An intelligent decision support system to anticipate the fall of a structure for paraplegic patients

The Human posture is defined by the ability to maintain a stable vertical position of balance while keeping the feet fixed relative to the ground. An accurate understanding of the equilibrium conditions is an essential point to dimension and model the forces in the exoskeleton structures. Several experiments were performed in this sense. However, the resulting data are quite vague and uncertain, which could contribute to the error in the equilibrium description. In this paper, we focus on the folding. Our objective is to find the factors and mechanisms that influence this fallback in order to improve its anticipation. We have considered an approach based on fuzzy logic to better explain the uncertain and ambiguous aspect of the data, and on data mining algorithms to find the factors that influenced the fall. The results of our methodology showed promising associations to anticipate this type of fall.