ID 2334: Depression detection using multi-domain features from motion sensors and voice
Depression is a common mental health disorder that often goes undetected or is diagnosed at a late stage, leading to long periods of suffering for individuals affected. Early identification allows for prompt intervention and treatment that can alleviate symptoms and improve overall mental well-being. Recent advancements in wearable technology have opened new possibilities for early and objective methods of depression detection.
The goal of this thesis is to design a multi-domain system to detect depression by combining motion data from ear-worn IMUs and voice features. A study to collect a representative data set should be planned and conducted. All developed algorithms should be data-driven and the impact of individual features as potential markers of depression should be analyzed.
Start Date: Sommer/Fall 2023
- Literature research
- Data recording
- Process data and extras relevant features for motion data and voice
- Design a system for the detection of depression based on motion and voice features
- Bring in your own ideas
- Fluent German skills (required for data collection)
- You should have empathy and show a lot of initiative
- Strong programming skills in Python
- Advanced knowledge and some experience in Machine Learning
- Advanced knowledge in signal processing
- Bachelor’s degree in a related field of study
Ann-Kristin Seifer, M. Sc.
Researcher & PhD Candidate
Please use the application form to apply for the topic. We will then get in contact with you.