ID 2513: Detection & prediction of freezing events in patients with Parkinson’s disease

Freezing of gait (FoG) is a common and disabling motor symptom in patients with Parkinson’s disease (PD), characterized by brief, episodic absence or marked reduction of forward progression of the feet despite the intention to walk. Patients often describe it as feeling as though their feet are “glued” to the ground.

In this project, we aim to detect and possibly predict freezing of gait (FoG) episodes in PD patients using data collected from inertial measurement unit (IMU) sensors. These wearable sensors, placed on the feet, capture detailed motion signals such as acceleration and angular velocity during walking. By analyzing these signals, we develop algorithms capable of identifying patterns associated with FoG events in recorded data from their stay at a rehabilitation facility.

By combining signal processing techniques with machine learning and AI models, we aim for an accurate detection of freezing episodes during daily activities. The ultimate goal of this project is to create a reliable, non-invasive monitoring tool that can support clinicians in assessing treatment efficacy and help patients manage their condition more effectively in daily life.

Here, you can see what a FoG looks like.

Tasks:

1. Literature Review
2. Data Handling
3. Feature Engineering
4. Model Development
5. Real-Time Considerations (Optional)
6. Validation and Interpretation
7. Documentation and Thesis Writing

Requirements:

1. Scientific writing and critical thinking.
2. Programming with Python & Data Handling using tools such as Pandas
3. Feature selection and engineering techniques
4. Machine Learning / Deep Learning models used in time series
5. English language

Supervisors

Hamid Moradi, M. Sc.

Researcher & PhD Candidate