Humans can easily recognize running, walking, shuffling or limping. For automatic classification of dynamic movement patterns, we will integrate analyses of multivariate time series and state-of-the-art techniques for pattern recognition. This web-site will provide an entry to our software solutions for analyzing gait patterns of (younger and older) healthy people but also people suffering from specific neurological disorders.

Primary host is the Research Institute MOVE, Faculty of Human Movement Sciences of the VU University Amsterdam but we expect many partner institutions to contribute to data acquisition and software development.

Whole-body kinematic recordings, (multivariate) EMG, accelerometer measurements, all contain different information about movement. For the gait analysis we extract features like step/stride length, step/stride frequency (or duration), left/right asymmetries, et cetera. These more conventional measures will here be supplemented by, e.g., data mining through principal or independent component analysis, and approaches used to analyze complex dynamical systems, especially stability-related Lyapunov or Floquet exponents and (detrended) fluctuation analysis or entropies.

Matlab and object-oriented programming make the package readily extendable.

Open-source is the only premise for our software development.

Visit this web-site frequently and check for updates. You can also subscribe to our mailing list, or follow us on Twitter.

Eager to try UPMOVE? It is currently in early alpha stage and only available to a limited testing group. People interested in joining this test group can register here.

©2010-2013 Andreas Daffertshofer & Ronald Kaptein