Source: European Journal of Internal Medicine, 2016, Vol 27 p. 48-56
Follow this link for the abstract
Date of publication: January 2016
Publication type: Journal article
In a nutshell: This study compared various performance criteria of different statistical models for fall risk in older community-dwellers. It tested six linear models and three non-linear models, including neuroevolution of augmenting topologies (NEAT) and the adaptive neuro fuzzy interference system (ANFIS). The study found that these latter two had the best performance criteria compared to other models, but sensitivity and specificity were unbalanced, underscoring that these models should be used respectively for the screening of fallers and the diagnosis of recurrent fallers.
Length of publication: 8 pages
Some important notes: Please contact your local NHS Library for the full text of the article. Follow this link to find your local NHS Library.