move - functional
Human movement patterns are a composition of unique and generic movement characteristics. Unique characteristics highlight those aspects that are specific to any given individual, while generic characteristics highlight aspects that are common to all individuals. Unique and generic characteristics can be isolated by applying layer-wise relevance propagation to an artificial neural network that was trained to recognize individuals based on their movement patterns.
If you are interested in the code and the data, please send me an email
The data can be downloaded here:
Data_and_Codes.rar (8.6GB)
If you prefere to download only the code or data please use the individual links below:
AnalysisOfOverlap.m (3kB)
Data.mat (3.5 GB)
GaitCharacteristics.m (5kB)
gliding_filter.m (2kB)
Model.mat (2.7 GB)
pcolap_ashape.m (2kB)
plotrelmap.m (4kB)
readme.docx (16kB)
Results.mat (1.7 GB)
lrp_toolbox-master.zip (300 MB)
The aim of this study was to test the similarity of running patterns of elite athlets. The study was designed to have a very similar situation, with a flat course and similar running speed. This study is currently under review in frontiers. If you are interested in the code and the data, please send me an email
The code to analyse the sample data is written in matlab version 2018a:
matlab based skript to analyse data
A sample of the data can be downloaded here (you need all 5 parts in order to run the skript):
Sample data part1(4 subjects) (5MB)
Sample data part2(4 subjects) (66MB)
Sample data part3(4 subjects) (66MB)
Sample data part4(4 subjects) (66MB)
Sample data part5(4 subjects) (66MB)
The code requires the statistical parametric mapping toolbox: spm1D