Non-invasive methods of computer vision in the posture evaluation of adolescent idiopathic scoliosis

摘要

Purpose: Reviewing techniques for non-invasive postural evaluation of adolescent idiopathic scoliosis (AIS) based on information extraction from images based on computer methods. Methods: The Scopus, Web of Science, MEDLINE, ScieLo and PubMed databases were used, for the period 2011-2015. Results: 131 articles were found based on keyword of which 15 articles met the established eligibility criteria. Of these, 4 were based on photogrammetry, and 11 based on laser, structured light, ultrasound, and Moire projection. In these studies, the methodological quality varied from low to high. Conclusions: The findings indicated diversity in methodologies; 14/15 articles reviewed were limited to the evaluation of the topography of the posterior back. A study, using two-dimensional photogrammetry, presented a whole body postural evaluation. As the asymmetry in AIS can be extended to the whole body, more attention should be given to develop full body assessment techniques to provide important additional data to aid in treatment decisions.