• Uncontrolled manifold
• kinematic synergy
• human walking
• motor control
• rehabilitation
• stroke

The goal of the present thesis is to verify if the mathematical tool of Uncontrolled Manifold (UCM) theory is suitable to detect the effects of a rehabilitation program in individuals with post-stroke hemiparesis, usually characterized by an atypical motor variability, resulting in a possible instability. The UCM hypothesis is that the CNS “co-varies” multiple elemental variables consistently in order to guarantee a stable trajectory for the task performance variable. The variance of elemental variables is split into two components: VUCM that does not affect the motor performance and VORT that does affect the time course of the task variable. The Ratio of VUCM and VORT is used as a measure of degree of stability: if Ratio>0, than the performance is said to be stabilized by kinematic synergy of the elemental variables. In addition, the Positive Rate of Ratio (PRR) was proposed as index of synthesis.
Kinematic data of stroke patients was recorded during walking on treadmill using an optoelectronic system in two sessions, pre- and post- rehabilitation, jointly to a clinical evaluation performed by a therapist. A 7-segment planar model of human walking was used for UCM analysis to characterize the movement in the sagittal plane during the stance phase.
According to the results, the PRR reflected both clinical and kinematic outcomes and therefore the UCM analysis may be useful to measure quantitatively the effects of rehabilitation on motor function of post-stroke individuals.