﻿This readme.txt file was generated on 2021-08-05 by SALIL APTE


GENERAL INFORMATION

1. Title of Dataset: Neglected physical human-robot interaction may explain variable outcomes in gait neurorehabilitation research

2. Author Information
	A. Principal Investigator Contact Information
		Name: H. Vallery
		Institutions: Delft University of Technology | Erasmus University Medical Center
		Address: Department of Biomechanical Engineering, Delft University of Technology, Mekelweg 2, 2628 CD Delft, The Netherlands | 
			 Department of Rehabilitation Medicine, Erasmus MC, Postbus 2040, 3000 CA Rotterdam, The Netherlands
		Email: h.vallery@tudelft.nl

	B. Associate or Co-investigator Contact Information
		Name: J. von Zitzewitz
		Institution:  ONWARD Medical SA | École Polytechnique Fédérale de Lausanne 
		Address: ONWARD (formerly known as GTX Medical), EPFL Innovation Park, Lausanne, Switzerland | 
			 Center for Neuroprosthetics (CNP) Valais, École Polytechnique Fédérale de Lausanne (EPFL), Switzerland
		Email: joachim.vonzitzewitz@onwd.com

	C. Alternate Contact Information
		Name: Michiel Plooij
		Institution: Delft University of Technology | Demcon
		Address: Demcon Advanced Mechatronics, Delfttechpark, Delft, The Netherlands
		Email: michiel.plooij@demcon.com

3. Date of data collection: 2018

4. Geographic location of data collection: Delft, Netherlands and Lausanne, Switzerland

5. Information about funding sources that supported the collection of the data: Eurostars project 10152, and Marie-Curie career integration grant PCIG13-GA-2013-618899


METHODOLOGICAL INFORMATION

1. Description of methods used for collection/generation of data: 
The data collection protocol is described in the related publication and its supplementary material.
"Neglected physical human-robot interaction may explain variable outcomes in gait neurorehabilitation research"

2. Methods for processing the data: 
Processed data was obtained from the Qualisys proprietary software, which collated the raw data from the Kistler force plates and Qualisys motion capture cameras.
Force and position data of the end effector of the RYSEN BWS device was collected simultaneously via the RYSEN sensors. Details about data processing can be found
in the original manuscript and its supplementary material.

3. Instrument- or software-specific information needed to interpret the data: 
All scripts have been tested in MATLAB R2019a. Data is stored in '.mat' files and the processing script are in '.m' format.
Subject specific information is stored in '.xls' files and can be accessed using Microsoft Excel.

USER INSTRUCTIONS
This folder contains the processing script and data from the three experiments used to analyze:
1. Device transparency/impedance (15 healthy participants) 
2. Ground reaction forces when walking with different anteroposterior support forces (15 healthy participants) 
3. Influence of the harness attachment (15 healthy participants + 3 patients)

Below is the folder directory with filenames and brief explanation of the purpose behind each file.

1. 'main.m' file
	- Running this script automatically loads all the sub-folders, runs the scripts inside them, generates the results and figures 
	  presented in the manuscript
2. 'forceDataProcessed.mat' file
	- Contains the processed data for the analysis of the vertical and antero-posterior ground reaction forces (VGRF and APGRF respectively)

Sub-folders:

1. paperFigures
	- Contains the figures for the device impedance/transparency experiments, as they are presented in the manuscript

2. paperOutcomeData
	- Contains the results for the device impedance/transparency experiments, as they are presented in the manuscript

3. AttachmentExperiments
	3.1 DataFromForms
		- Contains the data from the questionnaire for the preferred force and self-selected velocity experiments
		- Data is obtained for two sets of 15 healthy participants each
		- 'PreferredVelocity.xls' file contains details about this dataset
		- 'main.m' file processes this data
	3.2 dataQTM
		- Contains the data about the position and velocity of the sling-bar, obtained from the Qualisys QTM motion capture software
		- 'qtmDataExtraction.m' file processes this data
	3.3 dataRYSEN
		- Contains the position and velocity information about the sling-bar, obtained from the sensors in the RYSEN BWS device
	3.4 sortData.m
		- Loads the data from 'DataFromForms' and 'dataQTM' folders, and saves them in a combined '.mat' file called 'allDataB.mat'
	3.5 main.m
		- Matlab script to load the data from 'allDataB.mat', and process it to investigate the self-selected velocity, velocity dependent 
		  on the force, force needed to compensate for attachment, and the self-selected forcec. It also generates plots which are saved
		  in the 'paperFigures' sub-folder.

