Skip to content

Machine Learning model for detecting gait events in mocap data

License

Notifications You must be signed in to change notification settings

hmok/event-detector

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

16 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

event-detector

This software annotates .c3d gait trajectories with Heel-Strike and Foot-Off events. It uses neural networks (more precisely Long Short Term Memory networks) through keras and tensorflow packages.

Requirements

You only need python and btk (https://pypi.python.org/pypi/btk), all other dependencies will be installed automatically.

Linux, Mac OS

Good news! Linux and Mac OS come with python distributions ready to use. You can go to the Installation step

Windows

You need to install python, for example from here https://www.python.org/downloads/windows/

Instalation

Once python is installed, open terminal/command prompt and type

pip install eventdetector

This will install eventdetector scripts with all required dependencies

Running

Navigate Terminal or Command Prompt to a directory with .c3d files you wish to annotate with Heel-Strike and Foot-Off events. Then type

event-detector [file-in.c3d] [file-out.c3d]

where [file-in.c3d] is the name of the file to annotate and [file-out.c3d] is the name of the new file in which you want to store annotation.

Enjoy!

Credits

This research was sponsored by the Mobilize Center, a National Institutes of Health Big Data to Knowledge (BD2K) Center of Excellence supported through Grant U54EB020405. The model is trained on the data from Gillette Children's Specialty Healthcare, in accordance with the data sharing agreement. For the training scripts refer to https://github.com/kidzik/event-detector-train

About

Machine Learning model for detecting gait events in mocap data

Resources

License

Stars

Watchers

Forks

Packages

No packages published

Languages

  • Python 100.0%