Normal and pathological gait classification LSTM model

Margarita Khokhlova (1) , Cyrille Migniot (1) , Alexei Morozov (2) , Olga Sushkova (2) , Albert Dipanda (1)

1 CORES - Equipe CORES [ImViA - EA7535]
2 IRE - Kotelnikov Institute of Radio Engineering and Electronics

Computer vision-based clinical gait analysis is the subject of permanent research. However, there are very few datasets publicly available; hence the comparison of existing methods between each other is not straightforward. Even if the test data are in an open access, existing databases contain very few test subjects and single modality measurements, which limit their usage. The contributions of this paper are three-fold. First, we propose a new open-access multi-modal database acquired with the Kinect v.2 camera for the task of gait analysis. Second, we adapt to use the skeleton joint orientation data to calculate kinematic gait parameters to match golden-standard MOCAP systems. We propose a new set of features based on 3D low-limbs flexion dynamics to analyze the symmetry of a gait. Third, we design a Long-Short Term Memory (LSTM) ensemble model to create an unsupervised gait classification tool. The results show that joint orientation data provided by Kinect can be successfully used in an inexpensive clinical gait monitoring system, with the results moderately better than reported state-of-the-art for three normal/pathological gait classes. Copyright © 2019 Elsevier B.V. All rights reserved.

Mots clés: Kinect skeletons; Gait modeling; Gait assessment; LSTM; Low-limbs motion; RGB-D

Domaines: Traitement des images [eess.IV]; Environnements Informatiques pour l'Apprentissage Humain; Vision par ordinateur et reconnaissance de formes [cs.CV]

Margarita Khokhlova, Cyrille Migniot, Alexei Morozov, Olga Sushkova, Albert Dipanda. Normal and pathological gait classification LSTM model. Artificial Intelligence in Medicine, 2019, 94, pp.54-66. ⟨10.1016/j.artmed.2018.12.007⟩. ⟨hal-02064234⟩