Continuous time series analysis using a Hidden Markov Model enhances informations about joint angle functionality of human gait

Abstract

stance balance can be also associated with inadequate somatosensory perception in the residual limb [2]. Materials and methods: Sixteen patients with traumatic unilateral transtibial amputation (13males, 3 females; age 42.8±10.8 years, height 179.6±11.1 cm,weight 88.6±13.9 kg) participated in this study. The length of stump and stump sensation was assessed. For assessing the latter, the two-point discrimination was used. Todetermine thebiomechanical parameters (COP swayandCOP velocity movement) of postural stability during stance, the force plate Kistler9286AA was used. The stability was tested in modified situations: stance with eyes opened (EO), with eyes closed (EC), on a foam mat (EOM) and with eyes closed on a foam mat (ECM). For statistical comparison, the Spearmann correlation and the effect size (Cohen’s d) were used. Results: The results showed statistically significant correlation between stump length and COP sway in the anteroposterior direction – AP in EO (r=−0.526) and in EC (r=−0.506), and COP velocity movement in the mediolateral direction – ML (r=−0.597) in EC. In ECM, significant correlations were determined in all tested situations. The results between the group with hyposensitivity and the group with normal sensitivity in COP parameters showed the high effect of hyposensitivity on COP sway in AP (d=1.04) in EOM, medium effect on COP sway in AP (d=0.68) and ML (d=0.51) in EO and in ML (d=0.53) in EOM. Medium effect was found on COP velocity movement in AP in EO (d=0.50) and in EC (d=0.52). Discussion: The outputs indicate a significant correlation between stump length and COP parameters during stance modification, especially in the stance on foam mat with eyes closed. It shows that amputees with longer stump have better stability during stance [1]. Amputeeswith stumphyposensitivityhad less stable stance than amputees with normal sensitivity. Further research encompassing a larger number of subjects would be appropriate to validate these results.