Marker Position Prediction from raw IMU data

As a part of the Article: Bridging The Gap Between Optical Motion Capture and Inertial Measurement Unit Technology: A Deep Learning Approach to Joint Kinematic Modeling

Journal Name: IEEE Journal of Biomedical and Health informatics

DOI: (Will be added)

Installation

• Install Miniconda for your operating system
• Clone this repository
• Navigate to the repository folder on your local machine
• Install the virtual environment:

conda env create -f environment.yml
conda activate Marker-position

Note: you may need to add your environment to the list of jupyter kernels:

python -m ipykernel install --user --name=Marker-position

Running Jupyter notebooks

• Navigate to the notebooks subfolder
• Launch jupyter
• Navigate to appropriate notebook on your browser

Step 1 : Prediction of Marker positions from raw IMU data and Visulizing predicted markers

--> Run Marker_Position_Prediction_Notebook.ipynb

Step 2 : Calculating Joint angles in MATLAB and Comparing against optical motion capture joint angles.

--> Add biomechZoo toolbox in MATLAB search path.

--> Run Kinematics_calculation.m

Toolboxes and/or Supporting Materials

• Kinetics Toolkit
• biomechZoo