This is the repository that contains the notebooks for the Multiple-Sclerosis project at the Mobility and Fall Prevention Research Laboratory at the University of Illinois Urbana-Champaign.
The aim of this project is to use multi-class labeled gait (accelerometer) data to build a classifier for HoA individuals. Once relevant features are identified - we intend to look at the class-wise differences between HoA and MS individuals to attempt to build a MS classifier to aid in diagnoses. The data in question has is tri-axis accelerometer from an E4 wrist-band, sampled at 32Hz. This was collected during 2 hour clinical trials, where subjects performed a varited of ambulatory tasks.
This noteook was written in a Python 3.10.12 environment. The exact dependencies have been listed out in requirements.txt.
Ensure that a data/ directory exists. In case it does not, the read_csv() method may break, or the return value may be an empty list object.
You can fork the repository and submit Pull Requests to contribute. Please work on reature/dev branches while pushing.