
- Large Language Models:
- Semantic annotation and interpretation of nutrition data
- Automated reporting, documentation, and hypothesis formulation
- Contextual linking of food content, intake time, light exposure, and glucose response
Requirements:
- Strong motivation for interdisciplinary, data-driven health research
- Proficiency in Python; experience with OpenAI APIHuggingFace, or similar frameworks is an advantage
- Familiarity with machine learning or computer vision methods is desirable
- Interest in circadian biology, metabolic health, and digital prevention
- Willingness to explore AI tools and autonomous workflows
Conditions
- A valid certificate of study
- For foreign students: valid residence certificate, work permit, registration certificate
Application:
Please submit your application through our application portal, quoting GB-P-52259. The application deadline is August 27th 2025.
Contact:
For more detailed information please contact Prof. Dr. med. Carolin Victoria Schneider:
E-Mail: cschneider@ukaachen.de
We look forward to receiving your application!
This position is not gender specific.
The RWTH Aachen University Hospital promotes equal opportunities and diversity. Applications from women are expressly encouraged and if the applicant is suitable qualified, they will be given priority in accordance with the LGG. If suitably qualified, people with a registered disability will also receive priority.
Weekly hours are negotiable.
You should preferably use our digital application portal at www.karriere.ukaachen.de for your application. There you have the option of securing your documents in the electronic application folder to prevent unauthorized access. Applications that reach us by email to: bewerbung@ukaachen.de (this transmission path cannot be as effectively secured) will be transferred to the aforementioned portal and any accompanying documents will be disposed of in accordance with data protection regulations immediately after transfer. After the retention period has expired, the data in the portal will also be deleted. If you do not agree to a transfer to the Application portal your application cannot be considered.