Master’s Thesis – Analysis of Emotion-Related Pupillometry/Eyetracking in Parkinson’s Disease
Masterthesis / Biomedical Engineering / AI in Medicine
Status: Open
Time frame: Start date: September 2025 or upon agreement.
Master’s Thesis – Analysis of Emotion-Related Pupillometry/Eyetracking in Parkinson’s Disease
Project background
Parkinson’s disease (PD) is not only characterized by motor symptoms but also by non-motor symptoms such as depression, anxiety, apathy, and impulse control disorders, which may affect emotion processing (Weintraub & Mamikonyan, 2019). Deficits in recognizing negative emotions such as anger, disgust, fear, and sadness are frequently reported in PD (Argaud et al., 2018). Pupillometry is a promising marker for emotion processing, but raw pupil data are noisy and require careful preprocessing. Advanced statistical models, such as linear mixed models (LMM) and generalized additive mixed models (GAMM), provide a framework to capture both inter- and intra-individual variability, as well as non-linear trajectories over time, which have not yet been fully exploited in this context.
Aim
The aim of this thesis is to establish a high-quality, cleaned dataset of pupillometry responses and gaze location from PD patients during an emotional picture-rating task. Using advanced statistical models, you will investigate whether pupillary responses differ across pleasant, unpleasant, and neutral stimuli, and how they are modulated by clinical measures such as depression, anxiety, apathy, or impulse control disorders. In a sub-cohort, you will investigate the effect of deep brain stimulation (before/after surgery) on pupillometry and gaze during the task.
Materials and Methods
You will start from the raw pupillometry recordings acquired during the International Affective Picture System (IAPS) task in PD patients treated with subthalamic nucleus deep brain stimulation (STN-DBS). The first stage will involve the development of a rigorous preprocessing pipeline, including blink detection, interpolation, baseline correction/normalization, resulting in a cleaned dataset. You will start exploring the data with visual plotting and basic statics. Furthermore, you will apply advanced statistical modeling LMMs to account for repeated measures and nested data structures, and GAMM to model potential non-linear pupil dynamics across time. Predictors will include stimulus category (pleasant, unpleasant, neutral), or ratings, while controlling for measures of neuropsychiatric symptoms (HADS, QUIP-RS, Starkstein Apathy Scale, MoCA).
Nature of the Thesis
Data preprocessing & dataset construction: 50%
Statistical modelling & interpretation: 50%
Requirements
Good programming and data analysis skills in Python and/or R
Familiarity with time-series data
Experience with statistical methods (LMM/GAMM) will be an advantage (but is not necessary)
Interest in neuroscience and clinical research
Institutes
Gerontechnology and Rehabilitation Group, ARTORG Center for Biomedical Engineering Research, University of Bern
Center for Parkinson’s and Movement Disorder, Department of Neurology, Bern University Hospital and University of Bern
Supervisors
Prof. Dr. Tobias Nef (tobias.nef@unibe.ch)
Dr. Deborah Amstutz (deborah.amstutz@insel.ch)
Matilde Castelli (matilde.castelli@unibe.ch) & Aaron Colombo (aaron.colombo@unibe.ch)
Workplace: ARTORG Center (SITEM, Freiburgstrasse 3) on the Insel hospital campus.
Language: English
Contact: aaron.colombo@unibe.ch
Application: apply via E-Mail with the subject: “MT Eye Emotion PD – Firstname Lastname” and
CV
Course transcripts
A Short motivation letter (max. ¾ A4 page)
References
Argaud, S., Vérin, M., Sauleau, P., & Grandjean, D. (2018). Facial emotion recognition in Parkinson’s disease: A review and new hypotheses. Movement Disorders, 33(4), 554–567. https://doi.org/10.1002/mds.27305
Weintraub, D., & Mamikonyan, E. (2019). The Neuropsychiatry of Parkinson Disease: A Perfect Storm. The American Journal of Geriatric Psychiatry, 27(9), 998–1018. https://doi.org/10.1016/j.jagp.2019.03.002