Info: Zenodo’s user support line is staffed on regular business days between Dec 23 and Jan 5. Response times may be slightly longer than normal.

Published March 7, 2024 | Version v2
Other Open

Kinematics, Kinetic and EMG Dataset of overground locomotion with IMUs, optical motion tracking and EMG systems

  • 1. University of Twente

Description

A human movement experiment with four male post-stroke participants walking overground at two different speeds (comfortable and faster speed) was conducted with the ethic approved by the regional medical ethics committee of Eastern Netherlands (METC Oost-Nederland).

Each participant completed three tasks: a static standing pose in a neutral position, walking at a self-selected comfortable speed and walking as fast as possible. A minimum of 10 gait cycles in each walking condition were recorded.

EMG data were recorded using an EMG-sensorized garment embedded with a grid of 64 EMG electrodes equally distributed and surrounding the leg of the affected side. Seven muscle-specific activations (Tibialis anterior, peroneus longus, peroneus brevis, gastrocnemius lateralis and medialis, soleus, and extensor hallucis longus) were extracted via a previously developed muscle localization algorithm. 

Kinematics data were captured using five IMUs (XsensTM MTw, Movella, Enschede, The Netherlands) and a 3D motion tracking system (Vicon, New York (NY), USA).
IMUs were attached with velcro straps: on the sacrum at the midway point between the line connecting the left and right posterior superior iliac spine, on each foot on the midfoot region, and on the impaired leg on the thigh on the mid-femur lateral region, and on the tibia below the tibial tuberosity (Figure1). The MT Manager software (MT manager 2019.2, Movella, Enschede, The Netherlands) was used for wireless data recording from the IMUs, with a sampling rate of 100 Hz.
Thirty-seven reflective markers were placed on bony landmarks and leg segments. Data were acquired at 100 Hz using 16 optical
cameras (eight Vicon Bonita cameras, four Vicon Vero 2.2 cameras, and four Vicon Vue video cameras).
Simultaneously, 3D GRFs were recorded at 1000 Hz via two floor-embedded force plates (Kistler 9286BA, Kistler, Winterthur, Switzerland). 

In this shared repository, processing code (to be uploaded), and processed data are provided.

In each subject folder(s0x) are stored:
- quaternions from the IMUs
- joint angles obtained from OpenSim IK IMU
- GRFs computed using IMU accelerations and inverted pendulum approach
- Ankle dorsi-plantar flexion moment computed via OpenSim inverse dynamics driven by GRFs and joint angles computed from IMU data
- Reference GRF recorded using double force plates
- Reference joint angle output from OpenSim inverse kinematics driven by markers trajecotries
- Reference ankle moments computed using OpenSim inverse dynamics from reference GRFs and joint angles
- Opensim unscaed and scaled models
- EMG-driven musculoskeletal modeling output using as input reference data and IMU-derived data

For more information on the content of each folder and subfolder refer to the ReadMe file. 

 

Files

ReadMe.txt

Files (536.9 MB)

Name Size Download all
md5:0b74f514b6f085a286ae1a7de0a510a7
2.1 kB Preview Download
md5:f2f2c0407ffdf9aca2c51b8ea4935299
145.5 MB Preview Download
md5:314d2b39915a52368a063f8442273f7a
157.3 MB Preview Download
md5:928f0cd75af39b8064274fe36128c014
66.9 MB Preview Download
md5:bc405f6e5931e6126805d963130cbe81
167.2 MB Preview Download