4. patientExperiments
	4.1 Recording_comments.xlsx
		- Details about the data collection with patients. The goal was similar as the one in previous section ('AttachmentExperiments'), wherein the patients
		  answered a questionnaire for the preferred forward/backward force
	4.2 main.m
		- Matlab script to load the data from 'P'n'.mat' files ('n' for nth participant), process it to investigate the preferred forward/backword force and the self-selected walking velocity.
		  It also generates plots which are saved in the 'paperFigures' sub-folder.

5. patientExperiments
	5.1 main.m
		- Matlab script to simulate a static 2-D body model comprised of trunk, upper leg, and lower leg connected via rotational joints
		- The model is simulated under varying conditions of the attachment point and different forward/backward forces

6. GroundReactionForceExperiments
	6.1 Data
		- Contains the raw force plate data for each participants, arranged in one folder per participant
		- 'forceDataProcessed.mat' file contains the processed data for the analysis of the vertical and 
		  antero-posterior ground reaction forces (VGRF and APGRF respectively)
	6.2 library
		- Contains all the scripts used for processing the GRF data
		6.2.1 boxplot2.m
			- Function for plotting the processed GRF data
		6.2.2 extractGRFpeaks.m
			- Function for filtering the force plate signals and extracting the stepwise vertical GRF and 
			  the horizontal pushoff/braking impulse from the antero-posterior GRF
		6.2.3 plotGRF.m
			- This function plots the processed GRF data
		6.2.4 statisticalAnalysis.m
			- Statistical testing for the anteroposterior impulse values
		6.2.5 GRF_p'n'.m
			- Matlab script for loading and processing the raw data for participant 'n'
		6.2.6 loadGRFData.m
			- This function loads the GRF_p'n'.m script ('n' for nth participant, computes the mean and SD of vertical GRF,
			  and saves in a specific format in the file 'forceDataProcessed.mat'
	6.3 main.m
		- Matlab script to load and process the data using 'loadGRFData.m' to investigate the change in the vertical GRF profiles and horizontal impulse
		  with varying support vector conditions, plot it using 'plotGRFm' and conduct the statistical analysis using 'statisticalAnalysis.m'
	6.4 analyzedTrials.xlsx
			- Excel file containing the details of the trials selected for the analysis of the GRF for each participant.

7. ImpedanceExperiments1_Setup
	7.1 'DxC30A50-10.mat', 'DyC30A50-10.mat', 'DzC30A50-10.mat'
		- Matlab data files containing the data for X,Y, and Z directions from the disturbance rejection experiments using a dedicated experimental setup
	7.2 'FxC30.mat', 'FyC30.mat', 'FzC30.mat'
		- Matlab data files containing the data for X,Y, and Z directions from the force tracking experiments using a dedicated experimental setup
	7.3 main.m
		- Matlab script to load the data from the above-mentioned mat files and analyze the disturbance rejection and force tracking performance of the 
		  RYSEN robotic BWS device. Further, it saves the analysis results in the 'paperOutcomeData' folder.
	7.4 misCalibrationTest.m
		- As there was a suspected miscalibration of the RYSEN, additional experiments were performed to correct for this in the measurements. 
		  This matlab script checks for this suspected miscalibration and corrects it if necessary.

8. ImpedanceExperiments2_HumansRYSEN
	8.1 'RYSENData.mat', 'allDataPerStep.mat', 'allDataPerStepAndOrdered.mat'
		- Matlab data files containing the force and position data from the RYSEN device for the 'GroundReactionForceExperiments' conducted with 15 healthy participants.
		  The latter two files contain this data segmented by every step of the participants. 
	8.2 main.m
		- Matlab script to load the data from the above-mentioned mat files and analyze the force tracking performance of the RYSEN device and the bias of its controller. 
		  Further, it saves the analysis results in the 'paperOutcomeData' folder and the generated plots in the 'paperFigures' folder.

9. ImpedanceExperiments3_HumansFLOAT
	9.1 'parsedData.mat'
		- Matlab data file containing the recorded data (force plate, force, position from robot) from an earlier research [1] on influence of the support force vector. 
		  This work was conducted using the FLOAT robotic BWS device and the results presented in the current work serve to compare the transparency (force tracking performance)
		  of the FLOAT and RYSEN devices.
	9.2 main.m
		- Matlab script to load the data from the above-mentioned mat file and analyze the force tracking performance of the FLOAT robotic BWS device and the resulting ground reaction
		  impulses recored using the force plates. Further, it saves the generated plots in the 'paperFigures' folder.
		

References:
	1. Mignardot, Jean-Baptiste, Camille G. Le Goff, Rubia Van Den Brand, Marco Capogrosso, Nicolas Fumeaux, Heike Vallery, Selin Anil et al. 
	   "A multidirectional gravity-assist algorithm that enhances locomotor control in patients with stroke or spinal cord injury." Science translational medicine 9, no. 399 (2017).