ID 2430: EmpkinS D02 – Depression Detection from Body Posture and Movement

Depression is one of the most common mental health disorders, significantly reducing an individual’s quality of life. Psychologists use various methods to detect and assess depression, such as analyzing facial expressions, speech patterns, and head movements. In this research, our goal is to detect depression using upper body expressions and movements. This study is part of a more extensive series of research on depression, which has gathered data from interviews with over one hundred participants, including both depressed individuals and a control group. Specifically, we aim to use Kinect Azure data from these interviews to train a deep-learning model that can automatically detect depression.

Here, you can find more information about the D02 and EmpkinS projects.

These are some of the most relevant studies to this research:

1- Liu, Dongdong, et al. “Measuring depression severity based on facial expression and body movement using deep convolutional neural network.” Frontiers in psychiatry 13 (2022): 1017064.

2- Yu, Yanhong, et al. “Depression and severity detection based on body kinematic features: using Kinect recorded skeleton data of simple action.” Frontiers in Neurology 13 (2022): 905917.

3- Joshi, Jyoti, et al. “Relative body parts movement for automatic depression analysis.” 2013 Humaine association conference on affective computing and intelligent interaction. IEEE, 2013.

Tasks

• You will work with interdisciplinary researchers and should be ready to research topics outside your field.
• You will implement the existing pose estimation model on Kinect data (Depth/RGB recordings)
• You will perform statistical and machine learning data analysis on the pose data set.
• You should document your code in a clear and structured manner.
• You will report on your progress weekly and present your results and work several times to the MaD lab.
• As part of joint research with another group in EmpkinS, you should help another master’s student to measure the effects of compression in the results.

Requirements

• Knowledge in data processing, data analysis, and data visualization using Python
• Knowledge/willingness to learn Pose Estimation and Motion analysis.
• Interest in diving into the field of Depression and Psychology
• English proficiency

Supervisors

Amirreza Asemanrafat, M. Sc.

Researcher & PhD Candidate

Misha Sadeghi, M. Sc.

Researcher & PhD Candidate