A systematic review of integrated machine learning in posture recognition

Abstract Introduction Posture detection is used in various situations such as medical care, surveillance, virtual environment, indoor monitoring, virtual reality for animations and entertainment. The concept of machine learning has experienced great progress in the last two decades, from a curiosity started in the laboratory to a widespread practical technology for commercial use. Objective The aim of this paper is to review the literature on the use of machine learning algorithms in the medical field for posture recognition. Material and method Articles were collected from the following databases: Google Scholar, Science Direct, PubMed and Research Gate. We included only articles that were written in English, those that were available for download in full text, published after 2010, the year in which the industrialization of the idea of artificial learning began. Articles that did not assess or recognize the posture deficiencies were excluded. Results A total of 55 articles were eligible for the study. Following the inclusion criteria, and after sorting, using the exclusion criteria, a number of 16 articles remained to be analyzed, presented and discussed. Conclusions After the analysis of the articles included in this study, it can be concluded that using machine learning we can obtain very good results with high accuracy for posture recognition.