Abstract
BACKGROUND:
Most of the EMG analysis algorithms developed to date don't detect the whole sequence of rhythmic and subtle changes that take place during the process of trunk stabilization. Indeed, the few recent methods that are capable of assessing these important EMG characteristics are highly complex and not accessible in most applied clinic contexts.
OBJECTIVE:
To validate and disseminate a software program suitable for detecting multiple and relatively small EMG bursts during a trunk stabilization response.
METHODS:
Ninety EMG recordings randomly selected from 50 individuals (24 with chronic low back pain) were analysed by our algorithm based on means and standard deviations and an experienced examiner (as a gold standard). Concordance, sensitivity, specificity, positive predictive value and negative predictive value were considered to analyse reliability.
RESULTS:
Results showed a high degree of concordance between the two methods (87.2%), high sensitivity and specificity rates (79.5 and 89.2%), a moderate-low positive predicted value (66.9%) and a high negative predicted value (94.4%).
CONCLUSION:
The program provided is flexible and useful to detect EMG activity. The selected parameters of the program were able to detect onset/offset EMG bursts and were valid for the purpose of this study with a small tendency to over-detect bursts.
Get full access to this article
View all access options for this article.