A new EMG-based index towards the assessment of elbow spasticity for post-stroke patients

The commonly used method for grading spasticity in clinical applications is Modified Ashworth Scale (MAS). The MAS-based method depends on the subjective evaluations and the experience of physicians, which may lead to imprecise and inconsistent evaluations. In this study, we propose a novel index (A-ApA, which was calculated with the root mean square (RMS) of agonist muscle activity by the mean between the RMS of agonistic and antagonistic muscle activations extracted from surface electromyography (sEMG) signals to quantitatively assess elbow spasticity. 39 post-stroke patients with elbow spasticity caused by hemiplegia participated in the experiments, and their elbow spasticity was assessed with MAS by one experienced physiotherapist. Patients were thereafter divided into four groups according to the MAS scales. The sEMG signals were recorded simultaneously on the patients' biceps and triceps when they extended or bended their elbows passively. The correlations between MAS and RMS of sEMG signals as well as the newly proposed index were calculated. The results demonstrated that the correlation between the MAS and the sEMG-based index in the assessment of elbow spasticity was significant. This suggests that the EMG-based index would be helpful for the assessment of spasticity..